Pediatric

Surgery
Diagnosis and Management
Prem Puri
Michael E. Höllwarth
Editors
Second Edition

123
Pediatric Surgery
Prem Puri  •  Michael E. Höllwarth
Editors

Pediatric Surgery
Diagnosis and Management

Second Edition
Editors
Prem Puri Michael E. Höllwarth
Beacon Hospital University Clinic of Paediatric
University College Dublin and Adolescent Surgery
Dublin, Ireland Medical University of Graz
Graz, Austria

ISBN 978-3-030-81487-8    ISBN 978-3-030-81488-5 (eBook)

https://doi.org/10.1007/978-3-030-81488-5

© Springer Nature Switzerland AG 2009, 2023

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, expressed or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To Veena and Christa for their love, support, and inspiration
Preface to the Second Edition

It has been 14 years since the first edition of this book was published in 2009.
In these intervening years, major advances in prenatal diagnosis, imaging,
anesthesia, intensive care, and minimally invasive surgery including robotic
technology have radically altered the management of infants and children
with congenital and acquired surgical conditions.
This second edition of Pediatric Surgery has been substantially revised
and updated to reflect these advances in pediatric surgery. It contains 106
chapters from 234 contributors from five continents. Each chapter has been
written by internationally renowned pediatric surgeons with significant expe-
rience in their respective field of interest. Many younger surgeons who will
become the next generation of leaders in pediatric surgery were invited as
authors or co-authors. This edition contains nine new chapters on important
topics including respiratory management of the surgical patient, access to
enteral nutrition, surgical safety, surgical problems in children with disabili-
ties, surgical implications of HIV infection in children, esophageal replace-
ment, variant Hirschsprung’s disease, and long-term outcomes in pediatric
surgery. We have maintained the previous format, dividing the book into 13
sections. However, as the page extent of the second edition exceeds 1500
pages, we have split the book into two volumes for practical reasons.
The new edition of this book provides an authoritative, comprehensive,
and complete account of the pathophysiology and management of surgical
disorders in infants and children. The book is mainly intended for pediatric
surgical trainees and young pediatric surgeons, providing a comprehensive
description of various surgical conditions in children with a major emphasis
on diagnosis and management. The first edition was recognized worldwide as
an important textbook dealing with surgical conditions in children. We hope
that the thoroughly revised and updated second edition of the book will con-
tinue to act as a reference book for pediatric surgeons worldwide.
We wish to thank most sincerely all the contributors for their outstanding
work in producing this innovative textbook. We also wish to express our grat-
itude to the editorial staff of Springer, particularly Ms Melissa Morton and
Mr Rakesh Kumar Jotheeswaran for all their help during the preparation and
publication of the second edition of this book.

Dublin, Ireland Prem Puri

Graz, Austria Michael E. Höllwarth

vii
Contents

Part I General Principles

1 The
 Epidemiology of Birth Defects����������������������������������������������    3
Florian Friedmacher and Edwin C. Jesudason
2 Fetal
 Counselling for Surgical
Congenital Malformations������������������������������������������������������������   13
Kokila Lakhoo and Rebecca Black
3 Transport
 of the Surgical Neonate ����������������������������������������������   25
Udo Rolle and Prem Puri
4 Pre-operative Management and Vascular Access������������������������   33
Ancuta Muntean, Ionica Stoica, John Gillick, and Prem Puri
5 Anaesthesia and Analgesia������������������������������������������������������������   55
Coilin Collins Smyth and Suzanne Crowe
6 Respiratory
 Management of the Surgical Patient����������������������   71
Gregory Nolan and Suzanne Crowe
7 Fluid Management ������������������������������������������������������������������������   79
S. O’Sullivan and Suzanne Crowe
8 Sepsis ����������������������������������������������������������������������������������������������   85
Lexie H. Vaughn and Jeffrey S. Upperman
9 Nutrition������������������������������������������������������������������������������������������   97
Agostino Pierro and Simon Eaton
10 Access
 for Enteral Nutrition ��������������������������������������������������������  109
Julia Brendel and Michael W. L. Gauderer
11 Hematological
 Problems in Pediatric Surgery����������������������������  119
Peter McCarthy and Owen Patrick Smith
12 Genetics������������������������������������������������������������������������������������������  145
James J. O’Byrne and Andrew J. Green
13 Ethical
 Considerations in Pediatric Surgery ������������������������������  155
Rita D. Shelby, Donna A. Caniano, and Benedict C. Nwomeh

ix
x Contents

14 Minimal
 Access Surgery in Infants and Children����������������������  163
Amulya K. Saxena, Roberta V. Iacona, and Keith Georgeson
15 Surgical
 Safety in Children ����������������������������������������������������������  177
Iain Yardley and Holbrook Charlotte
16 Surgical
 Problems of Children with Physical Disabilities����������  185
Casey M. Calkins
17 Surgical
 Aspects of HIV Infection in Children����������������������������  203
Alastair J. W. Millar, Brian Eley, and Sharon Cox

Part II Trauma

18 Birth Trauma����������������������������������������������������������������������������������  219

Thambipillai Sri Paran and Prem Puri
19 Pediatric Thoracic Trauma ����������������������������������������������������������  229
David E. Sawaya, Michael W. Morris, and Paul M. Colombani
20 Abdominal and Genitourinary Trauma��������������������������������������  239
Claire D. Gerall, Vincent P. Duron, and Steven Stylianos
21 Surgical
 Treatment of Severe Head Trauma ������������������������������  261
Hans G. Eder
22 Pediatric Orthopedic Trauma������������������������������������������������������  273
Zacharias Zachariou, Eva E. Fischerauer, and Annelie M.
Weinberg
23 Injuries
 to the Tendons of the Hand��������������������������������������������  295
Georg Singer and Heidi Friedrich
24 Burns ����������������������������������������������������������������������������������������������  309
Alan David Rogers and Heinz Rode
25 Foreign Bodies��������������������������������������������������������������������������������  325
S. Shah, L. Nguyen, and R. Sun
26 Physical
 and Sexual Child Abuse��������������������������������������������������  337
Michael E. Höllwarth

Part III Head and Neck

27 Pierre Robin Sequence������������������������������������������������������������������  349

Udo Rolle, Aranka Ifert, and Robert Sader
28 Choanal Atresia������������������������������������������������������������������������������  359
R. Ben Speaker, Michael Harney, and John Russell
29 Thyroglossal
 and Branchial Cysts, Sinuses, and Fistulas����������  365
Michael E. Höllwarth
30 Tracheostomy ��������������������������������������������������������������������������������  373
Lina Woods, Thom E. Lobe, and John Russell
Contents xi

Part IV Chest

31 Chest Wall Deformities������������������������������������������������������������������  387

Robert E. Kelly and Donald Nuss
32 Breast
 Disorders in Children and Adolescents����������������������������  405
Steffi Mayer, Jan-Hendrik Gosemann, Benno M. Ure,
and Martin L. Metzelder
33 Congenital Airway Malformations ����������������������������������������������  413
Patricio Varela and Richard Azizkhan
34 Mediastinal
 Masses in Children ��������������������������������������������������  429
Maria Molina and Israel Fernandez-Pineda
35 Pleural
 Effusion and Empyema����������������������������������������������������  437
Michael Singh and Dakshesh Parikh
36 Congenital
 Malformations of the Lung����������������������������������������  447
Ali A. Mokdad, David M. Gourlay, and Keith T. Oldham
37 Congenital Diaphragmatic Hernia ����������������������������������������������  463
Prem Puri and Nana Nakazawa
38 Extracorporeal Membrane Oxygenation������������������������������������  475
Brian P. Fallon, Samir K. Gadepalli, Jason S. Frischer,
Charles J. H. Stolar, and Ronald B. Hirschl

Part V Spina Bifida and Hydrocephalus

39 Spina
 Bifida and Encephalocoele�������������������������������������������������  487
Martin T. Corbally
40 Hydrocephalus��������������������������������������������������������������������������������  499
Geraint Sunderland, Jonathan Ellenbogen,
and Conor Mallucci
41 Dermal
 Sinus Tract and Tethered Cord Syndrome��������������������  527
Geraint Sunderland and Jonathan Ellenbogen

Part VI Anterior Abdominal Wall Defects

42 Omphalomesenteric Duct Remnants��������������������������������������������  543

Ampaipan Boonthai, Dhanya Mullassery, and Paul D. Losty
43 Omphalocele and Gastroschisis����������������������������������������������������  551
Leah M. Sieren, Duane S. Duke, and Marshall Z. Schwartz
44 Conjoined Twins����������������������������������������������������������������������������  563
Juan A. Tovar and Leopoldo Martinez
xii Contents

Part VII Tumors

45 Vascular Anomalies������������������������������������������������������������������������  577

Anna McGuire, Steven J. Fishman, and Belinda Hsi Dickie
46 Congenital Nevi������������������������������������������������������������������������������  593
Hannes Prescher, Julia F. Corcoran, and Bruce S. Bauer
47 Lymphatic Malformations������������������������������������������������������������  609
Jeremy A. Goss, Mohammed Zamakhshary, Jacob C. Langer,
and Emily Christison-Lagay
48 Sacrococcygeal Teratoma��������������������������������������������������������������  625
Thambipillai Sri Paran and Prem Puri
49 Neuroblastoma ������������������������������������������������������������������������������  633
Edward Kiely
50 Soft Tissue Sarcomas ��������������������������������������������������������������������  643
Sandeep Agarwala and Robert Carachi
51 Lymphomas������������������������������������������������������������������������������������  661
Christian Urban
52 Wilms’ Tumor��������������������������������������������������������������������������������  673
Michael E. Höllwarth
53 Ovarian Tumors ����������������������������������������������������������������������������  685
Alicia G. Sykes, Mary E. Fallat, and Romeo C. Ignacio Jr.
54 Testicular Tumors��������������������������������������������������������������������������  699
Amanda F. Saltzman and Jonathan Ross

Part VIII Gastrointestinal

55 Esophageal Atresia and Tracheoesophageal Fistula ������������������  711

Michael E. Höllwarth and Paola Zaupa
56 Achalasia����������������������������������������������������������������������������������������  729
Fanny Yeung, Kenneth Wong, and Paul Tam
57 Esophageal
 Perforations and Caustic
Injuries in Children ����������������������������������������������������������������������  743
Shilpa Sharma and Devendra K. Gupta
58 Gastroesophageal Reflux Disease ������������������������������������������������  753
Michael E. Höllwarth and Valeria Solari
59 Esophageal Replacement��������������������������������������������������������������  777
Shilpa Sharma and Devendra K. Gupta
60 Infantile
 Hypertrophic Pyloric Stenosis��������������������������������������  799
Takashi Doi and Takao Fujimoto
Contents xiii

61 Gastrostomy and Jejunostomy ����������������������������������������������������  807

Julia Brendel and Michael W. L. Gauderer
62 Duodenal Obstruction ������������������������������������������������������������������  829
Yechiel Sweed and Arcady Vachyan
63 Intestinal Malrotation��������������������������������������������������������������������  849
Mark D Stringer and Prabal R. Mishra
64 Jejuno-Ileal Atresia������������������������������������������������������������������������  869
A. J. W. Millar, S. Cox, and A. Numanoglu
65 Meconium Ileus������������������������������������������������������������������������������  881
Valeria Solari and Massimo Rivosecchi
66 Duplications of the Alimentary Tract������������������������������������������  893
K. Hughes, A. Mortell, and Prem Puri
67 Necrotizing Enterocolitis ��������������������������������������������������������������  907
Heather L. Liebe, Henri R. Ford, Victoria Camerini,
and Catherine J. Hunter
68 Constipation�����������������������������������������������������������������������������������  919
Hannah M. E. Evans-Barns, Sebastian K. King,
Bridget R. Southwell, and John M. Hutson
69 Hirschsprung’s Disease������������������������������������������������������������������  933
Prem Puri
70 Variant Hirschsprung’s Disease����������������������������������������������������  949
Florian Friedmacher and Prem Puri
71 Anorectal Anomalies����������������������������������������������������������������������  967
Alberto Peña, Andrea Bischoff, and Luis De la Torre
72 Appendicitis������������������������������������������������������������������������������������  985
Markus Almström and Tomas Wester
73 Intussusception������������������������������������������������������������������������������  993
Holger Till and Erich Sorantin
74 Hernias�������������������������������������������������������������������������������������������� 1001
Anna Svenningsson and Tomas Wester
75 Short Bowel Syndrome������������������������������������������������������������������ 1015
Michael E. Höllwarth
76 Inflammatory Bowel Disease�������������������������������������������������������� 1031
Risto J. Rintala and Mikko P. Pakarinen
77 Paediatric Small Bowel Transplantation ������������������������������������ 1051
G. L. Gupte, K. Sharif, and A. J. W. Millar
78 Long-Term
 Outcomes in Pediatric Surgery�������������������������������� 1061
Risto J. Rintala and Mikko P. Pakarinen
xiv Contents

Part IX Hepatobiliary

79 Biliary Atresia�������������������������������������������������������������������������������� 1091

Hannah Thompson and Mark Davenport
80 Choledochal Cyst �������������������������������������������������������������������������� 1101
Hiroyuki Koga and Atsuyuki Yamataka
81 Hepatic Cysts and Abscesses �������������������������������������������������������� 1117
Priya Ramachandran
82 Portal Hypertension���������������������������������������������������������������������� 1123
Mark D Stringer
83 Gallbladder Disease ���������������������������������������������������������������������� 1141
Charlene Dekonenko, Shawn D. St. Peter,
and George W. Holcomb III
84 Pancreatic Disorders���������������������������������������������������������������������� 1155
Elke Zani-Ruttenstock and Augusto Zani
85 Splenic Disorders �������������������������������������������������������������������������� 1173
Takashi Doi and Thom E. Lobe
86 Hepatic Tumors in Childhood������������������������������������������������������ 1185
Thambipillai Sri Paran and Michael P. La Quaglia
87 Pediatric Liver Transplantation �������������������������������������������������� 1197
Khalid Sharif and Alastair J. W. Millar

Part X Genitourinary Disorders

88 Urinary Tract Infection ���������������������������������������������������������������� 1215

Thomas de los Reyes and Martin A. Koyle
89 Imaging
 of the Paediatric Urogenital Tract�������������������������������� 1227
Michael Riccabona
90 Management of Antenatal Hydronephrosis�������������������������������� 1249
Jack S. Elder
91 Upper Urinary Tract Obstructions���������������������������������������������� 1269
Leon Chertin and Boris Chertin
92 Ureteric Duplication Anomalies���������������������������������������������������� 1281
Thambipillai Sri Paran and Prem Puri
93 Vesicoureteral Reflux�������������������������������������������������������������������� 1291
Prem Puri and Balazs Kutasy
94 Posterior Urethral Valves�������������������������������������������������������������� 1307
Salvatore Cascio, David Coyle, Simona Nappo,
and Paolo Caione
Contents xv

95 Neuropathic Bladder��������������������������������������������������������������������� 1329

Alaa El Ghoneimi, Annabel Paye-Jaouen, Valeska Bidault,
Pauline Lopez, and Matthieu Peycelon
96 Bladder Exstrophy ������������������������������������������������������������������������ 1347
Wayland J. Wu and John P. Gearhart
97 Cloacal Exstrophy�������������������������������������������������������������������������� 1359
Duncan Wilcox, Andrea Bischoff, and Moritz M. Ziegler
98 Prune Belly Syndrome ������������������������������������������������������������������ 1371
Thambipillai Sri Paran and Prem Puri
99 End-Stage
 Renal Disease and Renal Transplantation���������������� 1377
Atif Awan and Michael Riordan
100 Different Sexual Development������������������������������������������������������ 1389
Maria Marcela Bailez, Gabriela Guercio, and Santiago Weller
101 Cryptorchidism������������������������������������������������������������������������������ 1415
John M. Hutson and Jaya Vikraman
102 Acute Scrotum�������������������������������������������������������������������������������� 1425
Amulya K. Saxena, Matthew Jobson, and Michael Höllwarth
103 Hypospadias����������������������������������������������������������������������������������� 1435
Agneta Nordenskjöld and Göran Läckgren
104 Circumcision and Buried Penis���������������������������������������������������� 1451
Martin Kaefer
105 Hydrometrocolpos�������������������������������������������������������������������������� 1469
Shilpa Sharma and Devendra K. Gupta
106 Gynaecologic Conditions of Childhood���������������������������������������� 1479
Hazel Isabella Learner and Sarah M. Creighton
Index�������������������������������������������������������������������������������������������������������� 1489
Contributors

Sandeep Agarwala  Department of Pediatric Surgery, All India Institute of

Medical Sciences, New Delhi, India
Markus  Almström Department of Pediatric Surgery, Astrid Lindgren
Children’s Hospital, Karolinska University Hospital, Stockholm, Sweden
Atif Awan  Department of Nephrology & Transplantation, Children’s Health
Ireland at Temple Street, Dublin, Ireland
Richard  Azizkhan Cincinnati Children’s Hospital Medical Center,
Cincinnati, OH, USA
Maria  Marcela  Bailez Department of Pediatric Surgery, Hospital de
Pediatría “Prof. Dr. J.P. Garrahan”, Buenos Aires, Argentina
Bruce S. Bauer  Section of Plastic and Reconstructive Surgery, University of
Chicago, Pritzker School of Medicine, Chicago, IL, USA
Valeska Bidault  Department of Pediatric Urology, Robert-Debré University
Hospital, APHP, National Reference Center of Rare Urinary Tract
Malformations (MARVU), Université Paris Cité, Paris, France
Andrea Bischoff  International Center for Colorectal and Urogenital Care,
Children’s Hospital Colorado, University of Colorado School of Medicine,
Aurora, CO, USA
Rebecca Black  Oxford University Hospitals NHS Foundation Trust, Oxford,
UK
Ampaipan  Boonthai Paediatric Surgery and Transplantation Consultant
Surgeon, Division of Pediatric Surgery, Ramathibodi Hospital, Mahidol
University, Bangkok, Thailand
Julia Brendel  Department of Pediatric Surgery, Hannover Medical School,
Hannover, Germany
Paolo Caione  Division of Paediatric Urology, “Bambino Gesù” Children’s
Hospital, Rome, Italy
Casey  M.  Calkins Division of Pediatric Surgery, Medical College of
Wisconsin, Children’s Hospital of Wisconsin, Milwaukee, WI, USA
Victoria  Camerini Children’s Hospital Los Angeles, Los Angeles, CA,
USA
xvii
xviii Contributors

Donna  A.  Caniano Department of Surgery and Pediatrics, Ohio State

University College of Medicine, Nationwide Children’s Hospital, Columbus,
OH, USA
Robert Carachi  Department of Surgical Pediatrics, University of Glasgow,
Queen Elizabeth University Hospital, Glasgow, UK
Salvatore Cascio  University College Dublin and Children’s Health Ireland,
Dublin, Ireland
Holbrook Charlotte  Evelina London Children’s Hospital, London, UK
Boris Chertin  Department of Urology, Hebrew University, Jerusalem, Israel
Leon  Chertin Sakler School of Medicine, Tel Aviv University, Tel Aviv,
Israel
Emily  Christison-Lagay Division of Pediatric Surgery, Department of
Surgery, Yale University, New Haven, CT, USA
Paul  M.  Colombani The Johns Hopkins Hospital Professor Emeritus,
Baltimore, MD, USA
Martin T. Corbally  Royal College of Surgeons in Ireland, Dublin, Ireland
Julia F. Corcoran  Section of Plastic and Reconstructive Surgery, University
of Illinois Chicago, Chicago, IL, USA
Sharon  Cox University of Cape Town and Red Cross War Memorial
Children’s Hospital, Cape Town, South Africa
David Coyle  Children’s Health Ireland, Dublin, Ireland
Sarah  M.  Creighton Department of Women’s Health University College
London Hospitals, London, UK
Suzanne  Crowe  Children’s Health Ireland, Crumlin, Dublin, Republic of
Ireland
Mark Davenport  Department of Paediatric Surgery, Kings College Hospital,
London, UK
Luis De la Torre  International Center for Colorectal and Urogenital Care,
Children’s Hospital Colorado, Aurora, CO, USA
Thomas  de los Reyes The Hospital for Sick Children and University of
Toronto, Toronto, ON, Canada
Charlene Dekonenko  Children’s Mercy Hospital, Kansas City, MO, USA
Belinda  Hsi  Dickie Vascular Anomalies Center, Department of Surgery,
Boston Children’s Hospital, Boston, MA, USA
Takashi Doi  Department of Pediatric Surgery, Kansai Medical University,
Osaka, Japan
Duane S. Duke  Children’s Hospital of the King’s Daughter, Norfolk, VA,
USA
Contributors xix

Vincent  P.  Duron Division of Pediatric Surgery, Columbia University

Vagelos College of Physicians & Surgeons, NYP-Morgan Stanley Children’s
Hospital, New York, NY, USA
Simon Eaton  Great Ormond Street Hospital and Institute of Child Health,
London, UK
Hans  G.  Eder Department of Neurosurgery Medical University of Graz,
Graz, Austria
Alaa  El Ghoneimi Department of Pediatric Urology, Robert-Debré
University Hospital, APHP, National Reference Center of Rare Urinary Tract
Malformations (MARVU), Université Paris Cité, Paris, France
Jack  S.  Elder Division of Pediatric Urology, Massachusetts General
Hospital, Boston, MA, USA
Brian  Eley University of Cape Town and Red Cross War Memorial
Children’s Hospital, Cape Town, South Africa
Jonathan Ellenbogen  Department of Neurosurgery, Alder Hey Children’s
NHS Foundation Trust, Liverpool, UK
Hannah M.E. Evans-Barns  Department of Paediatric Surgery, The Royal
Children’s Hospital and Surgical Research Group, Murdoch Children’s
Research Institute, Melbourne, VIC, Australia
Mary E. Fallat  The Hiram C. Polk, Jr., Department of Surgery/Division of
Pediatric Surgery, University of Louisville School of Medicine, Louisville,
KY, USA
Brian P. Fallon  University of Michigan, Ann Arbor, MI, USA
Israel  Fernandez-Pineda Department of Pediatric Surgery, Virgen del
Rocio Children’s Hospital, Sevilla, Spain
Eva  E.  Fischerauer University Clinic of Paediatric and Adolesecent
Surgery, Medical University, Graz, Austria
Steven  J.  Fishman Vascular Anomalies Center, Department of Surgery,
Boston Children’s Hospital, Boston, MA, USA
Henri R. Ford  University of Miami, Miami, CA, USA
Florian Friedmacher  Department of Pediatric Surgery, University Hospital
Frankfurt, Goethe University Frankfurt, Frankfurt (Main), Germany
Heidi  Friedrich University Clinic of Pediatric and Adolescent Surgery,
Medical University of Graz, Graz, Austria
Jason  S.  Frischer Cincinnati Children’s Hospital Medical Center,
Cincinnati, OH, USA
Takao Fujimoto  Fujimoto Children’s Clinic, Tokyo, Japan
Samir K. Gadepalli  University of Michigan, Ann Arbor, MI, USA
xx Contributors

Michael  W.L.  Gauderer Surgery and Pediatrics, University of South

Carolina, Greenville, SC, USA
University of South Carolina, School of Medicine Greenville, Greenville, SC,
USA
John  P.  Gearhart James Buchannan Brady Urological Institute, Roberts
Jeffs Division of Pediatric Urology, Charlotte Bloomberg Children’s Hospital,
Johns Hopkins University School of Medicine, Baltimore, MD, USA
Keith Georgeson  Department of Paediatric Surgery, University of Alabama
Health Services, Birmingham, AL, USA
Claire D. Gerall  Division of Pediatric Surgery, Columbia University Vagelos
College of Physicians & Surgeons, NYP-Morgan Stanley Children’s Hospital,
New York, NY, USA
John Gillick  Children’s Health Ireland at Temple Street, Dublin, Ireland
Jan-Hendrik  Gosemann Department of Pediatric Surgery, University of
Leipzig, Leipzig, Germany
Jeremy A. Goss  Division of Plastic and Reconstructive Surgery, Department
of Surgery, Yale University School of Medicine, New Haven, CT, USA
David  M.  Gourlay Division of Pediatric Surgery, Medical College of
Wisconsin, Milwaukee, WI, USA
Andrew  J.  Green School of Medicine and Medical Science, University
College Dublin, National Centre for Medical Genetics Our Lady’s Children’s
Hospital, Crumlin, Dublin, Ireland
Gabriela  Guercio Department of Endocrinology, Hospital de Pediatría
“Prof. Dr. J.P. Garrahan” y CONICET, Buenos Aires, Argentina
Devendra K. Gupta  Super Speciality Pediatric Hospital and Post Graduate
Teaching Institute, Noida, India
G.L. Gupte  Birmingham Children’s Hospital, Birmingham, UK
Michael Harney  Bon Secours Hospital, Cork, Ireland
Ronald B. Hirschl  C.S. Mott Children’s Hospital, University of Michigan,
Ann Arbor, MI, USA
Michael  E.  Höllwarth University Clinic of Paediatric and Adolescent
Surgery, Medical University of Graz, Graz, Austria
George  W.  Holcomb III Children’s Mercy Hospital, Kansas City, MO,
USA
K. Hughes  Children’s Health Ireland (CHI), Dublin, Ireland
Catherine  J.  Hunter Children’s Hospital at Oklahoma University,
Oklahoma City, OK, USA
Contributors xxi

John  M.  Hutson  University Clinic of Paediatric and Adolescent Surgery,

Medical University of Graz, Graz, Austria
Roberta  V.  Iacona Department of Paediatric Surgery, Chelsea and
Westminster Hospital NHS Foundation Trust, Imperial College London,
London, UK
Aranka Ifert  Carolinum, Institute of Dentistry, Frankfurt, Germany
Romeo C. Ignacio Jr  Department of Surgery/Division of Pediatric Surgery,
University of California San Diego School of Medicine, La Jolla, CA, USA
Edwin C. Jesudason  Department of Rehabilitation Medicine, Astley Ainslie
Hospital, Edinburgh, UK
University of Edinburgh, Edinburgh, UK
Matthew Jobson  NHS Trust: The Luton and Dunstable University Hospitals
NHS Foundation Trust, London, UK
Martin  Kaefer Department of Urology, Division of Pediatric Urology,
Indiana University School of Medicine, Indianapolis, IN, USA
Robert E. Kelly  Children’s Hospital of The King’s Daughters, Norfolk, VA,
USA
Eastern Virginia Medical School, Norfolk, VA, USA
Edward Kiely  Great Ormond Street Hospital for Children NHS Foundation
Trust, London, UK
Sebastian K. King  Department of Paediatric Surgery, The Royal Children’s
Hospital and Surgical Research Group, Murdoch Children’s Research
Institute, Melbourne, VIC, Australia
Surgical Research Group, Murdoch Children’s Research Institute, Melbourne,
VIC, Australia
Hiroyuki Koga  Department of Pediatric and Urogenital Surgery, Juntendo
University School of Medicine, Bunkyo-ku Tokyo, Japan
Martin A. Koyle  The Hospital for Sick Children and University of Toronto,
Toronto, ON, Canada
Balazs Kutasy  Astrid Lindgren’s Children’s Hospital, Karolinska University
Hospital, Stockholm, Sweden
Michael P. La Quaglia  Pediatric Surgery, Memorial Sloan-Kettering Cancer
Center, New York, NY, USA
Göran  Läckgren Department of Pediatric Surgery, Uppsala Academic
Hospital, Uppsala, Sweden
Kokila  Lakhoo  Children’s Hospital Oxford, Oxford University Hospitals,
University of Oxford, Oxford, UK
xxii Contributors

Jacob  C.  Langer  Division of General and Thoracic Surgery, Hospital for
Sick Children and Department of Surgery, School of Medicine Surgery,
University of Toronto, Toronto, ON, Canada
Hazel Isabella Learner  Department of Women’s Health University College
London Hospitals, London, UK
Heather L. Liebe  Children’s Hospital at Oklahoma University, Oklahoma
City, OK, USA
Thom E. Lobe  University of Tennessee Health Science Center, Memphis,
TN, USA
Pediatric Surgery, University of Illinois, Chicago, IL, USA
Pauline Lopez  Department of Pediatric Urology, Robert-Debré University
Hospital, APHP, National Reference Center of Rare Urinary Tract
Malformations (MARVU), Université Paris Cité, Paris, France
Paul D. Losty  Alder Hey Children’s Hospital NHS Foundation Trust, School
of Health and Life Science, University of Liverpool, Liverpool, UK
Conor Mallucci  Department of Neurosurgery, Alder Hey Children’s NHS
Foundation Trust, Liverpool, UK
Leopoldo  Martinez Departamento de Cirugía Pediátrica, Hospital
Universitario La Paz, Madrid, Spain
Steffi  Mayer Department of Pediatric Surgery, University of Leipzig,
Leipzig, Germany
Peter  McCarthy National Children’s Cancer Service, Children’s Health
Ireland at Crumlin, Dublin, Ireland
Systems Biology Ireland, School of Medicine, University College Dublin,
Dublin, Ireland
Anna McGuire  Vascular Anomalies Center, Department of Surgery, Boston
Children’s Hospital, Boston, MA, USA
Martin L. Metzelder  Department of Pediatric Surgery, University Clinic of
Surgery, Medical University of Vienna, Vienna, Austria
Alastair J.W. Millar  University of Cape Town and Red Cross War Memorial
Children’s Hospital, Cape Town, South Africa
Prabal  R.  Mishra Departments of Paediatric Surgery & Child Health,
Wellington Hospital, Wellington, New Zealand
Ali A. Mokdad  Division of Pediatric Surgery, Medical College of Wisconsin,
Milwaukee, WI, USA
Maria Molina  Department of Pediatric Surgery, Virgen del Rocio Children’s
Hospital, Sevilla, Spain
Michael W. Morris  University of Mississippi Medical Center, Jackson, MS,
USA
Contributors xxiii

A. Mortell  Children’s Health Ireland (CHI), Dublin, Ireland

Dhanya  Mullassery  Consultant Paediatric Surgeon, Great Ormond Street
Hospital for Sick Children, University College London, London, UK
Ancuta  Muntean Children’s Health Ireland at Temple Street, Dublin,
Ireland
Nana  Nakazawa Department of Pediatric Surgery, Juntendo University
Nerima Hospital, Tokyo, Japan
Simona Nappo  Division of Paediatric Urology, “Bambino Gesù” Children’s
Hospital, Rome, Italy
L. Nguyen  Montreal Children’s Hospital, Montreal, QC, Canada
Gregory  Nolan Children’s Health Ireland, Crumlin, Dublin, Republic of
Ireland
Agneta  Nordenskjöld Department of Pediatric Surgery, Karolinska
University Hospital, Stockholm, Sweden
Department of Women’s and Children’s Health, Karolinska Institutet,
Stockholm, Sweden
A.  Numanoglu University of Cape Town and Red Cross War Memorial
Children’s Hospital, Cape Town, South Africa
Donald  Nuss  Children’s Hospital of The King’s Daughters, Norfolk, VA,
USA
Eastern Virginia Medical School, Norfolk, VA, USA
Benedict  C.  Nwomeh Department of Surgery and Pediatrics, Ohio State
University College of Medicine, Nationwide Children’s Hospital, Columbus,
OH, USA
James J. O’Byrne  National Centre for Inherited Metabolic Disorders, Mater
Misericordiae Hospital, Dublin, Ireland and School of Medicine, University
College Dublin, Dublin, Ireland
Keith  T.  Oldham Division of Pediatric Surgery, Medical College of
Wisconsin, Milwaukee, WI, USA
S. O’Sullivan  Paediatric Intensive Care Unit, Our Lady’s Children’s Hospital
Crumlin, Dublin, Ireland
Mikko P. Pakarinen  Department of Pediatric Surgery, Hospital for Children
and Adolescents, University of Helsinki, Helsinki, Finland
Thambipillai  Sri  Paran Children’s Health Ireland at Crumlin, Dublin,
Ireland and Trinity College Dublin and University College Dublin, Dublin,
Ireland
Dakshesh Parikh  Department of Paediatric Surgery, Birmingham Children’s
Hospital, Birmingham, UK
xxiv Contributors

Annabel  Paye-Jaouen Department of Pediatric Urology, Robert-Debré

University Hospital, APHP, National Reference Center of Rare Urinary Tract
Malformations (MARVU), Université Paris Cité, Paris, France
Alberto  Peña International Center for Colorectal and Urogenital Care,
Children’s Hospital Colorado, University of Colorado School of Medicine,
Aurora, CO, USA
Matthieu  Peycelon Department of Pediatric Urology, Robert-Debré
University Hospital, APHP, National Reference Center of Rare Urinary Tract
Malformations (MARVU), Université Paris Cité, Paris, France
Agostino  Pierro  Division of General and Thoracic Surgery, Translational
Medicine Program, The Hospital for Sick Children, University of Toronto,
Toronto, ON, Canada
Hannes Prescher  University of Chicago Medicine & Biological Sciences,
Chicago, IL, USA
Prem Puri  Department of Pediatric Surgery, Beacon Hospital, and University
College Dublin, Dublin, Ireland
Priya Ramachandran  Department of Pediatric Surgery, Kanchi Kamakoti
CHILDS Trust Hospital, Chennai, Tamil Nadu, India
Michael Riccabona  Department of Pediatric Radiology, University Clinic
of Radiology, Medical University Graz, Graz, Austria
Risto  J.  Rintala Department of Pediatric Surgery, Children’s Hospital,
Helsinki University Central Hospital, Helsinki, Finland
Michael Riordan  Department of Nephrology & Transplantation, Children’s
Health Ireland at Temple Street, Dublin, Ireland
Massimo  Rivosecchi Department of Pediatric Surgery, “Bambino Gesù“
Children’s Hospital, Palidoro, Rome, Italy
Heinz  Rode Division of Paediatric Surgery, Department of Surgery,
University of Cape Town, Cape Town, South Africa and Red Cross War
Memorial Children’s Hospital, Cape Town, South Africa
Alan David Rogers  Ross Tilley Burn Centre, Sunnybrook Health Sciences
Centre, Toronto, Canada
Division of Plastic and Reconstructive Surgery, Department of Surgery,
University of Toronto, Toronto, ON, Canada
Udo Rolle  Department of Pediatric Surgery, University Hospital Frankfurt,
Frankfurt/Main, Germany
Department of Pediatric Surgery, Goethe University Frankfurt, Frankfurt,
Germany
Jonathan Ross  Department of Urology, Rush University, Chicago, IL, USA
Contributors xxv

John  Russell Department of Pediatric Otorhinolaryngology, Children’s

Health Ireland, Crumlin, Dublin, Ireland
Robert Sader  Department of Oral, Maxillofacial, and Plastic Facial Surgery,
Goethe University Frankfurt, Frankfurt, Germany
Amanda  F.  Saltzman Department of Urology, University of Kentucky,
Lexington, KY, USA
David E. Sawaya  University of Mississippi Medical Center, Jackson, MS,
USA
Amulya  K.  Saxena Department of Paediatric Surgery, Chelsea and
Westminster Hospital NHS Foundation Trust, Imperial College London,
London, UK
Marshall  Z.  Schwartz  Department of Surgery and Urology, Wake Forest
University School of Medicine, Wake Forest Institute for Regenerative
Medicine, Wake Forest Baptist Health, Winston-Salem, NC, USA
S. Shah  Baylor College of Medicine, Houston, TX, USA
Khalid Sharif  Birmingham Women’s and Children’s Hospital, Birmingham,
UK
Shilpa  Sharma Department of Pediatric Surgery, All India Institute of
Medical Sciences, New Delhi, India
Rita D. Shelby  Ohio State University Wexner Medical Center, Columbus,
OH, USA
Leah M. Sieren  Wake Forest Baptist Health, Winston-Salem, NC, USA
Georg  Singer University Clinic of Pediatric and Adolescent Surgery,
Medical University of Graz, Graz, Austria
Michael  Singh  Department of Paediatric Surgery, Birmingham Children’s
Hospital, Birmingham, UK
Owen Patrick Smith  National Children’s Cancer Service, Children’s Health
Ireland at Crumlin, Dublin, Ireland
Systems Biology Ireland, School of Medicine, University College Dublin,
Dublin, Ireland
Coilin  Collins  Smyth Our Ladys Children’s Hospital, Crumlin, Dublin,
Ireland
Valeria Solari  Department of Pediatric Surgery, Klinik Donaustadt, Vienna,
Austria
Erich  Sorantin  Department of Paediatric Radiology, University Clinic of
Radiology, Medical University of Graz, Graz, Austria
xxvi Contributors

Bridget  R.  Southwell Surgical Research Group, Murdoch Children’s

Research Institute, Melbourne, VIC, Australia
R. Ben Speaker  Royal College of Surgeons Ireland, Dublin, Ireland
Shawn D. St. Peter  Children’s Mercy Hospital, Kansas City, MO, USA
Ionica Stoica  Children’s Health Ireland at Temple Street, Dublin, Ireland
Charles  J.H.  Stolar  Stanley Morgan Children’s Hospital, New York, NY,
USA
Mark D Stringer  Department of Paediatric Surgery, Wellington Children’s
Hospital and Department of Paediatrics and Child Health, Wellington School
of Medicine, University of Otago, Wellington, New Zealand
Steven  Stylianos Division of Pediatric Surgery, Columbia University
Vagelos College of Physicians & Surgeons, NYP-Morgan Stanley Children’s
Hospital, New York, NY, USA
R. Sun  Texas Children’s Hospital, Houston, TX, USA
Geraint  Sunderland Department of Neurosurgery, Alder Hey Children’s
NHS Foundation Trust, Liverpool, UK
Anna Svenningsson  Department of Pediatric Surgery, Karolinska University
Hospital, Stockholm, Sweden
Yechiel  Sweed Department of Pediatric Surgery, Galilee Medical Center,
Naharia, Israel
Bar Ilan University, Safed, Israel
Alicia G. Sykes  Department of General Surgery, Naval Medical Center, San
Diego, CA, USA
Paul  Tam Macau University of Science and Technology, Macau, SAR,
China
Hannah  Thompson Department of Paediatric Surgery, Kings College
Hospital, London, UK
Holger Till  University Clinic of Pediatric and Adolescent Surgery, Medical
University of Graz, Graz, Austria
Juan A. Tovar  Departamento de Cirugía Pediátrica, Hospital Universitario
La Paz, Madrid, Spain
Jeffrey S. Upperman  Department of Pediatric Surgery, Monroe Carell Jr.
Children’s Hospital at Vanderbilt, Nashville, CA, USA
Christian  Urban  Department of Pediatric Hematology/Oncology,
University Clinic of Pediatrics and Adolescent Medicine, Medical University
of Graz, Graz, Austria
Benno M. Ure  Hannover Medical School, Hannover, Germany
Contributors xxvii

Arcady  Vachyan Pediatric Advanced Laparoscopic Surgical Unit, Ruth

Rappaport Children’s Hospital, Rambam Medical Center, Haifa, Israel
Patricio Varela  Dr Luis Calvo Mackenna Children’s Hospital, Division of
Pediatric Surgery, Pediatric Airway and Chest Wall Malformation Center,
University of Chile, Santiago, Chile
Division of Pediatric Surgery, Clinica Alemana Medical Center, Santiago,
Chile
Lexie  H.  Vaughn Department of General Surgery, Vanderbilt University
Medical Center, Nashville, CA, USA
Jaya  Vikraman Department of Paediatrics, University of Melbourne,
Melbourne, VIC, Australia
Annelie  M.  Weinberg Department of Pediatric Traumatology, Johannes
Gutenberg University, Mainz, Germany
Santiago  Weller Department of Pediatric Urology, Hospital de Pediatría
“Prof. Dr. J.P. Garrahan”, Buenos Aires, Argentina
Tomas Wester  Department of Pediatric Surgery, Astrid Lindgren Children’s
Hospital, Karolinska University Hospital, Stockholm, Sweden
Duncan  Wilcox Department of Pediatric Urology, Children’s Hospital
Colorado, University of Colorado School of Medicine, Aurora, CO, USA
Kenneth  Wong University of Hong Kong, Queen Mary Hospital, Hong
Kong, SAR, China
Lina  Woods Department of Pediatric Otorhinolaryngology, Children’s
Health Ireland, Dublin, Ireland
Wayland J. Wu  Division of Pediatric Urology, Cohen Children’s Medical
Center of New York, Zucker School of Medicine at Hofstra/Northwell, New
Hyde Park, NY, USA
James Buchannan Brady Urological Institute, Roberts Jeffs Division of
Pediatric Urology, Charlotte Bloomberg Children’s Hospital, Johns Hopkins
University School of Medicine, Baltimore, MD, USA
Atsuyuki  Yamataka Department of Pediatric and Urogenital Surgery,
Juntendo University School of Medicine, Bunkyo-ku Tokyo, Japan
Iain Yardley  Evelina London Children’s Hospital, London, UK
Fanny Yeung  Queen Mary Hospital, Hong Kong, SAR, China
Zacharias Zachariou  Department of Pediatric Surgery, Makarios Children’s
Hospital, Nicosia, Cyprus Medical School, University of Cyprus, Nicosia,
Cyprus
Mohammed  Zamakhshary Department of Pediatric Surgery, Alfaisal
University, Riyadh, Saudi Arabia
xxviii Contributors

Augusto Zani  Division of General and Thoracic Surgery, The Hospital for

Sick Children, Toronto, Canada
Department of Surgery, University of Toronto, Toronto, Canada
Elke  Zani-Ruttenstock Division of General and Thoracic Surgery, The
Hospital for Sick Children, Toronto, Canada
Department of Surgery, University of Toronto, Toronto, Canada
Paola Zaupa  University Clinic of Pediatric and Adolescent Surgery, Medical
University of Graz, Graz, Austria
Moritz  M.  Ziegler  Children’s Hospital Colorado, University of Colorado
School of Medicine, Aurora, CO, USA
 Part I
General Principles
 The Epidemiology of Birth Defects
Florian Friedmacher and Edwin C. Jesudason
1.1 Introduction
Surgical correction of birth defects helped to
establish the specialty of pediatric surgery during
the mid-twentieth century. Around this time, pio-
neering neonatal operations were successfully
performed for the first time, enabling survival of
newborns with life-threatening conditions like
congenital diaphragmatic hernia (CDH) or
esophageal atresia. Indeed, high survival rates
can now be achieved for many previously fatal
anomalies, due to innovations like parenteral
nutrition as well as the concentration of surgical,
anesthetic and intensive care expertise. For cer-
tain birth defects that still show high rates of
mortality and morbidity (e.g. CDH or myelome-
ningocele), fetal surgery represents a promising
approach to harm reduction.
For many pediatric surgeons, birth defects
represent only a small part of their overall work,
F. Friedmacher (*)
Department of Pediatric Surgery, University Hospital
Frankfurt, Goethe University Frankfurt,
Frankfurt (Main), Germany
e-mail: Florian.Friedmacher@nhs.net
E. C. Jesudason
Department of Rehabilitation Medicine, Astley
Ainslie Hospital, Edinburgh, UK
University of Edinburgh, Edinburgh, UK
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_1
1
because rates are relatively low or afflicted chil-
dren do not survive to reach them. However, if
neonatal surgeons are properly able to judge their
outcomes and to advise expectant parents, they
need to have good understanding of the epidemi-
ology of particular birth defects in their catch-
ment area and to have a feel for which of the
above best explains their local rates.
1.1.1 Birth Defects Are Leading
Causes of Infant Mortality
and Long-term Morbidity
Worldwide
Due to the advances made in the treatment of
infectious diseases, birth defects have emerged as
one of the leading causes of infant death. This
pertains not only to countries with well-funded
healthcare systems but in fact anywhere that
infant mortality has fallen below 50 per 1000
births (Carmona 2005). Prevention is possible for
some birth defects. For example, congenital
rubella syndrome may be virtually eradicated by
adoption of an effective maternal vaccination
program (Banatvala and Brown 2004).
Nevertheless, a subset of neural tube defects con-
tinues to occur perhaps due to inadequate peri-­
conceptional folic acid supplementation
(Khoshnood et al. 2015) and lack of fortification
policies (Arth et al. 2016). For many other birth
defects, prevention does not seem possible, fea-
3
4 F. Friedmacher and E. C. Jesudason
sible or effective. In these circumstances, the epi-
demiological challenge extends beyond the
known, preventable defects to newer issues and
their potential consequences (e.g., COVID-19
pandemic or Zika virus epidemic) (Dang et  al.
2020; Pierson and Diamond 2018). Hence, birth
defects will continue to cause not only infant
deaths but also premature birth and long-term
morbidity. Initially, pediatric services tended to
focus on perinatal management of birth defects,
but the focus has now widened to include improv-
ing quality of life and the transition toward
­independent adulthood. This is not helped by the
separate development of pediatric and adult spe-
cialists, which means there can often be a lack of
meaningful engagement and communication
between the two. After all, it is likely that pediat-
ric surgeons can do more to address transition
and mitigate long-term sequelae (Rothstein and
Dasgupta 2016).
1.1.2 Birth Defects Epidemiology
and Teratology Have Emerged
from Outbreak Investigations
Modern birth defects epidemiology and teratol-
ogy arose from two hazards that became appar-
ent during the course of the last century: the
recognition of congenital rubella syndrome
(noted by clinical ophthalmological examina-
tion) (Cooper 1985) and the thalidomide disaster
(phocomelia and other defects associated with
maternal thalidomide administration for morn-
ing sickness) (Botting 2002). These unfortunate
events vividly illustrated the devastating conse-
quences of prenatal infection and drug exposure,
respectively. They highlighted the urgent need to
formalize reliable birth defect surveillance.
Today, this serves a wide range of purposes,
including early warning of outbreaks, identifica-
tion of potential environmental or genetic causes,
rational planning of neonatal services, facilita-
tion of prenatal counseling and the comparison
of outcomes (as a guide toward best practice)
(Khoury 1989).
1.1.3 Causation of Birth Defects
Remains Often Complex
and Poorly Understood
Before considering the methods of birth defect
surveillance, it is worth outlining the develop-
mental biology and embryology that underlie
birth defects (Donnai and Read 2003). Causes of
birth defects can be considered as parental, fetal
and environmental. In reality, they will often
overlap. For instance, grandparental behaviors
(or exposures) may produce epigenetic modifica-
tions that only manifest themselves in the devel-
oping fetus.
A familiar example of a parental factor is the
impact of maternal age on the prevalence of
Down syndrome (Loane et al. 2013). Moreover,
maternal conditions such as diabetes (Bell et al.
2012) and overweight/obesity (Stothard et  al.
2009) are well-described risk factors for the for-
mation of birth defects. Unfortunately, the role of
paternal age and/or exposures is more difficult to
quantify (Yang et  al. 2007). Fetal causes may
include genetically determined inborn errors of
metabolism such as those causing intersex anom-
alies in congenital adrenal hyperplasia, chromo-
somal lesions such as Down or Edwards
syndrome and twinning with its increased risk of
birth anomalies. Environmental causes include
those related to prenatal addiction and drug expo-
sure (e.g. alcohol, smoking, illicit drugs, thalido-
mide, valproate, phenytoin, warfarin) as well as
the impact of intrauterine infections (e.g. toxo-
plasmosis, rubella, cytomegalovirus) (Blotière
et al. 2019; Hackshaw et al. 2011; Jentink et al.
2010; Ernhart et al. 1987). In turn, the effect of
assisted reproductive technologies such as
in vitro fertilization and intracytoplasmic sperm
injection on the prevalence of birth defects is
quite difficult to assess (Tararbit et al. 2013). The
suggestion that anomaly rates are higher in
assisted pregnancies needs to take into account
several confounding factors, including the
increased rates of multiple pregnancy, parental
abnormalities and predispositions that may have
resulted in the need for assisted reproduction.
1  The Epidemiology of Birth Defects
Other environmental contributors to birth defects
may include “endocrine disrupting chemicals”
(McLachlan 2001). These estrogenic compounds
are conjectured to contribute to anomalies of sex-
ual development in fetal males (e.g. hypospadias)
as well as putative impairment of adult male
sperm quality (Colborn et  al. 1993). In light of
such difficulties in attributing causes, it is simpler
to admit that only the minority of birth defects
are known to arise from a simple environmental
or genetic cause. At present, the majority of birth
defects appear to have multifactorial origins. In
such circumstances, it is helpful to consider birth
defect causation as the result of complex interac-
tions between genes and environment. Hence,
some cases of spina bifida may result from micro-
nutrient deficiency in the context of predisposing
enzyme polymorphisms (Brody et  al. 2002).
Similarly, teratogenic drugs may interact with
pharmacogenomic predispositions to help explain
why only certain pregnancies are affected (Leeder
and Mitchell 2007). Beyond consideration of
complex causation, it remains likely that simple
chance has a major role to play, similar to the sto-
chastic effects noted in radiation biology
(Whitaker et al. 2003).
1.1.4 Birth Defects Appear to Arise
Typically (But Not Exclusively)
in the First Trimester
Developmental biologists refer to “competence
windows” to describe periods in development
when particular cells and tissues are capable of
responding appropriately to certain growth and
transcription factors (Johansson et al. 2007; Kim
et  al. 2005). In a similar manner, developing
organs are assumed to have particular temporal
windows when an otherwise non-specific terato-
genic stimulus will impact disproportionately on
formation of that organ system.
During the first trimester, organ morphogene-
sis predominates, while later trimesters are
devoted to organ growth and maturation.
Therefore, sensitivity to teratogens is held to
5
peak during the first trimester. Hence, pregnant
women are advised to avoid medications during
this part of gestation in particular. Teleologically,
morning sickness, which peaks during the first
trimester, is postulated to help reduce ingestion
of potential teratogens during this period of max-
imum vulnerability. While the model of first-­
trimester teratogenesis appears appropriate for
many birth defects, it is now clear that certain
anomalies appear to arise later as a result of fetal
events (e.g. amniotic band formation, intussus-
ception or vascular insult). Gastroschisis and
intestinal atresia may be considered in this latter
category (Feldkamp et al. 2007). Indeed, the dif-
ference between gastroschisis and omphalocele
in terms of associated anomalies (and hence
prognosis) may be explained by the different
times they are held to originate during develop-
ment. Omphalocele is considered an embryonic
lesion that is accompanied by contemporaneous
lesions of organogenesis in other systems such as
the heart. In contrast, gastroschisis is thought to
result from a discrete fetal vascular accident (like
the associated intestinal atresia) and thus lacks
extraintestinal manifestations. An alternative
view, however, is that intestinal atresiae result
only rarely from fetal accidents such as intussus-
ception and are in fact better understood as fail-
ures of mesenteric vascular development (Shorter
et al. 2006.). Similarly, a contrast between duode-
nal atresia and small bowel atresia may likewise
be understood as the result of their differing
onsets and etiologies. Duodenal atresia was his-
torically explained as an embryonic failure of
luminal recanalization, although this “solid core”
theory has been contradicted by more recent ani-
mal studies. The strong association between duo-
denal atresia and other defects (e.g. cardiac
defects, esophageal atresia and Down syndrome)
supports an embryonic origin of this malforma-
tion (Meio et al. 2008). In contrast, small bowel
atresiae are thought to follow mesenteric vascular
occlusion usually in fetal life (Nichol et al. 2011).
Hence, aside from gastroschisis, intestinal atre-
siae are unlikely to be associated with other
structural lesions. Between these two extremes
6 F. Friedmacher and E. C. Jesudason

are birth defects where an embryonic lesion has Table 1.1 Prevalence rates of congenital anomalies
deleterious knock-on effects later in fetal devel- according to EUROCAT subgroups including genetic
anomalies (2011–2018) (Available at: https://eu-­rd-­
opment. Based on experimental models, the neu- platform.jrc.ec.europa.eu/eurocat/eurocat-­d ata/
rological sequelae of spina bifida are postulated prevalence_en)
to result not only from the primary failure of neu- Prevalence per 10,000 births
ral tube closure but also from consequent expo- EUROCAT subgroups (95% CI)a
sure of the neural placode to amniotic fluid All anomaliesb 255.28 (253.97–256.60)
(Stiefel and Meuli 2007). Similarly, pulmonary – Congenital heart 79.76 (79.02–80.50)
hypoplasia in CDH may be explained by two defects
– Chromosomal 44.61 (44.06–45.16)
insults, one affecting both lungs before dia-
– Limb 38.09 (37.58–38.60)
phragm development and one affecting the ipsi- – Urinary 35.12 (34.63–35.61)
lateral lung after defective diaphragm – Nervous system 26.12 (25.70–26.54)
development (Keijzer et  al. 2000). In circum- – Genital 20.78 (20.41–21.16)
stances such as these, where the pathology is – Digestive system 18.65 (18.30–19.01)
thought to progress during fetal life, prenatal sur- – Oro-facial clefts 14.40 (14.09–14.71)
gical correction has been a logical proposal to – Abdominal wall 6.63 (6.42–6.85)
defects
meet the challenge of refractory mortality and – Respiratory 4.03 (3.87–4.20)
morbidity (Jancelewicz and Harrison 2009). – Eye 3.98 (3.82–4.15)
– Ear, face and neck 1.77 (1.66–1.88)
a
 Including live births, fetal deaths/stillbirths (from
1.1.5 Classification of Birth Defects 20 weeks of gestation) and termination of pregnancy for
for Epidemiological Purposes congenital anomaly
b
 Excluding cases with only minor anomalies. Cases with
more than one anomaly are only counted once
In general, the epidemiology of birth defects
derives from the registration of anomalies by
type. For instance, a European network of Down syndrome), it is likely that in the future,
population-­based registries for the epidemiologic anomalies may be classified or at least sub-
surveillance of congenital anomalies (EUROCAT) grouped by genotypic differences rather than
surveys over 1.7 million births (29% of European anatomic details alone. In certain cases, such
birth population) per year from multiple sources distinctions may be prognostically and therapeu-
in 21 countries in Europe (Morris et al. 2018). All tically important: e.g. omphalocele in Beckwith-
EUROCAT registries use a classification scheme Wiedemann syndrome is associated with the
based around organ systems (Table 1.1), specific additional hazards of hypoglycemia, macrosomia
diagnoses and International Classification of and increased tumor risk due to disordered gene
Diseases codes (Table  1.2). The “International imprinting (Piedrahita 2011). Hence, the actual
Clearinghouse for Birth Defects” (http://www. anatomical defect (i.e. omphalocele) becomes
icbdsr.org) is another good starting resource for less important than the genetics and its multisys-
pediatric surgeons wishing to know more about tem sequelae. Similarly, it is postulated that sub-
the epidemiology of birth defects. groups of spina bifida may be folate-resistant due
Cooperation between registries helps by pool- to underlying genetic/enzymatic variation (Pitkin
ing data and also by building consensus on cer- 2007; Brody et al. 2002). The design of precon-
tain issues like exclusion of minor anomalies ceptional prophylaxis for birth defects may need
without major and/or long-term sequelae (e.g. to acknowledge pharmacogenomically distinct
cryptorchidism or congenital hydrocele) or how subgroups to avoid benefits within one subgroup
abnormalities of gut fixation in CDH may be being overlooked due to a surrounding non-
recorded. Although birth defects are currently responder cohort.
classified by their structural anomaly (e.g. CDH, Having a system of classification is, however,
esophageal atresia) or defined diagnosis (e.g. only one part of the task. Notification and classi-
1  The Epidemiology of Birth Defects 7

Table 1.2  EUROCAT prevalence rates of congenital cases to be assigned later after remote assessment
anomalies according to disease codes including genetic of images by experts (Li et al. 2003). In addition,
anomalies (2011–2018) (Available at: https://eu-­rd-­
platform.jrc.ec.europa.eu/eurocat/eurocat-­d ata/
the photographs potentially allow the classifiers
prevalence_en) to calibrate their judgments against those from
Prevalence per other registries.
10,000 births (95%
Anomalies CI)a
Down syndrome 24.43 (24.02–24.84) 1.1.6 Counting of Birth Defects Is
Hypospadias 17.77 (17.42–18.12) Affected by the Definition
Congenital hydronephosis 12.82 (12.53–13.12)
of Stillbirth
Edward syndrome 5.99 (5.79–6.19)
Spina bifida 4.97 (4.79–5.16)
Multicystic renal dysplasia 4.36 (4.19–4.53) Birth defect epidemiology becomes difficult
Omphalocele 3.63 (3.47–3.79) whenever the classification of defects is not uni-
Anorectal atresia and stenosis 3.37 (3.22–3.52) form or straightforward. However, an equal chal-
Congenital diaphragmatic hernia 2.93 (2.79–3.08) lenge remains the counting of birth defects. This
Esophageal atresia with/without 2.68 (2.55–2.82) task is complicated by practical barriers to case
tracheoesophageal fistula
Gastroschisis 2.56 (2.43–2.69)
ascertainment (e.g. inadequate resources), the
Duodenal atresia or stenosis 1.45 (1.35–1.55) definition of stillbirth, and the effects of prenatal
Hirschsprung disease 1.44 (1.34–1.54) diagnosis and termination of pregnancy.
Posterior urethral valve and/or 1.27 (1.18–1.37) Recording of anomaly prevalence lies at the
prune belly core of birth defect epidemiology. To account for
Congenital pulmonary airway 1.17 (1.08–1.26) the unknowable incidence of a defect amongst
malformations
Small bowel atresia or stenosis 0.94 (0.87–1.03)
vast numbers of naturally miscarried pregnan-
Situs inversus 0.81 (0.74–0.89) cies, epidemiologists measure the prevalences of
Bladder extrophy and/or 0.62 (0.55–0.68) defects within a defined birth cohort: i.e. the
epispadia number of live and stillborn cases of the defect,
Congenital construction bands/ 0.56 (0.50–0.62) as a proportion of all births. This definition
amniotic bands
depends on the artificial distinction between mis-
Indeterminate sex 0.52 (0.47–0.59)
Biliary atresia 0.33 (0.28–0.38)
carriage and stillbirth: EUROCAT’s recommen-
Conjoined twins 0.16 (0.13–0.19) dation is that spontaneous pregnancy losses prior
a
 
Including live births, fetal deaths/stillbirths (from to 20 weeks of gestation are counted as miscar-
20 weeks of gestation) and termination of pregnancy for riages (and therefore do not contribute to anom-
congenital anomaly aly prevalence), while similar losses at 20 weeks
of gestation and beyond are counted as stillbirths
fication are subject to local variations in practice. (and thus are included in prevalence statistics).
When resources exist for expert-led classification Despite these guidelines, several countries
of birth defects by diagnosis, this approach to have established different demarcations (e.g. 24
birth defect epidemiology appears the best cur- or 28 weeks or even 500 g birth weight). Clearly,
rently available (Lin et  al. 2006). Nevertheless, some estimate of prenatal birth defects is required
even some North American registries lack clini- to avoid serious underestimation of overall preva-
cian input in the classification and assignment of lence (Duke et al. 2009). However, the demarca-
observed birth defects. The consequences of this tion of stillbirths begins to complicate matters.
omission for data quality remains to be deter- Countries using later gestational cut-offs may
mined. In the contrasting circumstances of rural underestimate birth defect prevalence compared
China, expert-led assignment of cases has been to registries where 20 weeks of gestation is used.
substituted with simple photographic recording Hence, minor changes in convention can lead to
of malformations. This system not only allows large but artificial differences in anomaly preva-
the registry to function but also allows difficult lence. While a definition of stillbirths is needed
8 F. Friedmacher and E. C. Jesudason
for data collection, the sharp demarcation
(whether 20  weeks or later) also appears arbi-
trary from a biological perspective. Consider a
hypothetical prenatal medical therapy that
reduces the prevalence of a specific birth defect.
When the anomaly is rare (as most are), it may
be difficult to determine whether an observed
reduction in prevalence is truly due to fewer mal-
formations or instead due to the promotion of
earlier loss of affected pregnancies (i.e. prior to
the 20th week or other agreed-upon margin).
This latter phenomenon, termed “terathanasia”,
has been invoked to explain how folate supple-
mentation might influence the prevalence of neu-
ral tube defects (Godwin et al. 2008; Hook and
Czeizel 1997).
1.1.7 Prenatal Diagnosis:
The Greatest Challenge
to Birth Defect Epidemiology?
The classification of birth defects and the defini-
tion of stillbirth make anomaly surveillance com-
plex. However, the impact of prenatal diagnosis
is arguably still more important. Prenatal diagno-
sis (in particular, non-specific ultrasound screen-
ing) confounds birth defects surveillance in a
number of ways:
• Prenatal diagnosis increases identification of
birth defects within the cohort of assessment
by diagnosing those who may otherwise have
perished prenatally (and were uncounted) or
those who may have presented beyond the
neonatal period (if at all). For instance, prena-
tal identification of cystic lung lesions. Some
would never have been diagnosed (e.g. either
regressing spontaneously or persisting asymp-
tomatically). Others would have presented
later (beyond the scope of the birth defects
registry).
• Prenatal diagnosis alters antenatal manage-
ment and results in abortions or fetal interven-
tions, which in turn affect the numbers of birth
defects being counted. Therefore, most regis-
tries attempt to keep separate data on termina-
tions for congenital anomalies, but these data
are hard to find when abortions are prohibited
by law.
• Prenatal diagnosis may be inaccurate and
unchecked. Pathological verification after ter-
mination of pregnancy may be incomplete or
absent, yet the presumptive diagnosis is
included in the birth defect list.
• The resources and expertise to perform prena-
tal sonography vary with location, thereby
hampering international comparison of birth
defect prevalences.
In summary, the apparently simple task of
counting live and stillborn cases for birth defect
surveillance is fraught with difficulty once the
arbitrary definition of stillbirth is imposed and
ubiquitous prenatal imaging prompts both termi-
nations and identification of previously occult
cases.
Given these challenges in data collection, epi-
demiologists are aided by being able to compare
a variety of surveillance databases. Many
European registries are incorporated into the
EUROCAT initiative. Similarly, several other
registries feed into birth defect surveillance data
furnished by the World Health Organization
(WHO). Their “Atlas of selected congenital
anomalies” is an interesting publication available
in the public domain (https://apps.who.int/nutri-
tion/publications/birthdefects_atlas/en/index.
html). Most importantly, it is instructive to read
and consider the caveats that EUROCAT and
WHO place upon their data. These interpreta-
tional issues not only highlight the problems dis-
cussed in the previous sections but also allude to
the ongoing challenge of inadequate resources
and expertise for birth defect reporting. This in
turn impairs the data accuracy and may help
explain insufficient action upon findings. Major
studies reinforce the logistical shortcomings of
birth defect reporting in the United Kingdom
(Boyd et al. 2005).
1  The Epidemiology of Birth Defects
1.1.8 Pediatric Surgeons Often
Focus on Their Institutional
Series of Birth Defects
Small institutional series are the staple of pediat-
ric surgeons’ reporting. However, several studies
have shown how institutional series are vulnera-
ble to bias and confounding (Mah et  al. 2009).
Indeed, these studies face broader problems than
population-based registries. Ascertainment
remains a particular issue. For example, prenatal
diagnosis, terminations or fetal deaths/stillbirths
prior to transfer can each give the misleading
impression that the institution is improving out-
comes when, in fact, extramural changes are
responsible. Moreover, pediatric surgeons try to
stratify for disease severity to show that their
(good) results are not simply the product of a
low-risk caseload. But in such circumstances, it
can be misleading to use the frequency of inter-
ventions to stratify for severity in a birth defect
cohort. For instance, in CDH, the decision to
patch and/or use extracorporeal membrane oxy-
genation and/or nitric oxide may owe in reality
more to institutional protocols than differences in
pathophysiology between cases. Ultimately,
institutional series are often subject to substantial
bias with apparently poorer results perhaps not
even being submitted for publication.
1.1.9 A “Life-Course” Approach
to Birth Defects
Given the aforementioned precautions, where
can pediatric surgery make progress? Advances
in prenatal imaging may improve prenatal prog-
nostication and case selection for fetal therapies
(Jancelewicz and Harrison 2009). However, bet-
ter imaging may also identify more defects of
questionable significance. To help families bal-
ance these developments, pediatric surgeons will
need to keep abreast of birth defect epidemiology
and to collaborate with other specialties. A larger
prize may be realized if pediatric surgery follows
epidemiology in widely adopting “life-course”
studies to assess the impact over time of correct-
9
ing birth defects surgically. Embracing a life-­
course approach means that, like pediatric
oncology, further thought and training can be
focused on the needs of teenage and young adult
survivors. Pediatric surgery can also adopt the
biopsychosocial model and the creative, problem-­
solving approach exemplified by, e.g., trauma
rehabilitation. These efforts may help the team
broach issues well before the hurdles of transi-
tional care. It may also allow such issues to be
addressed using the full breadth of non-surgical,
non-pharmacological approaches.
1.2 Conclusion and Future
Directions
It has been argued that birth defects are best inter-
rogated with a systems approach rather than via
molecular biology. A similar shift away from lin-
ear drug-receptor paradigms in non-communica-
ble adult diseases may alleviate the current
stagnation within the blockbuster drug pipeline
(DiMasi et al. 2004).
Computing is more powerful than ever and
data storage is becoming relatively cheap.
Therefore, intriguing possibilities exist these
days to use bigger data to do in-depth research
into birth defects epidemiology. For example,
scientists have been able to use social media like
Facebook and Twitter to track the impacts of
adverse weather, earthquakes or even impeding
flu epidemics (Jesudason 2016). Store purchases
may also give an early hint of such events. Given
what store chains can divine about individual
lifestyles, it remains to be seen whether their
techniques, such as collaborative filtering, can be
used to identify obscure risk factors for birth
defects (Jesudason 2016). Certainly, the current
generation of children are perhaps the first in
which a store card records most purchases to
which they have ever been exposed. At present,
this seems like an area ripe for exploration.
Another advance that one may see is the intro-
duction of “near patient interfaces” that allow the
parent to enter more of their information and that
of their child in cases of birth defects. Shifting
10
data ownership may be an effective way to
increase participation, particularly when
resources for dedicated data gatherers are scarce.
The question then arises whether nations, rich
and poor, are building healthcare teams equipped
to do such work. Medicine is still taught around
the great empires of biomedicine. To help the
shift from biological reductionism, these new
challenges will require a reintroduction of engi-
neering know-how into medicine: to improve
quantitative modeling, to manipulate complex
systems, and to program computers to explore
these data-rich opportunities (Jesudason 2016).
Looking at wider policy, birth registries are too
important to neglect. Births show demographers
how the world’s population will look in the future.
By this standard, birth defects monitoring pro-
vides an early warning system for humanity as a
whole. Therefore, if climate change exerts subtle
effects, it may be that these will be detected first
in the birth prevalences of key defects (Van
Zutphen et al. 2012). Similarly, increased use of
genetically modified crops seems likely to fuel
public demand for good data on birth defects to
ensure that risk to humans is minimal (Maghari
and Ardekani 2011). Finally, areas of Iraq exposed
to depleted uranium shells report increased birth
defect rates (Hindin et al. 2005). Disturbingly, the
WHO is alleged to have been complicit in efforts
to suppress this “bad news” (Ahmed 2013). Birth
defects epidemiology is often difficult in peace-
time, so it is understandable that there would be
controversy about conflict-­related birth defects.
However, healthcare professionals have a respon-
sibility to speak truth to power even on these
uncomfortable matters.
References
Ahmed N (2013) How the World Health Organisation
covered up Iraq’s nuclear nightmare. The Guardian;
[internet]. Available from: https://www.theguard-
ian.com/environment/earth-­i nsight/2013/oct/13/
world-­health-­organisation-­iraq-­war-­depleted-­uranium
Arth A, Kancherla V, Pachón H, Zimmerman S, Johnson
Q, Oakley GP Jr (2016) A 2015 global update on folic
acid-preventable spina bifida and anencephaly. Birth
Defects Res A Clin Mol Teratol 106:520–529
Banatvala JE, Brown DWG (2004) Rubella. Lancet
363:1127–1137
F. Friedmacher and E. C. Jesudason
Bell R, Glinianaia SV, Tennant PW, Bilous RW, Rankin J
(2012) Peri-conception hyperglycaemia and nephrop-
athy are associated with risk of congenital anomaly in
women with pre-existing diabetes: a population-based
cohort study. Diabetologia 55:936–947
Blotière PO, Raguideau F, Weill A, Elefant E, Perthus I,
Goulet V et al (2019) Risks of 23 specific malforma-
tions associated with prenatal exposure to 10 antiepi-
leptic drugs. Neurology 93:e167–e180
Botting J (2002) The History of Thalidomide. Drug News
Perspect 15:604–611
Boyd PA, Armstrong B, Dolk H, Botting B, Pattenden S,
Abramsky L et al (2005) Congenital anomaly surveil-
lance in England  – ascertainment deficiencies in the
national system. BMJ 330:27
Brody LC, Conley M, Cox C, Kirke PN, McKeever MP,
Mills JL et  al (2002) A polymorphism, R653Q, in
the trifunctional enzyme methylenetetrahydrofolate
dehydrogenase/methenyltetrahydrofolate cyclohydro-
lase/formyltetrahydrofolate synthetase is a maternal
genetic risk factor for neural tube defects: report of
the Birth Defects Research Group. Am J Hum Genet
71:1207–1215
Carmona RH (2005) The global challenges of birth
defects and disabilities. Lancet 366:1142–1144
Colborn T, vom Saal FS, Soto AM (1993)
Developmental effects of endocrine-disrupting
chemicals in wildlife and humans. Environ Health
Perspect 101:378–384
Cooper LZ (1985) The history and medical consequences
of rubella. Rev Infect Dis 7:S2–S10
Dang D, Wang L, Zhang C, Li Z, Wu H (2020) Potential
effects of SARS-CoV-2 infection during pregnancy on
fetuses and newborns are worthy of attention. J Obstet
Gynaecol Res 46:1951–1957
DiMasi JA, Grabowski HG, Vernon J (2004) R&D costs
and returns by therapeutic category. Ther Innov Regul
Sci 38:211–223
Donnai D, Read AP (2003) How clinicians add to knowl-
edge of development. Lancet 362:477–484
Duke CW, Correa A, Romitti PA, Martin J, Kirby RS
(2009) Challenges and priorities for surveillance of
stillbirths: a report on two workshops. Public Health
Rep 124:652–659
Ernhart CB, Sokol RJ, Martier S, Moron P, Nadler D, Ager
JW et al (1987) Alcohol teratogenicity in the human: a
detailed assessment of specificity, critical period, and
threshold. Am J Obstet Gynecol 156:33–39
Feldkamp ML, Carey JC, Sadler TW (2007) Development
of gastroschisis: review of hypotheses, a novel hypoth-
esis, and implications for research. Am J Med Genet A
143A:639–652
Godwin KA, Sibbald B, Bedard T, Kuzeljevic B, Lowry
RB, Arbour L (2008) Changes in frequencies of select
congenital anomalies since the onset of folic acid for-
tification in a Canadian birth defect registry. Can J
Public Health 99:271–275
Hackshaw A, Rodeck C, Boniface S (2011) Maternal
smoking in pregnancy and birth defects: a systematic
review based on 173 687 malformed cases and 11.7
million controls. Hum Reprod Update 17:589–604
1  The Epidemiology of Birth Defects
Hindin R, Brugge D, Panikkar B (2005) Teratogenicity
of depleted uranium aerosols: a review from an epide-
miological perspective. Environ Health 4:17
Hook EB, Czeizel AE (1997) Can terathanasia explain the
protective effect of folic-acid supplementation on birth
defects? Lancet 350:513–515
Jancelewicz T, Harrison MR (2009) A history of fetal sur-
gery. Clin Perinatol 36:227
Jentink J, Loane MA, Dolk H, Barisic I, Garne E, Morris
JK et  al (2010) Valproic acid monotherapy in preg-
nancy and major congenital malformations. N Engl J
Med 362:2185–2193
Jesudason EC (2016) The epidemiology of birth defects.
In: Puri P (ed) Pediatric surgery. Springer, Berlin,
Heidelberg
Johansson KA, Dursun U, Jordan N, Gu G, Beermann
F, Gradwohl G et al (2007) Temporal control of neu-
rogenin3 activity in pancreas progenitors reveals
competence windows for the generation of different
endocrine cell types. Dev Cell 12:457–465
Keijzer R, Liu J, Deimling J, Tibboel D, Post M (2000)
Dual-hit hypothesis explains pulmonary hypoplasia in
the nitrofen model of congenital diaphragmatic hernia.
Am J Pathol 156:1299–1306
Khoshnood B, Loane M, de Walle H, Arriola L, Addor
MC, Barisic I et al (2015) Long term trends in preva-
lence of neural tube defects in Europe: population
based study. BMJ 351:h5949
Khoury MJ (1989) Epidemiology of birth defects.
Epidemiol Rev 11:244–248
Kim J, Wu HH, Lander AD, Lyons KM, Matzuk MM,
Calof AL (2005) GDF11 controls the timing of pro-
genitor cell competence in developing retina. Science
308:1927–1930
Leeder JS, Mitchell AA (2007) Application of pharma-
cogenomics strategies to the study of drug-induced
birth defects. Clin Pharmacol Ther 81:595–599
Li S, Moore CA, Li Z, Berry RJ, Gindler J, Hong SX et al
(2003) A population-based birth defects surveillance
system in the People’s Republic of China. Paediatr
Perinat Epidemiol 17:287–293
Lin AE, Forrester MB, Cunniff C, Higgins CA, Anderka
M (2006) Clinician reviewers in birth defects surveil-
lance programs: survey of the National Birth Defects
Prevention Network. Birth Defects Res A Clin Mol
Teratol 76:781–786
Loane M, Morris JK, Addor MC, Arriola L, Budd J, Doray
B et al (2013) Twenty-year trends in the prevalence of
Down syndrome and other trisomies in Europe: impact
of maternal age and prenatal screening. Eur J Hum
Genet 21:27–33
Maghari BM, Ardekani AM (2011) Genetically modified
foods and social concerns. Avicenna J Med Biotechnol
3:109–117
Mah VK, Zamakhshary M, Mah DY, Cameron B, Bass
J, Bohn D et al (2009) Absolute vs relative improve-
ments in congenital diaphragmatic hernia survival:
what happened to “hidden mortality”. J Pediatr Surg
44:877–882
11
McLachlan JA (2001) Environmental signaling: what
embryos and evolution teach us about endocrine dis-
rupting chemicals. Endocr Rev 22:319–341
Meio IB, Siviero I, Ferrante SMR, Carvalho JJ (2008)
Morphologic study of embryonic development of rat
duodenum through a computerized three-dimensional
reconstruction: critical analysis of solid core theory.
Pediatr Surg Int 24:561–565
Morris JK, Springett AL, Greenlees R, Loane M, Addor
MC, Arriola L et  al (2018) Trends in congenital
anomalies in Europe from 1980 to 2012. PLoS One
13:e0194986
Nichol PF, Reeder A, Botham R (2011) Humans, mice,
and mechanisms of intestinal atresias: a window
into understanding early intestinal development. J
Gastrointest Surg 15:694–700
Piedrahita JA (2011) The role of imprinted genes in fetal
growth abnormalities. Birth Defects Res A Clin Mol
Teratol 91:682–692
Pierson TC, Diamond MS (2018) The emergence of
Zika virus and its new clinical syndromes. Nature
560:573–581
Pitkin RM (2007) Folate and neural tube defects. Am J
Clin Nutr 85:285S–288S
Rothstein DH, Dasgupta R (2016) Delivery of Surgical
Care Committee of the American Academy of
Pediatrics Section on Surgery. Transition of care from
pediatric to adult surgery. Pediatrics 138:e20161303
Shorter NA, Georges A, Perenyi A, Garrow E (2006)
A proposed classification system for familial intes-
tinal atresia and its relevance to the understanding
of the etiology of jejunoileal atresia. J Pediatr Surg
41:1822–1825
Stiefel D, Meuli M (2007) Scanning electron microscopy
of fetal murine myelomeningocele reveals growth and
development of the spinal cord in early gestation and
neural tissue destruction around birth. J Pediatr Surg
42:1561–1565
Stothard KJ, Tennant PW, Bell R, Rankin J (2009)
Maternal overweight and obesity and the risk of
congenital anomalies: a systematic review and meta-­
analysis. JAMA 301:636–650
Tararbit K, Lelong N, Thieulin AC, Houyel L, Bonnet D,
Goffinet F et al (2013) The risk for four specific con-
genital heart defects associated with assisted repro-
ductive techniques: a population-based evaluation.
Hum Reprod 28:367–374
Van Zutphen AR, Lin S, Fletcher BA, Hwang SA (2012)
A population-based case-control study of extreme
summer temperature and birth defects. Environ Health
Perspect 120:1443–1449
Whitaker SY, Tran HT, Portier CJ (2003) Development
of a biologically-based controlled growth and differ-
entiation model for developmental toxicology. J Math
Biol 46:1–16
Yang Q, Wen SW, Leader A, Chen XK, Lipson J, Walker
M (2007) Paternal age and birth defects: How strong is
the association? Hum Reprod 22:696–701
 Fetal Counselling for Surgical
Congenital Malformations
Kokila Lakhoo and Rebecca Black
2.1 Introduction
Paediatric surgeons are often called to counsel
parents once a surgical abnormality is diagnosed
on a prenatal scan. The referral base for a paedi-
atric surgeon now includes the perinatal period.
Favourable impact of prenatal counselling has
been confirmed to influence the site of delivery in
37% of cases, change the mode of delivery in
6.8%, reverse the decision to terminate a preg-
nancy in 3.6% and influence the early delivery of
babies in 4.5%.
Counselling parents about prenatally sus-
pected surgically correctable anomalies should
not be solely performed by obstetricians or pae-
diatricians. Similarly, the paediatric surgeon per-
forming these prenatal consultations must be
aware of differences between the prenatal and
postnatal natural history of the anomaly. There is
often a lack of understanding regarding the natu-
ral history and prognosis of a condition present-
ing in the newborn and the same condition
diagnosed prenatally.
K. Lakhoo (*)
Children’s Hospital Oxford, Oxford University
Hospitals, University of Oxford, Oxford, UK
e-mail: kokila.lakhoo@paediatrics.ox.ac.uk
R. Black
Oxford University Hospitals NHS Foundation Trust,
Oxford, UK
e-mail: Rebecca.black@ouh.nhs.uk
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_2
2
The diagnosis and management of complex
fetal anomalies require a team effort by obstetri-
cians, neo-natologists, geneticists, paediatricians
and paediatric surgeons to deal with all the mater-
nal and fetal complexities of a diagnosis of a
structural defect. This team should be able to pro-
vide information to prospective parents on fetal
outcomes, possible interventions, appropriate
setting, time and route of delivery and expected
postnatal outcomes. The role of the surgical con-
sultant in this team is to present information
regarding the prenatal and postnatal natural his-
tory of an anomaly, its surgical management and
the long-term outcome (Lakhoo 2007).
2.2 Historical Overview
Prenatal diagnosis has remarkably improved our
understanding of surgically correctable congeni-
tal malformations. It has allowed us to influence
the delivery of the baby, offer prenatal surgical
management and discuss the options for termina-
tion of pregnancy in the case of seriously handi-
capping or lethal conditions. Antenatal diagnosis
has also defined an in utero mortality for some
lesions, such as diaphragmatic hernia and sacro-­
coccygeal teratoma, so that true outcomes can be
measured. Prenatal ultrasound scanning has
improved since its first use 50  years ago, thus
providing better screening programmes and more
accurate assessment of fetal anomalies. Screening
13
14
for Down’s syndrome may now be offered in the
first trimester e.g. the combined test (using a
combination of nuchal translucency measure-
ment and maternal blood markers) or second tri-
mester tests e.g. quadruple blood test. Better
ultrasound resolution has led to the recognition
of ultrasound soft markers that have increased the
detection rate of fetal anomalies, but at the
expense of higher false positive rates. Routine
ultrasound screening identifies anomalies and
places these pregnancies into a higher risk
­category. Such pregnancies may be referred to
Fetal Medicine Units for further scanning and
other investigations.
Parents may be offered further invasive diag-
nostic investigations, such as amniocentesis or
chorionic villous sampling. Some structural
abnormalities which are difficult to define on
ultrasound, such as hindbrain lesions or in the
presence of oligohydramnios, may be better
imaged with magnetic resonance imaging. With
the increasing range of options and sophistication
of diagnostic methods, parents today are faced
with more information, choice and decisions than
ever before, which can create as well as help to
solve dilemmas. The different tests and screening
procedures commonly in use are outlined below
under diagnosis.
2.3 Incidence
Congenital malformations account for one of the
major causes of perinatal mortality and morbid-
ity. Single major birth defects affect 3% of new-
borns and multiple defects affect 0.7% of babies.
The prenatal hidden mortality is higher since the
majority abort spontaneously. Despite improve-
ments in perinatal care, serious birth defects still
account for 20% of all deaths in the newborn
period and an even greater percentage of serious
morbidity later in infancy and childhood. The
major causes of congenital malformation are
chromosomal abnormalities, mutant genes, mul-
tifactorial disorders and teratogenic agents
(Lakhoo 2007).
K. Lakhoo and R. Black
2.4 Prenatal Diagnosis
2.4.1 Screening for Fetal Anomalies
The NHS fetal anomaly screening programme
(FASP) offers screening to all pregnant women in
England (NHS Fetal Anomaly Screening
Programme Handbook 2018). The first scan is
performed at 10 to 14 weeks of gestation. It can:
• Confirm viability.
• Accurately date the pregnancy.
• Diagnose multiple pregnancy and
chorionicity.
• Detect major structural anomalies, such as
anencephaly.
The combined test can be used to assess the
chance of the baby being born with Down’s syn-
drome (trisomy 21), Edward’s syndrome (tri-
somy 18) or Patau’s syndrome (trisomy 13). It
combines maternal age, gestational age, ultra-
sound measurement of the nuchal translucency
(Fig. 2.1) at between 11 and 14 weeks’ gestation
with two biochemical markers—PAPPA and free
beta hCG—to calculate the risk of the pregnancy
being affected by T21, 13 or 18.
If the nuchal translucency cannot be mea-
sured, a quadruple test can be offered. This mea-
sures four biochemical markers—AFP, hCG, uE3
and inhibin-A.  This test can be performed
between 14 + 2 weeks and 20 + 0 weeks.
Fig. 2.1  Nuchal translucancy scan
2  Fetal Counselling for Surgical Congenital Malformations
The combined test has a detection rate of 80%
for a screen positive rate of 2.5%. The quadruple
test has a detection rate of 80% for a screen posi-
tive rate of up to 3.5%.
A second scan (often referred to as ‘the anom-
aly scan’ is offered at 18 + 0 to 20 + 6 weeks of
pregnancy. This scan is designed to identify
anomalies which indicate:
• Conditions that may benefit from treatment
before or after birth.
• Conditions whose outcome may be improved
by planning an appropriate place, mode and
timing of birth, along with optimal postnatal
management.
• That the baby may not survive the neonatal
period.
As a minimum, the conditions screened for at
the anomaly scan, along with their detection
rates, are:
Anencephaly 98% detection rate
Open spina bifida 90%
Cleft lip 75%
Diaphragmatic hernia 60%
Gastroschisis 98%
Exomphalos 80%
Cardiac anomalies 50%
Bilateral renal agenesis 84%
Lethal skeletal dysplasia 60%
The quality of the images obtained is depen-
dent on many factors, including the skill of the
operator, maternal habitus and the position of the
fetus. Scan technology is improving images,
including 3-D and 4-D options.
Scanning for fetal cardiac defects remains
challenging. There is an association between an
increased nuchal translucency measurement and
the risk of a cardiac defect, or a wide range of
other syndromes, even in the context of a normal
karyotype. For this reason, those fetuses with a
raised NT and normal karyotype are offered fetal
echocardiography.
Some anomalies may not be visible at the time
of the anomaly scan, but only present later in the
third trimester. Examples include duodenal atre-
sia and other forms of bowel obstruction, and
15
non-lethal skeletal dysplasias; in achondroplasia,
the femur is of normal length at the time of the
anomaly scan.
2.4.2 Invasive Diagnostic Tests
Amniocentesis and chorionic villous sampling
(CVS) are the two most commonly performed
invasive diagnostic tests (Fetal Medicine
Foundation 2019). Since the introduction of the
national screening programme for common triso-
mies, the second trimester scan is interpreted dif-
ferently. Certain ultrasound findings, previously
referred to as ‘soft markers’ should not be used to
recalculate a risk for these trisomies. These find-
ings include choroid plexus cysts, a dilated cis-
terna magna, echogenic foci in the heart and a
2-vessel umbilical cord. However, other findings,
termed ‘markers’ are of significance and should
be referred for further assessment in a Fetal
Medicine Unit. These include an increased
nuchal fold, cerebral ventriculomegaly, echo-
genic bowel, renal pelvic dilatation and a small
fetus (<fifth centile).
2.4.2.1 Amniocentesis
Amniocentesis is commonly used for detecting
chromosomal abnormalities and less often for
molecular studies, metabolic studies and fetal
infection. It is performed after 15 weeks of gesta-
tion and carries a low risk of fetal injury or loss
(up to 1%). Full karyotype and microarray analy-
sis takes approximately 2  weeks but newer
RAPID techniques, using FISH (fluorescent in
situ hybridisation) or PCR (polymerase chain
reaction), can give limited (usually for trisomies
21, 18, 13) results within 2–3 days.
2.4.2.2 Chorionic Villous Sampling (CVS)
CVS is the most reliable method for first trimes-
ter diagnosis and may be performed after
11  weeks of gestation. The test involves
ultrasound-­guided biopsy of the chorionic villi.
The added risk for fetal loss is up to 1%. The
samples obtained may be subjected to a variety of
tests including full karyotype, rapid karyotyping
(FISH—PCR), microarray, enzyme analysis or
16 K. Lakhoo and R. Black

molecular studies. Approximate timing of chro- others have been abandoned and more are still
mosomal results is 1–2  weeks for karyotyping being evaluated. Some remain controversial due
and 2–3 days for FISH and PCR. the potential long-term effects on both mother
and baby (Harrison 2006).
2.4.2.3 Prenatal Maternal Serum
Screening 2.4.2.6 Genetic Diagnoses
It is possible to detect placental DNA in the Antenatal detection of genetic abnormalities is
mother’s bloodstream. This is termed total cell increasing, especially in high-risk pregnancies.
free DNA (cfDNA) and in most cases will be the Previously undiagnosed conditions such as cystic
same as fetal DNA. This means that a maternal fibrosis, Beckwith-Wiedemann syndrome,
blood test can be used to detect conditions such Hirshsprung’s disease and sickle cell disease may
as Trisomy 21. Many commercial options are be detected prenatally following invasive testing
currently available, as a screening test; invasive and genetic counselling and assessment offered
testing is still advised following a positive result. early in pregnancy.
The use of maternal blood to discover the fetal
blood group for mothers with antibodies such as 2.4.2.7 Future Developments
anti-D, and to determine fetal sex for mothers The aim of prenatal diagnosis and testing is to
who are carriers of sex-linked conditions such as ensure 100% accuracy without fetal loss or injury
haemophilia, is now in routine clinical practice. and no maternal risk. National strategies to
Invasive testing may still be advised following improve Down’s screening using ultrasound, bio-
the finding of an abnormality on ultrasound scan chemical combination tests and non-invasive pre-
because of the additional information that micro- natal testing (NIPT) are now in place in the UK
array testing may bring. ISUOG guidelines sug- (Fetal anomalies: screening, conditions, diagno-
gest that microarrays could detect clinically sis, treatment in the UK, 2019).
relevant aberrations in 6% of fetuses with normal Expansion of the use of NIPT is ongoing. Its
karyotype and structural defects. use in Rhesus disease and sex-linked conditions
is established. There are now tests available for
2.4.2.4 Fetal Blood Sampling (FBS) single gene disorders, such as thanatophoric
Rapid karyotyping of CVS and amniotic fluid dwarfism and Apert’s syndrome.
samples FISH and PCR have replaced fetal blood The field of genetics is also developing rap-
sampling for many conditions. However, FBS is idly. WES (whole exome sequencing) and other
still occasionally indicated for required for the genetic tests are becoming a clinical reality.
diagnosis and treatment of haematological condi- Imaging technology is improving. The use of
tions, some viral infections and to investigate 3D and 4D imaging is helping with diagnostics
chromosomal mosaicism after amniocentesis. and in communication with parents and families.
When required, it is best performed by ultrasound-­ Automation and the use of artificial intelligence
guided needle sampling after 18 weeks of gesta- (AI) to develop new ways of scanning and screen-
tion. Fetal loss from this procedure is reported to ing is also an active area of research.
be 1–2%.

2.4.2.5 Fetal Surgery 2.5 Specific Surgical Conditions

There is a spectrum of interventions ranging from
ultrasound-guided needling procedures, such as
cyst aspiration, through fetoscopic minimally
invasive techniques (Flake 2017) to hysterotomy
and open fetal surgery. Some have been subject
to rigorous randomised controlled trials and are
now incorporated into routine clinical practice;
2.5.1 Congenital Diaphragmatic
Hernia (CDH)
CDH accounts for 1 in 4000 live birth and chal-
lenges the neonatologist and paediatric surgeons
in the management of this high-risk condition.
2  Fetal Counselling for Surgical Congenital Malformations
Lung hypoplasia and pulmonary hypertension
account for most deaths in isolated CDH new-
borns. If isolated, survival is <15% for severe dis-
ease, 50% for moderate disease and  >  90% for
mild disease. Associated anomalies signify a
worse prognosis with a survival rate of less than
10%; chromosomal defects mainly trisomy 13 or
18 or 12p (Pallister Killian) syndrome are found
in 20% of fetuses with CDH, genetic syndromes
(e.g. Fryn’s syndrome) in 10% and other
­associated anomalies (mainly cardiac and cranio-
facial) in 20% of cases.
In the UK, most CDHs are diagnosed at the
20-week anomaly scan with a detection rate of
around 60%, although as early as 11  weeks’
gestation has been reported. 80% of CDHs
involve the left side of the diaphragm, 15% are
right sided and 5% are retrosternal; left sided
hernias are more easily diagnosed because the
stomach can be seen in the chest on routine grey
scale ultrasound. Magnetic resonance imaging
(MRI) can be useful in accurately differentiat-
ing CDH from cystic lung lesions. The lung
head ratio (LHR) of the fetus can be calculated
and has been used to predict prognosis. An LHR
of <1, or an observed compared to expected
ratio (O/E) of <25% along with early detection,
liver in the chest and polyhydramnios are poor
predictors of outcome. An O/E ratio of 26–45%
suggests moderate disease and >45%, mild dis-
ease. Once a diagnosis of CDH is made, detailed
scanning, including echocardiography, along
with invasive testing for karyotye and microar-
ray is offered.
For those fetuses with an isolated CDH and
normal karyotype, FETO (fetoscopic endolumi-
nal tracheal occlusion) can be offered. This
involves insertion of an inflatable balloon into the
fetal trachea using a fetoscope. The theory is that
the balloon prevents the net flow of fluid from the
lungs which occurs in fetal life, thereby promot-
ing lung growth. The balloon remains in situ
from about 26 to 34 weeks’ gestation. An interna-
tional randomised trial is currently underway to
investigate the effectiveness of FETO in compari-
son to routine postnatal management; results are
awaited (https://www.totaltrial.eu/). Women with
17
a fetus with CDH should deliver in a specialist
centre. CDH is not an indication for caesarean
section.
Postnatal management is aimed at reducing
barotrauma to the hypoplastic lung by introduc-
ing high frequency oscillatory ventilation
(HFOV) or permissive hypercapnea, and treating
severe pulmonary hypertension with nitric oxide.
No clear benefits for CDH with ECMO (extracor-
poreal membrane oxygenation) have been con-
cluded in a 2002 Cochrane ECMO study. The
value of ECMO is still unclear to date (McHoney
and Hammond 2018).
Surgery for CDH is no longer an emergency
procedure. Delayed repair following stabilisation
is employed in most paediatric surgical centres.
Primary repair using the trans-abdominal route is
achieved in 60–70% of patients with the rest
requiring a prosthetic patch. Complications of
sepsis or reherniation with prosthetic patch
requiring revision are recorded in 50% of survi-
vors. Minimally invasive techniques have been
successful in repairing diaphragmatic defects in
‘stable’ infants. Long-term survivors of CDH are
reported to develop chronic respiratory insuffi-
ciency (48%), gastro-oesophageal reflux (89%)
and neurodevelopment delay (30%).
2.5.2 Cystic Lung Lesions
Congenital pulmonary airway malformations
(CPAM) occur in approximately 1 in 4000 births.
They can be macrocystic—made up of one or more
large (>2 cm) cysts (type 1), microcystic (type 3) or
mixed (type 2). The vast majority (>95%) are uni-
lateral, involving one lobe or segment of the lung.
Less common lung anomalies include broncho-
genic cysts, congenital lobar emphysema and bron-
chial atresia. CPAMs are usually isolated. The risk
of chromosomal or genetic disorders are not
increased, so invasive testing is not usually offered.
Fetal echocardiography is performed in cases
where there is mediastinal shift, as this can make
cardiac assessment more challenging.
Many CPAMs will run a benign course. Some,
however, can be associated with hydrops (with a
18
poor prognosis) or polyhydramnios. If these
complications do develop and there is a large
cyst, aspiration or shunting (insertion of a pig tail
catheter connecting the cyst with the amniotic
fluid through the fetal chest) can improve out-
come. In the case of type 3 CPAM, a course of
antenatal steroids given to the mother can some-
times improve prognosis. Open fetal surgery with
excision of the CPAM has been performed in a
few cases.
CPAMs, particularly microcystic (type 3)
ones, can be difficult to see on ultrasound in the
third trimester as the fetus grows and the sur-
rounding lung becomes more echogenic. In 80%,
however, the lesion is still present and needs
postnatal follow-up. A small CPAM with mini-
mal medistainal shift should not affect the tim-
ing, place or mode of birth. A large lesion, where
postnatal surgery is contemplated, should deliver
in a tertiary unit. Accurate antenatal multiprofes-
sional planning is essential. Postnatal manage-
Fig. 2.2  Prenatal scan and postnatal radiological image of CPAM
K. Lakhoo and R. Black
ment is dictated by clinical status at birth.
Symptomatic lesions require urgent radiological
evaluation with chest radiograph and ideally a
CT scan (Fig. 2.2) followed by surgical excision.
In asymptomatic cases, postnatal investigation
consists of chest CT scan within 1 month of birth,
even if regression or resolution is noted on prena-
tal scanning. Plain radiography should not be
relied on, because it will miss and underestimate
many lesions.
Surgical excision of postnatal asymptomatic
lesions remains controversial, with some centres
opting for conservative management. The
approach to treating this asymptomatic group has
evolved in some centres, whereby a CT scan is
performed within 1 month post birth, followed by
surgery before 6 months of age due to the inher-
ent risk of infection and malignant transforma-
tion (Annunziata et al. 2019). Small lesions less
than 1 cm may be managed expectantly, as these
may not represent CCAM but artifact or end on
2  Fetal Counselling for Surgical Congenital Malformations 19

view of a vessel. True resolution of these lesions mortality in isolated exomphalos; however, the
is exceptional. Successful outcome of greater majority of these children survive to live normal
than 90% have been reported for these surgically lives.
managed asymptomatic lung lesions.
2.5.3.2 Gastroschisis
Gastroschisis is an isolated lesion that usually
2.5.3 Abdominal Wall Defects occurs on the right side of the umbilical defect
with evisceration of the abdominal contents
Exomphalos and gastroschisis are both common directly into the amniotic cavity. The incidence is
but distinct abdominal wall defects with an increasing from 1.66 per 10,000 births to 4.6 per
unclear aetiology and a controversial prognosis. 10,000 births affecting mainly young mothers
Antenatal detection rates for both conditions are typically less than 20  years old. Associated
high. Most will be found by the time of the anom- anomalies are noted in only 5–24% of cases with
aly scan; an increasing number is being found in bowel atresia the most common co-existing
the first trimester. abnormality. The incidence of chromosomal and
genetic syndromes is not increased. On prenatal
2.5.3.1 Exomphalos scan, with a detection rate approaching 100%,
Exomphalos is characteristically a midline the bowel appears to be free floating, and the
defect, at the insertion point of the umbilical loops may appear to be thickened due to damage
cord, with a viable sac composed of amnion and by amniotic fluid exposure causing a “peel” for-
peritoneum containing herniated abdominal con- mation. Dilated loops of bowel (Fig. 2.3) may be
tents. Incidence is known to be 1  in 4000 live seen from obstruction secondary to protrusion
births. Associated major abnormalities that from a defect or atresia due to intestinal
include trisomy 13, 18 and 21, Beckwith-­ ischaemia.
Wiedemann syndrome (macroglossia, gigan- Predicting outcome in fetuses with gastroschi-
tism, exomphalos), Pentology of Cantrell sis based on prenatal ultrasound finding remains
(sternal, pericar-dial, cardiac, abdominal wall a challenge. There is some evidence that internal
and diaphragmatic defect), cardiac, gastrointes- bowel dilatation may be predictive; however,
tinal and renal abnormalities are noted in thickened matted bowel and Doppler measure-
60–70% of cases; thus, karyotyping, in addition ments of the superior mesenteric artery are not
to detailed sonographic review and fetal echo- accurate predictors of outcome. Fetal growth
cardiogram, is essential for complete prenatal restriction is common (30–60% of cases) and
screening. Fetal intervention is unlikely in this more difficult to monitor because the abdominal
condition. If termination is not considered, nor-
mal vaginal delivery at a centre with neonatal
surgical expertise is recommended and delivery
by caesarean section only is reserved for large
exomphalos with exteriorised liver to prevent
damage.
Surgical repair includes primary closure or a
staged repair with a silo for giant defects.
Occasionally, in vulnerable infants with severe
pulmonary hypoplasia or complex cardiac abnor-
malities the exomphalos may be left intact and
allowed to slowly granulate and epithelialise by
application of antiseptic solution. Postnatal mor-
bidity occurs in 5–10% of cases. Malrotation and Fig. 2.3  Prenatal ultrasound of dilated bowel in gastros-
adhesive bowel obstruction does contribute to chisis
20
circumference is more difficult to measure and
interpret. Some centres will opt for elective cae-
sarean section for all, but most will offer induc-
tion of labour by around 37 weeks of gestation.
Delivery needs to be at a centre with paediatric
surgical facilities.
Various methods of postnatal surgical repair
include the traditional primary closure, reduction
of bowel without anaesthesia, reduction by pre-
formed silo, or by means of a traditional silo.
Co-existing intestinal atresia could be repaired
by primary anastamosis or staged with stoma for-
mation. Variation in achieving full enteral feed-
ing due to prolonged gut dysmotility is expected
in all cases.
The long-term outcome in gastroschisis is
dependent on the condition of the bowel. In
uncomplicated cases, the outcome is excellent in
more than 90% of cases. The mortality of live
born infants is 5%, with further 5% suffering
short bowel syndrome and 10% requiring surgery
for adhesive bowel obstruction. Late third trimes-
ter fetal loss should always be mentioned during
fetal counselling (Gamba and Midrio 2014).
2.5.4 Tracheo-Oesophageal Fistula
(TOF) and Oesophageal
Atresia (OA)
Repair of TOF/OA is a condition that measures
the skill of paediatric surgeons from trainees to
independent surgeons. The incidence is estimated
at 1 in 3000 births.
Prenatally, the condition may be suspected
from maternal polyhydramnios and absence of a
fetal stomach bubble at any time from the
20-week anomaly scan. However, if there is an
associated tracheoesophageal fistula, the stomach
may appear normal on ultrasound scan. It is
therefore estimated that oesophageal atresia is
suspected prenatally in only about 40% of cases
(Bradshaw et al. 2016).
Additional diagnostic clues are provided by
associated anomalies, such as trisomy (13, 18,
21), VACTERL sequence (vertebral, anorectal,
cardiac, tracheo-oesophageal, renal, limbs) and
CHARGE association (coloboma, heart defects,
K. Lakhoo and R. Black
atresia choanae, retarded development, genital
hypoplasia, ear abnormality). Associated anoma-
lies, mainly cardiac, are present in more than 50%
of cases and worsen the prognosis; fetal echocar-
diography and invasive testing are therefore usu-
ally offered. Duodenal atresia may co-­exist with
TOF/OA. Amnioreduction (draining of the amni-
otic fluid) can be offered, particularly for symp-
tomatic relief for the mother, but carries a risk of
preterm birth and is only a temporary measure as
the fluid will reaccumulate. The risk of recurrence
in subsequent pregnancies for isolated TOF/OA is
less than 1%. Delivery is advised at a specialised
centre with neonatal surgical input.
Postnatal surgical management is dependent
on the size and condition of the baby, length of
the oesophageal gap and associated anomalies.
Primary repair of the oesophagus is the treatment
of choice; however, if not achieved, staged repair
with upper oesophageal pouch care and gastros-
tomy or organ replacement with stomach or large
bowel are other options. Associated anomalies
require evaluation and treatment. Advanced pae-
diatric endosurgical centres may offer minimally
invasive thoracoscopic approach to the repair of
TOF.  Early outcome of a high leak rate and
oesophageal stricture requiring dilatation in 50%
of cases are expected where the anastamosis of
the oesophagus is created under tension.
Improved perinatal management and inherent
structural and functional defects in the trachea
and oesophagus indicate long-term outcome. In
early life, growth of the child is reported to be
below the 25th centile in 50% of cases, respira-
tory symptoms in two-thirds of TOF/OA and
gastro-oesophageal reflux recorded in 50% of
patients. Quality of life is better in the isolated
group with successful primary repair compared
to those with associated anomalies and delayed
repair.
2.5.5 Gastrointestinal Lesions
The presence of dilated loops of bowel
(>17 mm in length and 7 mm in diameter) on
prenatal ultrasound scan is indicative of bowel
obstruction.
2  Fetal Counselling for Surgical Congenital Malformations
Duodenal atresia has a characteristic ‘double
bubble’ appearance on prenatal scan, resulting
from the simultaneous dilatation of the stomach
and proximal duodenum. This characteristic sign
is, however, usually only present after 24 weeks’
gestation, so does not get picked up at the time of
the routine anomaly scan. Associated anomalies
are present in approximately 50% of cases, most
notably trisomy 21  in 30% of cases, cardiac
anomalies in 20% and the presence of VACTERL
association (vertebral, anorectal, cardiac, tracheo-­
oesophageal, renal and limbs).
The incidence of duodenal atresia is 1 in 5000
live births. The postnatal survival rate is >95% with
associated anomalies, low birth weight and prema-
turity contributing to the <5% mortality. Temporary
delay in enteral feeding occurs due to the dysmotil-
ity in the dilated stomach and duodenum.
Many bowel abnormalities may be noted on
prenatal scanning (dilated bowel, ascites, cystic
masses, hyperparistalsis, poyhydramnios and
echogenic bowel); however, none is absolutely
predictive of postnatal outcome. Patients with
obstruction frequently have findings (especially in
the third trimester) of bowel dilatation, polyhy-
dramnios and hyperparistalsis, but ultrasound is
much less sensitive in diagnosing anomalies in the
large bowel than those in the small bowel. Since
the large bowel is mostly a reservoir, with no phys-
iologic function in utero, defects in this region
such as anorectal malformations or Hirschsprung’s
disease are very difficult to detect. Bowel dilata-
tion and echogenic bowel may be associated with
cystic fibrosis; therefore, all such fetuses should
undergo postnatal evaluation for this disease.
Prenatally diagnosed small bowel atresia does not
select for a group with a worse prognosis and sur-
vival rates are 95–100% (Lau et al. 2017).
2.5.6 Sacrococcygeal Teratoma
Sacrococcygeal teratoma (SCT) is the common-
est neonatal teratoma with an incidence of around
1 in 20,000 births. Four types have been defined:
Type 1 external tumour with a small presacral
component.
21
Type 2 external tumours with a large presacral
component.
Type 3 predominantly presacral with a small
external component.
Type 4 entirely presacral.
The latter carry the worst prognosis due to
delay in diagnosis and malignant presentation.
Overall perinatal mortality is around 50%.
Doppler ultrasound is a useful diagnostic tool to
assess tumour vascularity; fetal MRI can provide
better definition of the intrapelvic component.
Most teratomas are extremely vascular and the
fetus may develop high cardiac output failure,
polyhydramnios, anaemia and ultimately hydrops
with a mortality of almost 100%. Fetal blood
transfusions and amniodrainage may be indi-
cated. Minimally invasive techniques including
fetoscopic or ultrasound-guided laser coagula-
tion of blood vessels within the tumour have been
tried with limited benefits; the rate of in utero
demise or preterm birth post procedure is high
(Alalfy et al. 2019).
Caesarean section may be offered to patients
with large tumours to avoid the risk of bleeding
during delivery. Postnatal outcomes following
surgery in type 1 and 2 lesions are favourable;
however, type 3 and 4 tumours may present with
urological problems and less favourable out-
comes. Long-term follow-up with alpha fetopro-
tein and serial pelvic ultrasounds are mandatory
to exclude recurrence of the disease.
2.5.7 Renal Anomalies
Urogenital abnormalities are among the com-
monest disorders seen in the perinatal period and
account for almost 20% of all prenatally diag-
nosed anomalies. The routine use of antenatal
ultrasound scans has resulted in the early detec-
tion of these conditions, and in selected cases has
led to the development of management strategies
including fetal intervention aimed at preservation
of renal function. Two major issues are the indi-
cations for intervention in bladder outlet obstruc-
tion and early pyeloplasty in infancy in cases
with hydronephrosis.
22
Prenatal evaluation of a dilated urinary tract is
based on serial ultrasound scans as well as mea-
surement of urinary electrolytes. Ultrasonography
provides measurements of the renal pelvis,
assessment of the renal parenchyma as well as
the detection of cysts in the cortex. In severe
­disease, lack of amniotic fluid may make ultra-
sound assessment of the renal tract difficult and
MRI may be helpful. Oligohydramnios is indica-
tive of poor renal function and poor prognosis
owing to the associated pulmonary hypoplasia.
Urogenital anomalies co-exist with many other
congenital abnormalities and amniocentesis
should be offered in appropriate cases. It is esti-
mated that 3% of infants will have an abnormal-
ity of the urogenital system and half of these will
require some form of surgical intervention (Yulia
and Winyard 2018).
2.5.7.1 Upper Urinary Tract
Obstruction
Antenatal hydronephrosis accounts for 0.6–0.65%
pregnancies. The most common cause of prenatal
hydronephrosis is pelvi-ureteric junction obstruc-
tion (PUJ), others being transient hydronephrosis,
physiological hydronephrosis, multicystic kidney,
posterior urethral valves, ureterocele, ectopic ure-
ter, etc. The prognosis of antenatally diagnosed
hydronephrosis in unilateral disease and with a
renal pelvic diameter of <10  mm is excellent.
Spontaneous resolution is noted in 20% of patients
at birth and 80% at 3 years of age. Only around
20% of prenatally diagnosed hydronephrosis need
surgical intervention.
Postnatal management of hydronephrosis
requires ultrasound at birth and at 1 month of age,
and further evaluation with radiology and scin-
tigraphy if an abnormality is suspected. The
non-­operative treatment of antenatally detected
hydro-nephrosis has been carefully monitored over
a 17-year period, and from an analysis of six patient
series the conclusion is that this approach is safe.
2.5.7.2 Lower Urinary Tract
Obstruction
Posterior urethral valves (PUV) are the most
common cause for lower urinary tract obstruction
in boys with an incidence of 1 in 1500 live male
K. Lakhoo and R. Black
births. The diagnosis of PUV is suspected with a
prenatal ultrasound finding of lower urinary tract
obstruction (LUTO). With LUTO, there is a
thick-walled bladder with evidence of urethral
obstruction (a ‘keyhole’ sign). Depending on the
degree of obstruction, there may be back pressure
causing dilatation of the ureters and renal pel-
vices and damage to the renal cortices. The blad-
der may rupture, causing urinary ascites. Poor
prognostic features of PUVs include antenatal
detection, damage to the renal cortex and a
reduced liquor volume.
It is technically straightforward to insert a
shunt between the fetal bladder and amniotic
fluid in an attempt to relieve pressure on the uri-
nary system and protect renal function. However,
a randomised trial of this procedure versus stan-
dard postnatal management struggled to recruit;
those babies who were involved fared badly in
both arms of the trial and there were high rates of
renal failure at 2 years of age in both groups.
Postnatal management includes ultrasound
confirmation of the diagnosis, bladder drainage
via a suprapubic or urethral route and contrast
imaging of the urethra. Primary PUV ablation,
vesicostomy or ureterostomy are postnatal surgi-
cal options. The overall outcome from this dis-
ease is unfavourable.
2.6 Conclusion
The boundaries of paediatric surgical practice
have been extended by prenatal diagnosis. The
care of patients with surgically correctable
defects can now be planned prenatally with the
collaborative effort of obstetricians, specialist
midwives, geneticists, neonatologists and paedi-
atric surgeons. The understanding of the specific
surgical condition’s prenatal natural history, the
limitations of prenatal diagnosis, the detection of
associated anomalies, the risks and indications of
fetal intervention programmes and postnatal out-
comes are essential to prenatal counselling.
Prenatal counselling is an essential component of
paediatric surgical practice and should be
included in the training programme for future
paediatric surgeons.
2  Fetal Counselling for Surgical Congenital Malformations 23

References Gamba P, Midrio P (2014) Abdominal wall defects:

Prenatal diagnosis, newborn management, and long-­
term outcomes. Semin Pediatr Surg 23(5):283–290
Alalfy M, Elebrashy A, Azmy O et  al (2019) Review
Harrison MR (2006) The fetus as a patient. In: Grosfeld
article: prenatal diagnosis and management of sacro-
JL, O’Neill JA, Fonkalsrud EW, Coran AG (eds)
coccygeal teratoma, a review of literature. Obstet
Pediatric surgery. Mosby, Philadelphia, pp 77–88
Gynecol Int J 10(1):47–49. https://doi.org/10.15406/
Lakhoo K (2007) Fetal counselling for congenital malfor-
ogij.2019.10.00412
mations. PSI 23:509–519
Annunziata F, Bush A, Borgia F, Raimondi F, Montella S,
Lau PE, Cruz S, Cassady CI et al (2017) Prenatal diagno-
Poeta M, Borrelli M, Santamaria F (2019) Congenital
sis and outcome of fetal gastrointestinal obstruction. J
lung malformations: unresolved issues and unan-
Pediatr Surg 52(5):722–725. https://doi.org/10.1016/j.
swered questions. Front Pediatr 7:239. https://doi.
jpedsurg.2017.01.028
org/10.3389/fped.2019.00239
McHoney M, Hammond P (2018) Role of ECMO in con-
Bradshaw CJ, Thakkar H, Knutzen L, Lakhoo K et  al
genital diaphragmatic hernia Archives of Disease in
(2016) Accuracy of prenatal detection of tracheo-
Childhood  – Fetal and Neonatal Edition 103:F178–
esophageal fistula and oesophageal atresia. J Pediatr
F181. (https://www.totaltrial.eu/)
Surg 51(8):1268–1272. https://doi.org/10.1016/j.
NHS Fetal Anomaly Screening Programme Handbook
jpedsurg.2016.02.001
(2018). https://assets.publishing.service.gov.uk/gov-
Fetal anomalies: screening, conditions, diagnosis,
ernment/uploads/system/uploads/attachment_data/
treatment (2019). https://www.gov.uk/government/
file/749742/NHS_fetal_anomaly_screening_pro-
collections/fetal-­a nomalies-­s creening-­c onditions-­
gramme_handbook_FINAL1.2_18.10.18.pdf
diagnosis-­treatment
Yulia A, Winyard P (2018) Management of antenatally
Fetal Medicine Foundation (2019). https://fetalmedi-
detected kidney malformations. Early Hum Dev 126:38–
cine.org
46. https://doi.org/10.1016/j.earlhumdev.2018.08.017
Flake AW (2017) Fetal intervention. Semin Fetal Neonatal
Med 22(6), December 2017
 Transport of the Surgical Neonate
Udo Rolle and Prem Puri
3.1 Introduction
The successful outcome of surgery on newborns
with congenital anomalies or severe surgical dis-
ease depends not only on the skills of the paedi-
atric surgeon but also on that of a large team
consisting of a paediatrician, paediatric anaes-
thetist, paediatric radiologist, paediatric patholo-
gist, biochemist, nurses and others who are
necessary to satisfactorily deal with the newborn
infant undergoing surgery (Puri and Doodnath
2011). Advances in neonatal intensive care dic-
tate that effective and efficient treatment of the
sickest neonates can only be available by con-
centrating resources, such as equipment and
skilled staff, in a few specialist paediatric centres
that have responsibilities for a particular region.
Newborns who are severely ill and/or have con-
genital anomalies may present in hospitals that
may not be equipped to manage complex paedi-
atric medical requirements; thus, emergency
transport to a paediatric tertiary care facility
U. Rolle
Department of Pediatric Surgery, University Hospital
Frankfurt, Frankfurt/Main, Germany
Department of Pediatric Surgery, Goethe University
Frankfurt, Frankfurt, Germany
e-mail: udo.rolle@kgu.de
P. Puri (*)
Department of Pediatric Surgery, Beacon Hospital,
and University College Dublin, Dublin, Ireland
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_3
3
must be organized (Quinn et al. 2015). It has pre-
viously been shown that the outcome of criti-
cally ill neonates is better if they are cared for in
specialized tertiary centres. In addition, the
prognosis of ill neonates is better if postnatal
care is given in the same hospital in which they
were born. However, not every neonate with
congenital anomalies can be transferred antena-
tally. Those who can must be transported safely
to the specialized centres, sometimes over a con-
siderable distance. The objective is to transport
the critically ill newborn to a tertiary hospital in
the proper condition and at the proper time by an
appropriately trained and skilled team of health
professionals. Moreover, it has also been shown
that the transport per se is not a risk factor for
severe brain injury in outborn infants less than
33 weeks of gestation, whereas the condition at
birth and local postnatal management represent
risk factors (Redpath et al. 2020).
3.2 Prenatal Transfer
The safest and best way to care for both the
mother and newborn is the transfer of the preg-
nant woman to a high-risk perinatal centre before
delivery (Messner 2011). This practice is espe-
cially important for high-risk foetuses, such as
extremely preterm and very low birth weight
foetuses and those with life-threatening neonatal
surgical problems (Puri and Doodnath 2011).
25
26 U. Rolle and P. Puri

Hypothermia remains a main problem in these Table 3.1  Standard protocol for neonatal transportation
babies and adversely affects neonatal outcome. (Modified from Fenton et al. 2004)
Poor post-transfer temperature seems to be an Airway/breathing
independent predictor of death. Therefore, Intubation must be considered before transfer if the
following criteria are met:
whenever possible, threatening preterm delivery
    Oxygen requirement >50%
before 28  weeks of gestation should be con-     Increasing PaCO2
verted to in utero transport (Puri and Doodnath     Recurrent apnoea
2011; Goldsmit et al. 2012; McCall et al. 2008).     <30 weeks of gestation
It is important that the prenatal transport of foe- Control of the correct tube position
tuses with antenatally diagnosed preterm deliv- Respiratory support
ery or surgical abnormalities does not negatively Circulation
Arterial access
influence the outcome. The quality of care before
    Repeated blood gases
and during transport must be appropriate     Measurement of blood pressure
(Longhini et al. 2015). Many studies support the Intravenous access
in utero transportation of high-risk foetuses, par-     Fluids
ticularly very low birth weight (VLBW) babies     Inotropes
and those with life-­threatening neonatal surgical Coagulation
problems (Colaco et  al. 2017). For example,     Vitamin K
Gastrointestinal
infants with transposition of the great arteries
    Nasogastric tube
born outside a specialist centre do not have dif- Temperature
ferent outcomes if they have been transferred to     Support for transport
a specialist centre by a specialist team (Veal Blood glucose
et al. 2019). Furthermore, it has been shown that Measurement and stabilization
very preterm foetuses exhibit increased mortal- Secure intravenous access
ity if born in a non-tertiary centre, whereas Infection
Screening if indicated
infants born in a tertiary centre or after antenatal
Start of treatment
transfer to a tertiary centre exhibit lower mortal- Parent information
ity (Shah et al. 2020). Transfer team/receiving centre information

ent’s information and transfer team information

3.3 Transfer Management
3.3.1 Pre-transfer Management
The transport of a sick neonate is associated with
increased morbidity and mortality. Therefore,
proper pre-transfer stabilization is needed. The
golden rule is that no neonate should be trans-
ported unless the newborn has been sufficiently
stabilized. Careful attention to the pre-transfer
management will provide a higher margin of
safety during the journey when access to the
patient is restricted, and it may be difficult to pro-
vide adequate treatment should problems arise.
Therefore, a standard protocol for pre-transfer
management is recommended.
The key issues in stabilizing neonates for
transfer includes airways and breathing, circula-
tion, temperature, blood glucose, infection, par-
(Table 3.1).
Essential procedures prior to transportation
include peripheral intravenous insertion, arterial
blood gas sampling and endotracheal tube inser-
tion. Other procedures, such as umbilical cathe-
ter insertion, should be spared to save time
(Chakkarapani et al. 2020).
Furthermore, essential data should be trans-
ferred with the infant. Special standardized report-
ing systems are advised (Shadid et al. 2020).
3.3.2 Transfer Team/Transfer
Vehicle
The transport team and transport vehicle should
be a natural physical extension of the intensive
care unit and should be able to supply advanced
critical care management for children of all ages
3  Transport of the Surgical Neonate
at remote sites and during transport to a tertiary
hospital (Ajizian and Nakagawa 2007). Effective
and efficient treatment can only be available by
optimizing resources, medical and technical
equipment, and skilled staff in a few specialist
paediatric centres that have responsibilities for a
particular region (Messner 2011).
Although it is widely accepted that specific
transport training is required for staff transferring
neonatal patients (Fenton and Leslie 2009; Orr
et al. 2009; Stroud et al. 2015), the benefit of spe-
cialist transport teams remains controversial
(Meyer et al. 2016). Several studies have shown
that transport morbidity during high-risk trans-
fers is reduced by having paediatric transport
teams (PTT) on board, due to fewer adverse
events (e.g., improper endotracheal intubation or
loss of vascular access) (Orr et al. 2009; Calhoun
et al. 2017; Edge et al. 1994; Ramnarayan et al.
2010). In contrast, Meyer et al. found no signifi-
cant difference in adjusted 48-h paediatric ICU
mortality for children transported by paediatric
transport teams (Meyer et al. 2016). Furthermore,
a recent Cochrane analysis has shown that there
is no credible evidence from literature-based ran-
domized trials to support or confute the benefits
of specialist neonatal transport staff for neonatal
outcome in terms of morbidity and mortality,
given that there are currently no eligible trials to
compare (Chang et al. 2015).
The mode of transport depends on the travel dis-
tance, geography, weather conditions, ground traf-
fic, vehicle availability, size of the transport team,
nature of the children’s problem and need for speed
(Messner 2011). One should be aware that a dete-
rioration of the patient’s medical condition may be
influenced by transport-related factors, such as the
response and stabilization time or the transport
vehicle used (Puri and Doodnath 2011; Ramnarayan
et al. 2010; Borrows et al. 2010).
3.3.3 Receiving Centre
The continuation of care is essential to improve
neonatal outcome. On arrival at the tertiary cen-
tre, a brief report of prenatal, labour and delivery
history should be given by the transport nurse to
the newborn intensive care nurse, together with
27
details of the newborn’s resuscitation and any
problems experienced during transfer. The
accompanying paediatrician should review the
baby and all documents, together with the accept-
ing surgeon and anaesthetist if necessary. The
parents should be introduced to all staff who will
be involved in the care of their baby. Every pro-
cedure should be explained in a clear and com-
prehensive language to avoid confusion and
parental fear. The consent form should be updated
if necessary. Blood tests and radiological exami-
nations can be ordered subsequently.
3.4 Special Considerations
3.4.1 Gastroschisis
There is clear evidence that the antenatal transfer
of a child with abdominal wall defects is prefer-
able but not essential. Appropriate postnatal
transfer has no negative influence on the outcome
(Drewett et al. 2006; How et al. 2000).
Nevertheless, it has been shown that inborn
babies with uncomplicated gastroschisis have an
improved outcome (Dalton et al. 2017).
A baby with gastroschisis is at a higher risk of
hypothermia, excessive fluid loss and shock,
infection, intestinal strangulation, necrosis and
obstruction due to the small size of the paraum-
bilical defect and the lack of a covering perito-
neal/amniotic membrane (Christison-Lagay et al.
2011; Kastenberg and Dutta 2013). Treatment
starts immediately after delivery to prevent water
and heat loss. Heat loss is a frequent problem,
and hypothermia can result. Therefore, radiant
heating should be available in the room, and the
baby should be kept in a warmed incubator with
the temperature monitored frequently. Intubation
and ventilation are performed if needed. A naso-
gastric tube is passed for intestinal decompres-
sion and the prevention of pulmonary aspiration.
The exposed bowel and the nasogastric aspirates
might result in fluid and protein losses of up to
30 ml/kg/day. (Howell 1998). Therefore, imme-
diate resuscitation with adequate IV fluids
(120 ml/kg/24 h) to overcome substantial water,
electrolyte and protein losses is initiated. Pulse
rate and mean arterial pressure is observed, and
28
blood and glucose measurements are taken.
Vitamin K and broad-spectrum antibiotics (ampi-
cillin, gentamicin and metronidazole) are
­simultaneously administered to reduce contami-
nation of the exposed intestinal loops. A urinary
catheter is passed to decompress the bladder and
to monitor urinary output. The bowel is localized
in the centre of the abdomen, and clingfilm is
used to encircle the exposed intestine and is
wrapped around the baby. Dry sterile gauze
dressing is draped around the clingfilm to support
and protect the highly mobile bowel and to pre-
vent mesenteric injury or venous congestion.
3.4.2 Omphalocele
The initial objectives for the neonatologist are to
assess and treat respiratory distress, protect the
sac from rupture and infection and minimize heat
loss (Chance 1978; Caplan and MacGregor 1989;
McNair et al. 2006). A nasogastric tube is passed
immediately to decompress the stomach and
bowel. The sac should be stabilized in the middle
of the abdomen to prevent kinking of the vessels
and covered with a sterile, dry, non-adherent
dressing to prevent trauma and heat loss.
Intravenous fluids, broad-spectrum antibiotics
and vitamin K should be started.
3.4.3 Pierre Robin Syndrome
Babies with Pierre Robin syndrome carry a high
risk of tongue swallowing and asphyxiation. These
babies should be nursed prone and have an appro-
priate assessment of the anatomical site of the air-
way obstruction (Schaefer and Gosain 2003). A
secure oropharyngeal airway needs to be estab-
lished for transportation (Trevisanuto et al. 2005).
3.4.4 Choanal Atresia
Neonates with choanal atresia suffer from inter-
mittent hypoxia. They should be nursed with an
appropriate sized oral airway to keep the mouth
open (i.e., laryngeal mask airway) or a tooth with
U. Rolle and P. Puri
the end cut off (Trevisanuto et  al. 2005; Lloyd
1996).
3.4.5 Myelomeningocele
An infant with myelomeningocele should be
nursed prone to prevent trauma and pressure on
the spinal area (Lockridge et al. 2002). A warm,
sterile, saline-soaked dressing should be placed
over the lesion, and clingfilm can be wrapped
around the baby to prevent drying and dehis-
cence. If the sac is ruptured and cerebrospinal
fluid (CSF) is leaking or if the myelomeningo-
cele is open, it should be covered with Betadine-­
soaked gauze, and broad-spectrum antibiotics
should be started. Care must be taken to prevent
faecal contamination in sacral lesions. Careful
observation and documentation of neurological
function is essential before, during and after
transport, including the evaluation of the senso-
rimotor level and assessment of the degree of
hydrocephalus. (Fenton and Leslie 2009; Chance
1978) Furthermore, it is important to provide
latex-free care for these neonates since children
with spina bifida have a high index of latex
allergy (Rendeli et al. 2006).
3.4.6 Bladder Exstrophy
At birth, the umbilical cord should be ligated
close to the abdominal wall, and the umbilical
clamp should be removed to prevent mechanical
damage to the bladder mucosa and excoriation of
the bladder surface (Gearhart and Ben-Chaim
1998; Gearhart 2001). Trauma and damage to the
exposed bladder mucosa and plate should be
avoided by covering the defect with the clingfilm
wrapped around the baby to prevent the mucosa
from sticking to clothing or diapers (Gearhart
2001). This procedure allows urine to escape,
while establishing a barrier between the environ-
ment and the fragile bladder mucosa. Old urine,
mucus and any detritus should be washed from
the surface of the bladder with sterile, body tem-
perature saline at each nappy change, and a clean
layer of clingfilm should be applied during trans-
3  Transport of the Surgical Neonate
fer (Gearhart and Ben-Chaim 1998; Howell
1998). Prophylactic antibiotics should be started
immediately (Mercy and Brady-Fryer 2004).
3.4.7 Cloacal Exstrophy
The same measures to protect the omphalocele
sac as discussed in the “Omphalocele” section
are applicable.
3.4.8 Oesophageal Atresia
Once a diagnosis of oesophageal atresia is sus-
pected, the baby should be transferred to a ter-
tiary referral centre for further investigation and
surgery (Spitz 2007). Some babies will require
endotracheal intubation and ventilation
(Table 3.1). These infants are particularly at risk
because mechanical ventilation is relatively inef-
fective due to the presence of a fistula. Therefore,
the tip of the endotracheal tube should be placed
proximal to the carina but distal to the fistula.
Urgent transfer and ligation of the fistula are
essential. Generally, the infant should be handled
with care, and crying should be avoided to reduce
the risk of aspiration, abdominal distension and
subsequent respiratory distress. Moreover, the
baby should be well oxygenated at all times and
kept in a warm environment. Regurgitation of
gastric contents through the fistula during trans-
port can be prevented by keeping the head of the
baby in a slightly elevated position or by nursing
the baby prone or in a right lateral position,
thereby decreasing the work of breathing and
improving oxygenation (Spitz et  al. 1984). The
blind upper oesophageal pouch should be kept
empty. A Replogle suction catheter should be
placed in the pouch and connected to low-­
intermittent or low-continuous suction to prevent
the accumulation of saliva. The perforations
along the side of the catheter are only located
near the tip and therefore minimize the possibil-
ity of suctioning oxygenated air away from the
larynx (O’Neill et al. 1998). However, these dou-
ble lumen oesophageal tubes have a tendency to
become blocked with mucus and therefore should
29
be irrigated at frequent intervals during transport.
Intravenous fluids should be started to provide
maintenance and supplemental fluids and electro-
lytes to compensate oesophageal secretion losses.
Infection should be prevented, and any existing
pneumonitis should be treated using broad-­
spectrum antibiotics. Vitamin K should be admin-
istered prior to transfer.
Recently, it has been established that oesopha-
geal atresia type A can be detected antenatally in
the vast majority of the cases, which leads to bet-
ter antenatal counselling and reduction of postna-
tal transfers (Garabedian et al. 2018).
3.4.9 Congenital Diaphragmatic
Hernia
The initial objective for the neonatologist and
anaesthetist is to stabilize the critically ill neo-
nate before transport to the referral centre. A
large calibre nasogastric tube should be passed
immediately on diagnosis to decompress the gas-
trointestinal tract and prevent further compres-
sion of the lung. Endotracheal intubation should
be performed promptly in a baby with respiratory
difficulty or poor gas exchange. Full sedation and
paralysis will reduce the risk of barotrauma.
Mask ventilation should be avoided because it
will distend the stomach and further compromise
the respiratory status. Hyperventilation using low
pressure and high oxygen content, correction of
acidosis and prevention of thermal and metabolic
stress are recommended to prevent pulmonary
hypertension. IV fluids, fresh frozen plasma and
dopamine, if necessary, should be started to
maintain adequate peripheral perfusion.
Prophylactic antibiotics should be started, and
vitamin K should be administered. Venous access
through the umbilicus is useful for obtaining
mixed venous blood gas specimens and monitor-
ing central venous pressures if passed across the
liver into the right atrium. Arterial access with an
umbilical artery catheter will allow the monitor-
ing of systemic blood pressure and blood gas
measurements at the post ductal level. The baby
will also need a right radial arterial line to mea-
sure preductal blood gasses. This line can be
30
inserted on arrival at the referral centre. Acute
deterioration of the infant’s condition can occur
during transfer due to a pneumothorax.
Equipment for intercostal drainage must be avail-
able as it can be a lifesaving manoeuvre (Spitz
et al. 1984; O’Neill et al. 1998). There has been a
tremendous growth in the use of extracorporeal
life support for neonatal cardiopulmonary failure
in the last two decades. Extracorporeal mem-
brane oxygenation (ECMO) has been used as a
salvage procedure, with 80% survival in high-­
risk neonates with congenital diaphragmatic her-
nia who fail to respond to mechanical ventilation
and meet entry criteria. The number of centres
providing ECMO remain limited, so special ser-
vices are needed to transport critically ill neo-
nates to these centres. These special transport
teams should be familiar with the pathophysiol-
ogy of cardiac and respiratory failure and be
equipped to continue the monitoring and treat-
ment started at the referring centre to maintain
that level of care during transport and treat com-
plications of the disease or therapy itself (Day
and Chapman 1992; Cornish et  al. 1991).
Transportable ECMO systems currently exist that
can effectively stabilize and transport high-risk
neonates to an ECMO competent centre (Faulkner
1995; Wilson et al. 2002). ECMO transportation
has demonstrated acceptable survival but is a
high-risk modality and should not replace early
referral to an ECMO centre (Faulkner 1995;
Kirby and Keijzer 2020).
3.4.10 Intestinal Obstruction
Intestinal obstruction can occur as a result of a
number of conditions, e.g., malrotation, duplica-
tions of the alimentary tract, intestinal atresias, nec-
rotizing enterocolitis, Hirschsprung’s disease,
meconium ileus and anorectal anomalies. The prin-
ciples of care are the same, irrespective of the level
or cause of the obstruction. The main objectives are
to decompress the bowel and prevent aspiration,
accurately estimate and correct fluid losses and
minimize heat loss. A nasogastric tube should be
passed. Suction should be performed every
15–30 min and left on free drainage between aspi-
U. Rolle and P. Puri
rations prior to and during transport. Intravenous
fluids should be started to correct acid-base and
volume deficits and reviewed and adjusted on a 6–8
hourly basis according to the needs of the infant.
Broad-spectrum antibiotics should be started pro-
phylactically (De la Hunt 2006).
Newborn infants with bilious vomiting require
a specialist investigation for potential surgical
pathologies, but this does not seem to be “time-­
critical” and allows previous clinical collabora-
tion (Ojha et al. 2017).
In order to avoid the unnecessary transport of
newborns presenting with bilious vomiting, a
contrast fluoroscopy is advised before transport
(Cullis et al. 2018).
3.4.11 Necrotizing Enterocolitis
Neonates with necrotizing enterocolitis are typi-
cally transferred only if surgery is required in the
case of perforation of gangrenous bowel resulting
in pneumoperitoneum or progressive clinical
deterioration with evidence of peritonitis. These
neonates are typically critically ill with sepsis
and shock. Preferably, the transfer is done while
the infant’s condition is as stable as possible.
Resuscitation with crystalloids, colloids or blood
to correct acidosis is started prior to departure.
Ventilation with intermittent positive pressure
and inotropic support is often required. A sump
nasogastric tube on continuous suction is passed
and suctioned regularly prior and during trans-
port. Broad-spectrum antibiotics are started.
3.5 Conclusion
The approach to the care of high-risk newborns
has changed dramatically in the past 30  years.
Newborn infants with a serious congenital mal-
formation require assessment and stabilization by
experienced staff prior to and during transport to
the receiving centre. Standard protocols need to
be established for the appropriate stabilization of
high-risk newborns before transport, which will
result in a reduction in perinatal morbidity and
mortality.
3  Transport of the Surgical Neonate 31

References Faulkner SC (1995) Mobile extracorporeal mem-

brane oxygenation. Crit Care Nurs Clin North Am
7(2):259–266
Ajizian SJ, Nakagawa TA (2007) Interfacility trans-
Fenton AC, Leslie A (2009) Who should staff neonatal
port of the critically ill pediatric patient. Chest
transport teams? Early Hum Dev 85(8):487–490
132(4):1361–1367
Fenton AC, Leslie A, Skeoch CH (2004) Optimising
Borrows EL, Lutman DH, Montgomery MA, Petros AJ,
neonatal transfer. Arch Dis Child Fetal Neonatal Ed
Ramnarayan P (2010) Effect of patient- and team-­
89:F215–F219
related factors on stabilization time during pediat-
Garabedian C, Bonnard A, Rousseau V, Sfeir R, Drumez
ric intensive care transport. Pediatr Crit Care Med
E, Michaud L, Gottrand F, Houfflin-Debarge V (2018)
11(4):451–456
Management and outcome of neonates with a prenatal
Calhoun A, Keller M, Shi J, Brancato C, Donovan K,
diagnosis of esophageal atresia type A: a population-­
Kraus D et  al (2017) Do pediatric teams affect out-
based study. Prenat Diagn 38:517–522
comes of injured children requiring inter-hospital
Gearhart JP (2001) The Bladder extrophy-epispadias-­
transport? Prehosp Emerg Care 21(2):192–200
cloacal extrophy complex. In: Gearhart, Rink,
Caplan MS, MacGregor SN (1989) Perinatal manage-
Mouriquand (eds) Pediatric urology. WB Saunders
ment of congenital diaphragmatic hernia and anterior
Company, Pennsylvania, pp 511–546
abdominal wall defects. Clin Perinatol 16:917–938
Gearhart JP, Ben-Chaim J (1998) Extrophy and Epispadias.
Chakkarapani AA, Whyte HE, Masse E, Castaldo M,
In: King LR (ed) Urologic surgery in infants and chil-
Yang J, Lee KS (2020) Canadian neonatal trans-
dren. WB Saunders Company, Pennsylvania, p 106
port network. Air Med J 39(4):276–282. https://doi.
Goldsmit G, Rabasa C, Rodriguez S, Aguirre Y, Valdes M,
org/10.1016/j.amj.2020.04.007
Pretz D et  al (2012) Risk factors associated to clini-
Chance GW (1978) Thermal environment in transport. In:
cal deterioration during the transport of sick newborn
Sinclair JC (ed) Temperature regulation and energy
infants. Arch Argent Pediatr 110(4):304–309
metabolism in the newborn. Grune and Stratton,
How HY, Harris BJ, Pietrantoni M, Evans JC, Dutton
New York, pp 227–239
S, Khoury J et  al (2000) Is vaginal delivery prefer-
Chang ASM, Berry A, Jones LJ, Sivasangari S (2015)
able to elective caesarean delivery in fetuses with a
Specialist teams for neonatal transport to neonatal
known ventral wall defect? Am J Obstet Gynecol
intensive care units for prevention of morbidity and
182(6):1527–1525
mortality. Cochrane Database Syst Rev 28:CD007485
Howell KK (1998) Understanding gastroschisis: an
Christison-Lagay ER, Kelleher CM, Langer JC (2011)
abdominal wall defect. Neonatal Netw 17(8):17–25
Neonatal abdominal wall defects. Semin Fetal Neonat
Kastenberg ZJ, Dutta S (2013) Ventral abdominal wall
Med 16:164–172
defects. NeoReviews 14:e402–e411
Colaco SM, Karande T, Bobhate PR, Jiyani R, Rao SG,
Kirby E, Keijzer R (2020) Congenital diaphragmatic
Kulkarni S (2017) Neonates with critical congenital
hernia: current management strategies from antena-
heart defects: impact of fetal diagnosis on immedi-
tal diagnosis to long-term follow up. Pediatr Surg Int
ate and short-term outcomes. Ann Pediatr Cardiol
36:415–429
10:126–130
Lloyd DA (1996) Transfer of the surgical newborn infant.
Cornish JD, Carter JM, Gerstmann DR et  al (1991)
Semin Neonatol 1:241–248
Extracorporeal membrane oxygenation as a means
Lockridge T, Caldwell AD, Jason P (2002) Neonatal sur-
of stabilizing and transporting high risk neonates.
gical emergencies: stabilization and management. J
ASAIO Trans 37:564–556
Obstet Gynecol Neonatal Nurs 31(3):328–339
Cullis PS, Mullan E, Jackson A, Walker G (2018) An audit
Longhini F, Jourdain G, Ammar F, Mokthari M, Boithias
of bilious vomiting in term neonates referred for pedi-
C, Romain O, Letamendia E, Tissieres P, Chabernaud
atric surgical assessment: can we reduce unnecessary
JL, De Luca D (2015) Outcomes of preterm neonates
transfers? J Pediatr Surg 53(11):21-23-27
transferred between tertiary perinatal centers. Pediatr
Dalton BG, Gonzalez KW, Reddy SR, Hendrickson RJ,
Crit Care Med 16:733–738
Iqbal CW (2017) Improved outcomes for inborn
McCall EM, Alderdice FA, Halliday HL, Jenkins JG,
babies with uncomplicated gastroschisis. J Pediatr
Vohra S (2008) Interventions to prevent hypothermia
Surg 52:1132–1134
at birth in preterm and/or low birthweight infants.
Day SE, Chapman RA (1992) Transport of the critically
Cochrane Database Syst Rev 1:CD004210
ill patients in need of extracorporeal life support. Crit
McNair C, Hawes J, Urguhart H (2006) Caring for the
Care Clin 8:581–596
newborn with an omphalocele. Neonatal Netw
De la Hunt MN (2006) The acute abdomen in the new-
25(5):319–327
born. Sem Fetal Neonatal Med 11:191–197
Mercy N, Brady-Fryer B (2004) Bladder exstrophy: a chal-
Drewett M, Michailidis GD, Burge D (2006) The peri-
lenge for nursing care. J Wound Ostomy Continence
natal management of gastroschisis. Early Hum Dev
Nurs 31(5):293–298
82:305–312
Messner H (2011) Neonatal transport: a review of the cur-
Edge WE, Kanter RK, Weigle CG, Walsh RF (1994)
rent evidence. Early Hum Dev 87(Suppl 1):S77
Reduction of morbidity in interhospital transport by
Meyer MT, Mikhailov TA, Kuhn EM, Collins MM,
specialized pediatric staff. Crit Care 22(7):1186–1191
Scanlon MC (2016) Pediatric specialty transport
32
teams are not associated with decreased 48-hour pedi-
atric intensive care unit mortality: a propensity analy-
sis of the VPS. LLC Database Air Med J 35(2):73–78
O’Neill JA, Rowe MI, Grosfeld JL et al (1998) Pediatric
surgery. Mosby - Year Book, St. Louis
Ojha S, Sand L, Ratnavel N, Kempley ST, Sinha AK,
Mohinuddin S, Budge H, Leslie A (2017) Newborn
infants with bilious vomiting: a national audit of neo-
natal transport services. Arch Dis Child Fetal Neonatal
Ed 102:F515–F518
Orr RA, Felmet KA, Han Y, McCloskey KA, Dragotta
MA, Bills DM et al (2009) Pediatric specialized trans-
port teams are associated with improved outcomes.
Pediatrics 124(1):40–48
Puri P, Doodnath R (2011) Transport of the surgical neo-
nate. In: Puri P (ed) Newborn surgery 3E. CRC Press,
London, pp 83–90
Quinn JM, Pierce MC, Adler M (2015) Factors associ-
ated with mode of transport decision making for
pediatric-neo-natal interfacility transport. Air Med J
34(1):44–51
Ramnarayan P, Thiru K, Parslow RC, Harrison DA,
Draper ES, Rowan KM (2010) Effect of specialist
retrieval teams on outcomes in children admitted to
paediatric intensive care units in England and Wales: a
retrospective cohort study. Lancet 376(9742):698–704
Redpath S, Shah PS, Moore GP, Yang J, Toye J, Perreault
T, Lee KS (2020) Canadian neonatal transport net-
work and Canadian neonatal network investigators. J
Perinatol 40(3):385–393
Rendeli C, Nucera E, Ausili E, Tabacco F, Roncallo C,
Pollastrini E, Scorzoni M, Schiavino D, Caldarelli M,
Pietrini D, Patriarca G (2006) Latex sensitisation and
allergy in children with myelomeningocele. Childs
Nerv Syst 22(1):28–32
U. Rolle and P. Puri
Schaefer RB, Gosain AK (2003) Airway management in
patients with isolated Pierre Robin sequence during
the first year of life. J Craniofac Surg 14(4):462–467
Shadid S, Thabane L, Marrin M, Schattauer K, Silenze
L, Borhan S, Singh B, Thoma C, Thomas S (2020)
Evaluation of a modified SBAR report to physi-
cian tool to standardize communication on neonatal
transport. Am J Perinatol 2020 Aug 20. https://doi.
org/10.1055/s-­0050-­1715524
Shah KP, de Regnier RAO, Grobman WA, Bennett AC
(2020) Neonatal mortality after interhospital transfer
of pregnant women for imminent very preterm birth in
Illinois. JAMA Pediatr 174:358–365
Spitz L (2007) Oesophageal atresia. Orphanet J Rare Dis
2:24
Spitz L, Wallis M, Graves HF (1984) Transport of the sur-
gical neonate. Arch Dis Child 59:284–288
Stroud MH, Sanders RCJR, Moss MM, Sullivan JE,
Prodhan P, Melguizo-Castro M et  al (2015) Goal-­
directed resuscitative interventions during pediatric
interfacility transport. Crit Care Med 43(8):1692–1698
Trevisanuto D, Verghese C, Doglioni N, Ferrarese
P, Zanardo V (2005) Laryngeal mask airway for
the interhospital transport of neonates. Pediatrics
115(1):e109–e111
Veal C, Hunt R, Tume LN (2019) Do infants with trans-
position of the great arteries born outside a special-
ist centre have different outcomes? Cardiol Young
29:1030–1035
Wilson BJ, Heiman HS, Butler TJ, Negaard KA,
DiGeronimo R (2002) A 16-year neonatal/pediatric
extracorporeal membrane oxygenation transport expe-
rience. Pediatrics 109(2):189–193
 Pre-operative Management
and Vascular Access
4
Ancuta Muntean, Ionica Stoica, John Gillick,
and Prem Puri

4.1 Introduction nosed early in pregnancy, allowing for familial

counselling, followed by termination in cases of
The surgical procedure performed by the paediat- anomalies that are incompatible with life, plan-
ric surgeon forms only a part of the continuum of ning of foetal surgery in selected cases and hav-
care provided by paediatric surgeons to their ing an optimal delivery plan to ensure survival
patients. The operation is preceded by pre-­ and access to the required treatment for the neo-
operative management involving the assessment nate (Hutson et al. 2014).
and optimization of each individual patient for Therefore, prenatal counselling regarding a
the surgical procedure. The operation itself is baby with an antenatally diagnosed surgical con-
then followed by a post-operative course where dition should be done by a team of specialists in
the patient’s recovery is followed and managed prenatal and perinatal management and should
appropriately. This chapter deals with pre-­ essentially include a paediatric surgeon (Raboei
operative management in the paediatric patient 2008). The parents are often more reassured if the
with specific reference to vascular access. paediatric surgeon that will correct their child’s
anomaly and follow them up from birth is
involved starting in the antenatal period and takes
4.2 Prenatal Diagnosis part in counselling and management of preg-
nancy and birth. Counselling by a paediatric sur-
In the past, paediatric surgeons were not privy to geon that is familiar with the congenital anomaly
the full spectrum of some congenital defects, as during the pregnancy will allow the family to
some foetuses or neonates would not survive come to terms with their baby’s condition, to
long enough to be assessed by a paediatric sur- understand the steps needed to correct it and to
geon. Nowadays, with the significant advances in continue forming a bond with their child (Raboei
prenatal imaging, these pathologies can be diag- 2008; Benachi and Sarnacki 2014). Parents can
see photos of what the defect looks like, before
and after surgery, can have their questions
A. Muntean · I. Stoica · J. Gillick (*)
Children’s Health Ireland at Temple Street, answered, time to process the information given
Dublin, Ireland and their emotions regarding this event.
e-mail: john.gillick@cuh.ie Counselling should aim at offering the pro-
P. Puri spective parents evidence-based information
Department of Pediatric Surgery, Beacon Hospital, regarding the foetal anomaly, its implications on
and University College Dublin, Dublin, Ireland prenatal development and postnatal outcomes,
e-mail: prem.puri@ucd.ie

© Springer Nature Switzerland AG 2023 33

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_4
34
and planning possible foetal interventions,
requirements for delivery (i.e. timing and mode
of delivery that are appropriate for the specific
foetal malformation, maternal transport to an
appropriate specialized centre) and postnatal sur-
gical management. In cases of severe antenatal
anomalies that would be incompatible with post-­
natal life, the option of terminating the pregnancy
can be discussed with the parents; this should
involve a multidisciplinary forum of discussion,
together with the parents and should comply with
the country’s legal basis. Counselling should also
address the risks to subsequent pregnancies
(Berman et al. 2018; Choudhury 2020).
Pathologies that are frequently diagnosed
antenatally and require surgical treatment for
correction, either foetal or postnatal, include
abdominal wall defects, congenital diaphrag-
matic hernia (CDH), lower urinary tract obstruc-
tion (LUTO), congenital tumours and neural tube
defects.
The prognosis of an antenatally diagnosed
anomaly should be evaluated depending on mul-
tiple factors (Hutson et  al. 2014; Choudhury
2020):
a. Type and severity of the malformation: from a
severe anomaly, incompatible with life (i.e.
anencephaly, early intrauterine renal failure
secondary to urethral valves), to correctable
malformations, either through foetal interven-
tions or postnatal surgical treatment, both of
which require a specific timeline and special-
ized centres, in order to optimize outcome.
b. Solitary or multiple anomalies, as multiple
associated malformations tend to have a
poorer prognosis.
c. Syndromic conditions.
d. Associated chromosomal anomalies.
Tennant et al. found a 92% survival at 20 years
for live-born babies with abdominal wall defects
(87.2% for omphalocele and 93.7% for gastros-
chisis) while congenital diaphragmatic hernia
(CDH) has one of the lowest survivals of the
antenatally diagnosed surgical anomalies, being
reported at 44% by Choudhury et al. and 57.1%
by Tennant et al. (Choudhury 2020; Tennant et al.
A. Muntean et al.
2010). Congenital pulmonary airway malforma-
tions (CPAM) that were diagnosed antenatally
associated a survival of 71%, with almost 100%
survival of live-born babies (Choudhury 2020).
Renal anomalies had a poorer prognosis when
diagnosed antenatally at an early stage, when
bilateral and when they were associated with oli-
gohydramnios or with other anomalies
(Choudhury 2020).
During the past two decades, there have been
significant advances in modes and techniques
for prenatal diagnosis. These include amniocen-
tesis, amniography, fetoscopy, foetal sampling
and ultrasonography. As well as these, screening
for chromosomal abnormalities may include
testing for markers in maternal serum and the
non-­ invasive cell-free foetal DNA analysis
(Sinkar et  al. 2020; Minarik et  al. 2015). The
latter has been emerging as a screening tool,
decreasing the use of amniocentesis and chori-
onic villus sampling, both of which carry a risk
for loss of pregnancy, and is indicated in moth-
ers over age 35, known family history or
increased risk of trisomies, aneuploidy risk on
ultrasonography or early pregnancy screening
tests (Committee Opinion No. 640 2015). These
have improved the safety and efficacy of screen-
ing for congenital anomalies for both the foetus
and the mother.
Prenatal ultrasound remains a powerful tool in
the diagnosis and surveillance of foetal anoma-
lies, as it has a widespread use and is safe for both
mother and child. Advances in ultrasound tech-
nology, with enhancement in resolution, is allow-
ing increased detection of antenatal anomalies.
Major anomalies, like anencephaly or cystic
hygromas, are more frequently detected at the
first trimester ultrasound, along with assessment
of viability of foetus and gemelarity, while
second-­ trimester ultrasonography at 18 to
20 weeks evaluates for anatomic anomalies (Fetal
Diagnosis and Treatment Committee 2013;
iFIND n.d.). Real-time sonography as well as
serial ultrasounds can assess changes in the func-
tion or anatomy of a foetal organ and evaluate the
progression of a foetal abnormality (Berkley
et  al. 2009). However, ultrasound remains
operator-­dependent and access to these advances
4  Pre-operative Management and Vascular Access
in technology may be limited in some regions,
leading to deficiencies in prenatal diagnosis.
A new research project called iFIND
(Intelligent Fetal Imaging and Diagnosis), based
in the UK and scheduled to finalize in 2021, is
aiming to develop a robotic automated system
that uses advanced artificial intelligence tech-
niques to help sonographers screen for foetal
abnormalities and increase the rate of their detec-
tion (iFIND n.d.).
Ultrafast foetal MRI is another useful inves-
tigation that can increase detection rate of asso-
ciated abnormalities. In CDH it can identify
liver herniation and estimate lung volumes
(Ramasauskaite 2018).
In the case of certain congenital anomalies, a
foetal intervention may be sought. This specialty
has evolved in the past decades from solitary
selected cases of severe abnormalities to a new
multidisciplinary specialty aimed at decreasing
the morbidity associated with some foetal diag-
noses and ensuring postnatal survival (Fetal
Diagnosis and Treatment Committee 2013).
This field has also seen a shift from open foe-
tal surgery to minimally invasive foetoscopic sur-
gery with different degrees of success. Although
outcomes vary by condition, severity and
approach, good results have been published for
foetal surgery in spina bifida, with promising out-
comes in foetoscopic repair as well, while the
PLUTO trial for vesico-amniotic shunting in foe-
tal lower urinary tract obstructions has shown
improved postnatal survival but no significant
amelioration of long-term renal function or sur-
vival. Addition of serial amnioinfusions may pre-
vent the severe pulmonary hypoplasia associated
with oligohydramnios, although clinical trials are
still underway (Ruano 2020).
In CDH, temporary foetoscopic endoluminal
tracheal occlusion (FETO) has been shown to
increase survival for severe CDH while for
extremely severe cases, early placement of the
balloon compared to standard FETO seems to
improve the outcomes for these foetuses.
However, as preterm delivery and premature rup-
ture of membranes are still significant complica-
tions, further research is needed to optimize the
appropriate selection of cases and increase its
35
safety. One such study is the TOTAL trial, cur-
rently ongoing in Europe, for moderate and
severe forms of CDH. However, studies seem to
suggest that FETO may promote foetal lung
regeneration (Ruano 2020).
Centres that specialize in foetal surgery can
now offer a range of other procedures like: foeto-
scopic laser photocoagulation for twin-twin
transfusion syndrome, radiofrequency ablation of
foetal tumours, foetal shunt placement, amniotic
band resections, foetal cardiac intervention, open
foetal surgery for sacrococcygeal teratoma (SCT)
or CPAM resections and Ex-utero Intrapartum
Treatment (EXIT) (Choudhury 2020; Ruano
2020).
Many neonates born with congenital malfor-
mations still require urgent postnatal surgery and,
in addition to their surgical problem, may suffer
from a multitude of medical problems.
Furthermore, they are at a period when signifi-
cant physiological and maturational changes of
transition from foetal to extrauterine life occur.
Surgical and anaesthetic intervention at this time
may affect this transition by interfering with nor-
mal homeostatic controls of circulation, ventila-
tion, temperature, fluid and metabolic balance.
Thus, all prenatal information is an integral part
of the pre-operative assessment of a newborn
with a congenital anomaly.
To facilitate a smooth pre-operative course,
close coordination between the obstetrician, neo-
natologist, paediatric surgeon and paediatric
anaesthesiologist is necessary. An ideally planned
delivery should take place in a centre of paediat-
ric surgical excellence.
All neonates undergoing surgery must be
carefully assessed pre-operatively, particularly
paying attention to the following:
• History and physical examination.
• Maintenance of body temperature.
• Respiratory function.
• Cardiovascular status.
• Metabolic status.
• Coagulation abnormalities.
• Laboratory investigations.
• Fluid and electrolytes, and metabolic responses.
• Vascular access.
36
4.3 History and Physical
Examination
The history of the newborn starts months before
delivery, as many of the congenital malforma-
tions (e.g. CDH, omphalocele, gastroschisis and
SCT) nowadays are known to the paediatric
­surgeon, prenatally. Not only are the anatomical
and structural anomalies important, but meta-
bolic abnormalities or chromosomal aberrations
as well, which must be diagnosed prenatally or
immediately after birth.
Anticipation of a problem in the delivery room
is often based on prenatal diagnosis. For exam-
ple, identification of a trisomy 21  in the foetus
will increase the neonatologist’s awareness in
evaluating the infant for those abnormalities
closely associated with this chromosomal defect,
e.g. evaluation for duodenal atresia and congeni-
tal heart disease. Conversely, correct prenatal
identification of a duodenal obstruction and dif-
ferentiating it from a lower obstruction, should
trigger investigation for the associated chromo-
somal abnormality (Hutson et al. 2014).
The most recent advance in prenatal detection
of anatomical problems has been the develop-
ment of foetal ultrasonography. However, this
anatomical prenatal diagnosis is only one of the
tools that aids in planning postnatal management.
An accurate and well-documented family history
may increase the suspicion that an infant is at risk
for an anatomical defect linked to an inherited
disorder. In other cases, the evidence of polyhy-
dramnios or oligohydramnios should signifi-
cantly increase the suspicion of a congenital
anomaly.
Most conditions are best managed expectantly
by natural labour and vaginal delivery at term.
Improved outcomes have been reported for the
caesarean mode of delivery in antenatally diag-
nosed anomalies; however, it should be retained
only for selected cases of meningomyelocele
(nonvertex position of foetus at labour, hydro-
cephalus or large sac), hydrocephalus with con-
comitant macrocephaly, large anterior wall
defects with extracorporeal liver, large sacrococ-
A. Muntean et al.
cygeal teratomas, hydrops foetalis or large foetal
tumours that may cause maternal dystocia
(Wataganara et  al. 2017). Gastroschisis alone
should not be an indication for c-section and it
has been suggested that delivery at 37–38 weeks,
can decrease neonatal morbidity and prevent
term stillbirths. In some situations, like congeni-
tal high airway obstruction (CHAOS) which can
be secondary to laryngeal or tracheal stenosis/
atresia or obstructing foetal neck lesions, an
Ex-utero Intrapartum Treatment (EXIT) proce-
dure may be indicated. The infant is delivered by
elective caesarean section, while maintaining
foeto-placental circulation, which allows secur-
ing the child’s airway in a controlled manner
(Wataganara et al. 2017).
After birth, the assessment of the degree of
prematurity, which is an integral part of the phys-
ical examination, and the specific type of con-
genital anomaly must be identified and recorded
because of the profound anaesthetic and post-­
operative implications that are involved. The nor-
mal full-term infant has a gestational age of
37 weeks or more and a body weight greater than
2500 g. Infants born with a birth weight of less
than 2500  g are defined as being of low birth
weight (LBW). Babies may be of LBW because
they have been born too early (preterm—earlier
than 37  weeks’ gestational age) or because of
intrauterine abnormalities affecting growth
(growth retardation). “Small-for-gestational-age”
(SGA) infants are those whose birth weight is
less than the tenth percentile for their age. Infants
may, of course, be both growth retarded and born
preterm.
The principle features of prematurity are as
follows:
• A head circumference below the 50th
percentile.
• A thin, semi-transparent skin.
• Soft, malleable ears.
• Absence of breast tissue.
• Absence of plantar creases.
• Undescended testicles with flat scrotum and,
in females, relatively enlarged labia minora.
4  Pre-operative Management and Vascular Access
The physiological and clinical characteristics
of these babies are as follows:
• Apnoeic spells.
• Bradycardia.
• Hypothermia.
• Sepsis.
• Hyaline membrane disease.
• Blindness and lung injury due to use of high
levels of oxygen.
• Patent ductus arteriosus.
In the SGA infant, although the body weight is
low, the body length and head circumference
approach are that of an infant of normal weight
for age. These babies are older and more mature.
Their clinical and physiological characteristics
are given below:
• Higher metabolic rate.
• Hypoglycaemia.
• Thermal instability.
• Polycythemia.
• Increased risk of meconium aspiration syn-
drome.
4.4 Maintenance of Body
Temperature
The World Health Organization (WHO) defines
hypothermia as a newborn’s temperature that
drops below 36.5  °C and classifies it as cold
stress or mild hypothermia (36.0–36.4 °C), mod-
erate hypothermia (32.0–35.9  °C) and severe
hypothermia (below 32  °C) (World Health
Organization 1997). Further research showed
that the mean normal temperature for newborns
is lower than previously described, with a value
of 36.5 ± 0.6 °C (Takayama et al. 2000).
Newborn infants, particularly premature
infants, have poor thermal stability because of a
higher surface area-to-weight ratio, a thin layer
of insulating subcutaneous fat and a high thermo-
neutral temperature zone. The newborn readily
loses heat by conduction, convection, radiation
37
and evaporation, with the major mechanism
being radiation (World Health Organization
1997). Shivering thermogenesis is absent in the
neonate, and the heat producing mechanism is
limited to non-shivering thermogenesis through
the metabolising of brown fat (Silverman and
Sinclair 1966). Cold stress-induced thermogene-
sis in neonates leads to an increased metabolic
rate, oxygen and calories consumption to main-
tain body temperature. If prolonged, this leads to
depletion of the limited energy reserve and pre-
disposes to hypothermia and increased mortality.
Hypothermia can also suggest infection and
should trigger diagnostic evaluation and antibi-
otic treatment if required (Takayama et al. 2000).
Body temperature is a strong predictor of mor-
tality and morbidity across all gestational ages
(Mccall et al. 2018). Risk factors for developing
hypothermia include: delivery room tempera-
ture < 25 °C, maternal temperature < 36 °C, cae-
sarean delivery, low birth weight, low Apgar
score and failure in using preventive measures
such as plastics wrap and head covering (Miller
et al. 2011; De Almeida et al. 2014). Illness in the
newborn, particularly when associated with pre-
maturity, further compounds the problems with
maintenance of body temperature. The classic
example for such an illness is the newborn with
omphalocele or gastroschisis. In a group of 23
neonates with gastroschisis, Muraji et  al. found
that hypothermia (31–35.4 °C), which was found
in seven patients on arrival at the hospital, was
the most serious pre-operative problem (Muraji
et  al. 1989). To minimize heat losses, the neo-
nates should be nursed in thermally neutral envi-
ronments, which allows maintenance of a normal
body temperature with minimal oxygen con-
sumption and caloric expenditure. The equip-
ment available to maintain a thermal neutral
environment are represented by incubators, over-
head radiant warmers, thermal mattresses and
warming blankets (World Health Organization
1997). The incubators are efficient for maintain-
ing the baby’s temperature, but do not allow ade-
quate access to the sick baby for active
resuscitation and observation. Overhead radiant
38
heaters, servo-controlled by a temperature probe
on the baby’s skin, are preferred and effective in
maintaining the baby’s temperatures; they also
provide visual and electronic monitoring and
access for nursing and medical procedures. In an
open environment such as theatre, the neonate
should be kept warm by increasing the room
­temperature, using a warming blanket such as
Bear Huggers and use of warmed ventilator gas,
intravenous and irrigation fluids (Buisson et  al.
2004). Hyperthermia should be avoided, because
it is associated with perinatal respiratory depres-
sion, seizures and hypoxic brain injury
(Lieberman et al. 2000). Caution must be taken to
avoid iatrogenic hyperthermia, particularly when
multiple interventions to prevent hypothermia are
used simultaneously, such as thermal bags and
thermal mattresses (Mccall et al. 2018). Low and
high admission temperatures are more frequent
in extremely premature infants (Laptook et  al.
2018).
The environmental temperature must be main-
tained near the appropriate thermoneutral zone
for each individual patient because the increase
in oxygen consumption is proportional to the gra-
dient between the skin and the environmental
temperature. This is 34–35 °C for LBW infants
up to 12 days of age and 31–32 °C at 6 weeks of
age. Infants weighing 2000–3000 g have a ther-
moneutral zone of 31–34 °C at birth and 29–31 °C
at 12  days. The neutral temperature during the
first week of life is dependent on gestational and
postnatal age, whereas after the first week of life
it is dependent on body weight and postnatal age
(Sauer et  al. 1984). In an incubator, either the
ambient temperature of the incubator can be
monitored and maintained at thermoneutrality, or
a servo-system can be used. The latter regulates
the incubator temperature according to the
patient’s skin temperature, which is monitored by
means of a skin probe on the infant. The normal
skin temperature for a full-term infant is 36.2 °C,
but because of many reasons, benign factors such
as excessive bundling and ambient temperature
may affect body temperature. Diurnal and sea-
sonal variations in body temperature have also
been described (Takayama et al. 2000). Thus, the
control of the thermal environment of the new-
A. Muntean et al.
born and especially the ill baby with congenital
malformations is of utmost importance to the
outcome.
4.5 Respiratory Function
Assessment of respiratory function is essential in
all neonates undergoing surgery. The main clini-
cal features of respiratory distress are restless-
ness, tachypnoea, grunting, nasal flaring, chest
wall recession, cyanosis and apnoea. These
symptoms are occasionally present in the deliv-
ery room due to the anatomical abnormalities
involving the airway and lungs and require the
most urgent therapy, as failure to recognize them
can lead to fatal complications (Gleason and Juul
2017). There are multiple conditions that can
cause respiratory distress post-delivery, the most
common being transient tachypnoea of newborn,
respiratory distress syndrome (RDS), meconium
aspiration syndrome, pneumothorax and cardiac
failure secondary to congenital heart disease.
Surgical conditions include diaphragmatic hernia
(Bochdalek), bronchopulmonary malformations
(congenital pulmonary airway malformations,
bronchopulmonary sequestration, lobar emphy-
sema), oesophageal atresia with or without
tracheo-­esophageal fistula, choanal atresia
(Edwards et al. 2013). It is important to recognize
that more than one condition may be present in
the same patient. First-line investigations in the
assessment of a baby with signs of respiratory
insufficiency should include pulse oximetry,
bloods tests (full blood count, CRP, blood gas
and blood cultures) and a chest radiograph
(Edwards et  al. 2013). It is recommended for a
radio-opaque nasogastric tube to be passed and a
thoraco-abdominal radiograph to be performed in
order to determine the cause of the respiratory
distress and to localize the oesophagus, stomach
and bowel gas, avoiding late diagnosis or misdi-
agnosis of a surgical pathology such as oesopha-
geal atresia or congenital diaphragmatic hernia.
Blood gas studies are essential in the diagno-
sis and management of respiratory distress.
Arterial PO2 and PCO2 indicate the state of oxy-
genation and ventilation, respectively. In the
4  Pre-operative Management and Vascular Access
newborn, repeated arterial blood samples may be
obtained either by catheterization of an umbilical
artery or by cannulation of radial, brachial or
posterior tibial arteries (Askin 1997a). An impor-
tant alternative is non-invasive monitoring
­technique with transcutaneous PO2 monitors or
pulse oximeters. More recently, combined trans-
cutaneous measurement of both SaO2 and PaCO2
has proved possible. Monitoring of arterial pH is
also essential in patients with respiratory distress.
Acidosis in the neonate produces pulmonary
arterial vasoconstriction and myocardial depres-
sion. Respiratory alkalosis causes decreased car-
diac output, decreased cerebral blood flow,
diminished oxyhaemoglobin dissociation and
increased airway resistance with diminished pul-
monary compliance.
Respiratory failure is the leading cause of
death in the neonate, but significant progress
has been made in the neonatal respiratory sup-
port field. Multiple devices can be used to sup-
port a patient with respiratory failure, from
simple oxygen delivery to use of surfactant,
inhaled nitric oxide (iNO), high-frequency ven-
tilation (HFOV), and extracorporeal life sup-
port (ECLS). These interventions have been
shown to improve survival dramatically in
selected neonates.
Surfactant replacement therapy plays an
essential role in the clinical management of neo-
nates with RDS. The European Consensus guide-
line on the management of RDS recommends
administration of animal-derived surfactant prep-
aration early in the course of the disease in neo-
nates with RDS. Traditionally administration of
surfactant is intratracheal through the endotra-
cheal tube. However, a less invasive surfactant
administration (LISA) method has been devel-
oped, using a fine catheter placed in the trachea
while the newborn is spontaneously breathing on
CPAP (Sweet et al. 2019). It may also be effec-
tive in other forms of lung disease, such as meco-
nium aspiration syndrome (MAS), neonatal
pneumonia, the “adult” form of acute respiratory
distress syndrome (ARDS) and congenital dia-
phragmatic hernia (CDH). It ensures that alveolar
stability is promoted, atelectasis is reduced,
39
oedema formation is decreased and the overall
work of respiration is minimized (McCabe et al.
2000).
Inhaled nitric oxide (iNO) results in preferen-
tial pulmonary arterial vasodilation and decreased
pulmonary vascular resistance leading to dimin-
ished extrapulmonary shunt and has a microselec-
tive effect that improves ventilation and perfusion
matching. It is available for treatment of persistent
pulmonary hypertension of the neonate (PPHTN),
but its usefulness depends on the underlying cause
of pulmonary hypertension (PHTN). In infants
with MAS or RDS, iNO has shown to be effective
in reducing pulmonary vascular resistance and the
need for ECMO (Pandya and Puligandla 2015). Its
beneficial role in the treatment of PHTN associ-
ated with congenital diaphragmatic hernia has
become more uncertain, currently iNO is not rec-
ommended routinely in CDH patients (Puligandla
et al. 2015; Finer et al. 2017). Recent research sug-
gests that iNO treatment may have a role in the
management of a subpopulation of patients with
CDH with PHTN and normal left ventricular sys-
tolic function (Lawrence et al. 2020).
In newborns with severe lung disease, HFOV
is frequently used to optimize lung inflation and
minimize lung injury. The combination of HFOV
and iNO is reported to be more successful in
improvement in oxygenation in newborns with
severe PPHN complicated by diffuse parenchy-
mal lung disease and under inflation (Kinsella
et  al. 1997). A Cochrane review published in
2015 compared elective HFOV versus conven-
tional ventilation in preterm neonates with acute
pulmonary dysfunction and the authors con-
cluded that the use of HFOV results in a small
reduction of chronic lung disease, but the evi-
dence was weak. (Cools et al. 2015).
ECLS provides long-term cardiopulmonary
support for patients with reversible pulmonary
and cardiac insufficiency, therefore it is a sup-
portive and not a curative therapy. ECLS and
extracorporeal membrane oxygenation (ECMO)
are terms often used interchangeably, but at pres-
ent ECLS in the preferred terminology. It is well
accepted as a standard of treatment for neonatal
respiratory and/or cardiac failure with potentially
40
reversible aetiology, refractory to conventional
techniques of pulmonary support. The most com-
mon neonatal indications are congenital dia-
phragmatic hernia, meconium aspiration
syndrome, persistent pulmonary hypertension
and sepsis (Wild et al. 2020).
Extracorporeal Life Support Organisation
(ELSO) is a registry established in 1989 with the
aim to collect data on use of ECLS at global level
and in 2019 ELSO collected data from 463 cen-
tres across the world. The report published in
July 2020 summarized an overall experience of a
total of 133.371 patients being managed with
ECLS, of which 32.634 were neonates with
respiratory failure, of which 73% survived to dis-
charge or transfer (Extracorporeal Life Support
Organisation 2020).
The indications and contraindications for
ECLS have changed over time. The ESLO neo-
natal respiratory guideline published in 2020 rec-
ommends ECLS in neonates with one or more of
the following: inadequate tissue oxygen delivery
despite maximal therapy (rising lactate, worsen-
ing metabolic acidosis, signs of end organ dys-
function), severe hypoxic respiratory failure with
acute decompensation (PaO2 < 40 mmHg), oxy-
genation index with sustained elevation and no
improvement and severe pulmonary hypertension
with evidence of right ventricular dysfunction
and/or left ventricular dysfunction (Wild et  al.
2020). The current absolute contraindications are
represented by lethal chromosomal disorder or
other lethal anomaly, severe brain damage,
uncontrollable bleeding, significant intraventric-
ular haemorrhage and vessel size too small for
cannulation. The relative contraindications are
irreversible organ damage, gestational age
<34 weeks and weight <2000 g (Wild et al. 2020).
In 2008, a Cochrane review comparing ECMO
and conventional therapy was published and the
authors’ conclusions were that using ECMO sup-
port in infants with severe but potentially revers-
ible respiratory failure results in significantly
improved survival. Of the total 244 infants, 77%
survived in the ECMO group compared with
44% in the conventional therapy group (Mugford
et al. 2008).
A. Muntean et al.
In summary, the type of respiratory care in
neonates will always depend on clinical and
radiological findings supported by blood gas
estimations.
4.6 Cardiovascular Status
At birth, the circulation undergoes rapid transi-
tion from foetal to neonatal pattern. In full-term
neonates, the ductus arteriosus normally closes
functionally by smooth muscle contraction
within 18 to 24 hours after birth, while anatomi-
cal occlusion of the lumen occurs over the next
few days or weeks (Gournay 2011). Prior to birth,
the pulmonary arterioles are relatively muscular
and constricted. With the first breath, total pul-
monary resistance falls rapidly because of the
unkinking of vessels with the expansion of the
lungs and also because of the vasodilatory effect
of inspired oxygen. However, during the first few
weeks of life, the muscular pulmonary arterioles
retain a significant capacity for constriction, and
any constricting influences such as hypoxia may
result in rapid return of pulmonary hypertension
(Gleason and Juul 2017).
The management of neonates with congenital
malformations is frequently complicated by the
presence of congenital heart disease. At this time
of life, recognition of heart disease is particularly
difficult. There may be no murmur audible on
first examination, but a loud murmur can be audi-
ble a few hours, days or a week later (McNamara
1990). A newborn with congenital anomalies
undergoing surgery should have a full cardiovas-
cular examination and a chest X-ray. In the
asymptomatic patient, a thorough physical exam
may be sufficient pre-operatively, but if there is
suspicion of a cardiac anomaly, the neonate
should have an echocardiogram and be examined
by a paediatric cardiologist (Petroze and
Puligandla 2019). In recent years, the use of
detailed foetal echocardiography performed by
skilled physicians has increased the prenatal
detection rate of major congenital heart disease,
providing valuable information to optimize peri-
natal management (Chu et al. 2017).
4  Pre-operative Management and Vascular Access
4.7 Metabolic Status
4.7.1 Acid–Base Balance
The buffer system, renal function and respiratory
function are the three major mechanisms respon-
sible for the maintenance of normal acid–base
balance in body fluids. Most newborn infants can
adapt competently to the physiological stresses
of extrauterine life and have a normal acid–base
balance, with a maintenance of the systemic arte-
rial pH between 7.35 and 7.45 (Iacobelli and
Guignard 2020). However, clinical conditions
such as RDS, sepsis, congenital renal disorders
and gastrointestinal disorders may result in gross
acid–base disturbances in the newborn. The four
basic disturbances of acid–base physiology are
metabolic acidosis, metabolic alkalosis, respira-
tory acidosis and respiratory alkalosis. In a new-
born undergoing surgery, identification of the
type of disorder, whether metabolic or respira-
tory, simple or mixed, is of great practical impor-
tance to permit the most suitable choice of
therapy, and for it to be initiated in a timely fash-
ion (Askin 1997b). The acid–base state should be
determined by arterial blood gases and pH esti-
mation, and must be corrected by appropriate
metabolic or respiratory measures prior to
operation.
4.7.2 Hypoglycaemia
The mechanisms of glucose homeostasis are not
well developed in the early post-natal period; this
predisposes the neonate, especially the premature
neonate, to the risk of both hypoglycaemia and
hyperglycaemia. Prenatally, the glucose require-
ments of the foetus are obtained almost entirely
from the mother, with very little derived from
foetal gluconeogenesis. Following delivery, the
plasma glucose concentrations drop to a level of
3–3.3 mmol/l (55–60 mg/dl) by 1–2 hours of life,
and then steadily increase over the next few days
(Stanley et al. 2015). The limited liver glycogen
stores are rapidly depleted and the blood glucose
level depends on the infant’s capacity for gluco-
neogenesis, the adequacy of substitute stores and
41
energy requirements. Three mechanisms may
lead to infantile hypoglycaemia (Hutson et  al.
2014): limited glycogen stores (Raboei 2008),
hyperinsulinism and (Benachi and Sarnacki
2014) diminished glucose production. It is impor-
tant to differentiate between transitional neonatal
glucose regulation and hypoglycaemia that per-
sists or occurs after 2–3 days of life, as the latter
needs effective treatment in order to avoid brain
injury and long-term neurodevelopmental defi-
cits (Thornton et al. 2015).
Risk factors for developing hypoglycaemia
include prematurity, foetal distress, birth
asphyxia, low Apgar score, LBW, especially
SGA, macrosomia, infants of toxaemic or dia-
betic mothers, Beckwith-Wiedemann syndrome
and infants requiring surgery who are unable to
take oral nutrition and who have the additional
metabolic stresses of their disease (Stanley et al.
2015; Sweet et al. 2013).
The Pediatric Endocrine Society recommends
evaluation and treatment for hypoglycaemia in
infants with a plasma glucose below 3.3 mmol/l
(<60  mg/dl), which is the threshold for neuro-
genic responses (Thornton et  al. 2015). The
symptomatic infant should be treated urgently
with intravenous glucose as a bolus of 200 mg/kg
(dextrose 10% at 2 mL/kg), followed by continu-
ous infusion of dextrose 10% at 5 to 8 mg/kg per
minute (80 to 100  mL/kg per day) to maintain
blood glucose levels of 40 to 50 mg/dl (Adamkin
et al. 2011).
During the perioperative period, dextrose sup-
plementation is necessary in neonates and infants
to avoid hypoglycaemia, hypercatabolism, keto-
genesis and delayed hyperglyceamia (Datta and
Aravindan 2017).
Neonates are susceptible to electrolyte imbal-
ances such as hypocalcaemia and hypomagnese-
mia, particularly in association with prematurity
and SGA.  Correction of these deficiencies may
be required in the pre-operative period.
4.7.3 Hyperbilirubinaemia
Jaundice in the newborn is a common physiologi-
cal problem that can be observed in up to 84% of
42
infants, approximately 60% of term and 80% of
preterm babies, during the first week of life and
generally resolves by 14  days of life (Bhutani
et al. 2013; NICE 2016).
Physiological neonatal jaundice is the result
of a combination of decreased red cell survival
and increased enterohepatic circulation of biliru-
bin, with a consequent increase in bilirubin load,
an immature glucuronyl-transferase enzyme sys-
tem with a limited capacity for conjugating bili-
rubin and a decreased hepatic uptake of bilirubin
from plasma (Maisels 2006). Thus, a degree of
hyperbilirubinemia secondary to a normal physi-
ological process is common in neonates, espe-
cially in breastfed babies, and it is harmless
(NICE 2016).
Jaundice usually becomes apparent in the face
and then progresses distally as the bilirubin con-
centration rises. Visual inspection is not a reliable
means to determine the severity of hyperbilirubi-
nemia or to discriminate between conjugated and
unconjugated bilirubin fraction (Gleason and
Juul 2017). Although visual assessment is an
important clinical practice, it is recommended
that all newborns have a total serum bilirubin
(TSB) or a transcutaneous bilirubin measured
before discharge (Maisels 2006).
Hyperbilirubinaemia in the newborn may
have a pathological basis such as severe sepsis,
blood group incompatibility (most commonly Rh
and ABO incompatibility), congenital haemo-
lytic anaemias, bruising and metabolic disorders
(NICE 2016). Neonatal haemolytic jaundice usu-
ally appears during the first 24 h of life, whereas
physiological jaundice, reaches a peak between 2
and 5  days of life. Other causes for prolonged
hyperbilirubinaemia, including those often asso-
ciated with surgical conditions are biliary
obstruction, hepatocellular dysfunction and
upper intestinal tract obstruction. Extra-hepatic
biliary obstruction should be diagnosed as early
as possible, because early operation for biliary
atresia is essential to obtain good short-term as
well as long-term results (Hung et al. 2006).
Predisposing factors include male gender, pre-
maturity, infections (TORCH), hypoalbuminae-
mia (circulating bilirubin is bound to albumin),
hypothermia, acidosis, hypoglycaemia, hypoxia,
A. Muntean et al.
caloric deprivation and the use of drugs (e.g.
chloramphenicol, gentamicin, digoxin and furo-
semide) (Porter and Dennis 2002).
When the serum bilirubin concentration is
high, hyperbilirubinemia must be treated as the
infant can develop acute bilirubin encephalopa-
thy and kernicterus (American Academy of
Pediatrics Subcommittee on Hyperbilirubinemia
2004). The bilirubin level is used to determine
the management of hyperbilirubinemia. In most
patients, other than those with severe haemolysis,
phototherapy is a safe and effective method and
should be initiated at a level of TSB of 20 mg/dl
(342 micromol/l) or more (Maisels 2006).
Exchange transfusion is indicated when TSB
rises early and rapidly despite intensive photo-
therapy or when the infant shows signs of acute
bilirubin encephalopathy (American Academy of
Pediatrics Subcommittee on Hyperbilirubinemia
2004).
4.7.4 Coagulation Abnormalities
Coagulation abnormalities in the neonate should
be sought pre-operatively and treated. The new-
born is deficient in vitamin K and 1  mg of the
same should be administered prior to the opera-
tion in order to prevent hypoprothrombinaemia
and haemorrhagic disease in the newborn.
Although given routinely in most developed
countries immediately after birth, its administra-
tion should be confirmed with the labour suite.
Neonates, especially the premature babies, have a
high risk of developing disseminated intravascu-
lar coagulopathy when there is an underlying pre-
disposing condition such as sepsis, birth asphyxia,
RDS and necrotizing enterocolitis. The treatment
strategy includes addressing the underlying
pathology and normalizing the coagulation sys-
tem by administering pre-operatively fresh-­
frozen plasma, fresh blood or platelet concentrate
(Veldman et al. 2010).
Bleeding is one of the major risks associated
with ECLS and can occur in up to 70% of patients
with different frequency between sites (Dalton
et al. 2017). The most common localizations are
the surgical site, the cannula site and chest drain
4  Pre-operative Management and Vascular Access 43

site, occurring in around one-third of the patients.acidosis, hypernatremia and cardiovascular col-
Central nervous system haemorrhage occurs in lapse. Administration of excessive fluid may
around 16% of patients and can have a particu- result in pulmonary oedema, congestive heart
larly devastating outcome (Wild et  al. 2020; failure, opening of ductal shunts, bronchopulmo-
Dalton et al. 2017). Bleeding is usually the result nary dysphasia and cerebral intraventricular
of systemic heparinization, and further research haemorrhage.
to both establish standardized anticoagulation In order to plan accurate fluid and electrolyte
regimens and monitoring for haemolysis (Dalton therapy for the newborn, it is essential to under-
et al. 2017) and to develop an extracorporeal cir- stand the normal body “water” consumption and
cuit that can function as the endothelium so that the routes through which water and solute are lost
systemic anticoagulation can be evaded from the baby. In foetal life around 16  weeks’
(Ontaneda and Annich 2018) may help identify gestation, total body water (TBW) represents
the optimal care practice. approximately 90% of total body weight, and the
proportions of extracellular and intracellular
water components are 65% and 25%, respec-
4.7.5 Laboratory Investigations tively (Friis-Hansen 1983). At term, these two
compartments constitute about 45% and 30%,
A newborn undergoing surgery should have respectively, of total body weight, indicating that
blood drawn on admission for various investiga- (Hutson et  al. 2014) a shift from extracellular
tions, including full blood count, serum sodium, water to intracellular water occurs during devel-
potassium and chloride, urea, calcium, magne- opment from foetal to neonatal life and (Raboei
sium, glucose, bilirubin and group and cross-­ 2008) relative total body and extracellular fluid
match. Blood gases and pH estimation should volume both decrease with increasing gestational
also be obtained to assess acid–base state and the age (Friis-Hansen 1983).
status of gas exchange. The availability of micro-­ In very small premature infants, water consti-
methods in the laboratory has minimized the tutes as much as 85% of total body weight and in
amount of blood required to do the above blood the term infants it represents 75% of body weight.
tests. The coagulation status of infants who have The total body water decreases progressively
suffered asphyxia or sepsis may be abnormal and during the first few months of life, falling to 65%
can result in disseminated intravascular clotting of body weight at the age of 12  months, after
and severe thrombocytopenia (Veldman et  al. which it remains fairly constant (Statter 1992).
2010). A platelet count <50,000/mm3 in the neo- The extracellular and intracellular fluid volumes
nate is an indication for pre-operative platelet also change with growth.
transfusion. Blood cultures should be obtained The objectives of parenteral fluid therapy are
whenever there is any suspicion of sepsis. to provide the following:

• Maintenance fluid requirements needed by the

4.7.6 Fluid and Electrolytes,
and Metabolic Responses
Estimation of the parenteral fluid and electrolyte
requirements is an essential part of the manage-
ment of newborn infants with surgical conditions.
Inaccurate assessment of fluid requirements,
especially in premature babies and LBW infants,
may result in a number of serious complications
(Bhatia 2006). Inadequate fluid intake may lead
to dehydration, hypotension, poor perfusion with
body to maintain vital functions.
• Replacement of pre-existing deficits and
abnormal losses.
• Basic maintenance requirement of water for
growth.
Maintenance fluid requirement consists of
water and electrolytes that are normally lost
through insensible loss, sweat, urine and stools.
The amount lost through various sources must be
calculated to determine the volume of fluid to be
44
administered. Insensible loss is the loss of water
from the pulmonary system and evaporative loss
from the skin. Approximately 30% of the insen-
sible water loss (IWL) occurs through the pulmo-
nary system as moisture in the expired gas; the
remainder (about 70%) is lost through the skin
(Statter 1992). Numerous factors are known to
influence the magnitude of insensible water loss.
These include the infant’s environment (ambient
humidity and temperature), metabolic rate, respi-
ratory rate, gestational maturity, surface area and
presence of skin defects, fever and the use of
radiant warmers and phototherapy (Oh 2012). A
precise value of IWL can be calculated with
accurate recording over 24 hours of intake, out-
put and weight, using the formula: Intake—
Output (mainly urine during the first week of
life)—Difference in weight (grams). (Oh 2012).
The respiratory water loss is approximately
5 ml/kg/24 h and is negligible when infants are
intubated and on a ventilator. Water loss through
sweat is generally negligible in the newborn
except in patients with cystic fibrosis, severe con-
gestive heart failure or high environmental tem-
perature. Faecal water losses are 5–20 ml/kg/day.
As neonates have such variable fluid require-
ments, depending on all the factors discussed
above, fluid administration should be calculated
with precision and should aim to maintain a urine
output above 1  mL/kg/h (Eaton et  al. 2018). In
the first days of life, the baby should receive only
10% dextrose solution, with specific rates
depending on prematurity and also accounting
for the initial weight loss. Transition is usually
made most commonly to 5% dextrose with ½
normal saline and the rates of administration vary
depending on the neonates’ day of life and gesta-
tional age. However, losses will usually be
replaced with normal saline, only adding potas-
sium once urine output is established.
Bodyweight, serum electrolytes and an exact
input–output balance should be monitored with
regularity.
If it is expected that enteral feeding will be
delayed for more than 4–5 days, as with neonatal
digestive obstructions, parenteral nutrition should
be commenced. However, precocious initiation
of enteral feeds is recommended, as there is proof
A. Muntean et al.
that trophic enteral intake can stimulate bowel
adaptation, digestion and immune function and
will also reduce time to reaching full enteral
feeds (Eaton et al. 2018; Penman et al. 2017).
In surgical neonates, feeding will mainly
depend on the pathology, need for bowel resec-
tion and for stoma formation. Nasogastric aspi-
rates usually inform the physician if the bowel is
ready for feeding, with colour changing from
dark green to yellow and the quantity decreasing.
Freshly expressed maternal milk is usually the
first choice for feeding, though in certain situa-
tions like limited length of small bowel remain-
ing after surgery or a proximal stoma, special
hydrolysed formulas may be needed to facilitate
absorption. In some situations, like necrotising
enterocolitis (NEC), bowel rest will remain indi-
cated (Penman et al. 2017).
4.7.7 Renal Function, Urine Volume
and Concentration
in the Newborn
The kidneys are the final pathway regulating fluid
and electrolyte balance of the body. The urine
volume is dependent on water intake, the quantity
of solute for excretion and the maximal concen-
trating and diluting abilities of the kidneys. Renal
function in the newborn varies with gestational
age and should be evaluated in this context. Very
preterm infants younger than 34  weeks gesta-
tional age have reduced glomerular filtration rate
(GFR) and tubular immaturity in the handling of
the filtered solutes when compared to term
infants. Premature infants between 34 and
37 weeks gestational age undergo rapid matura-
tion of renal function similar to term infants with
rapid establishment of glomerulo-tubular balance
early in the post-natal period (Shaffer and
Norman 1989). The full-term newborn can con-
centrate urine to an osmolarity of 800 mOsm/kg,
while preterm newborn only to 600  mOsm/kg.
Urinary dilution capability is normal in term neo-
nates but decreased in preterm neonates (Chien
et al. 2020).
If the volume of fluid administered is inade-
quate, urine volume falls and concentration
4  Pre-operative Management and Vascular Access
increases. With excess fluid administration, the
opposite occurs. We aim to achieve a urine output
of 2 ml/kg/h, which will maintain a urine osmo-
larity of 250–290  mmol/kg (specific gravity
1009–1012) in newborn infants. For older infants
and children, hydration is adequate if the urine
output is 1–2  ml/kg/h, with an osmolarity
between 280 and 300 mmol/kg.
Creatinine can also be an indicator of glomer-
ular filtration, although it has its limitations, as it
mirrors maternal serum creatinine in the first
days of life and it can be falsely elevated in pre-
mature babies from days to weeks, due to its
tubular resorption. New markers for glomerular
filtration like cystatin C are being studied and
have proven some limited use although it may
provide better results in conjunction with creati-
nine (Kastl 2017).
Accurate measurements of urine flow and
concentration are fundamental to the manage-
ment of critically ill infants and children, espe-
cially those with surgical conditions and extensive
tissue destruction or with infusion of high osmo-
larity solutions. In these situations, it is recom-
mended that urine volume be collected and
measured accurately.
4.8 Pain in Neonates
Neonates frequently experience discomfort and
pain as a result of a disease, a diagnostic or thera-
peutic procedure. The prevention and alleviation
of pain in neonates is important as firstly it is an
ethical requirement and secondly repetitive pain
has been associated with long-term neurodevel-
opmental, behavioural and cognitive sequelae
(Pediatrics 2016).
Although assessing pain in a neonate can be
very challenging, currently there are a number of
validated neonatal pain assessment tools avail-
able (Giordano et  al. 2019). The American
Academy of Pediatrics recommends using the
following five scales: PIPP (Premature Infant
Pain Profile), NFCS (Neonatal Facial Coding
System), N-PASS (Neonatal Pain, Agitation and
Sedation scale), DAN (Douleur Aiguë
Nouveau-né scale) and BIIP (Behavioural
45
Indicators of Infant Pain) (Pediatrics 2016). A
unique assessment tool has not been adopted uni-
versally because each scale was developed and
validated for a specific population and clinical
setting (Giordano et al. 2019).
Using nonpharmacologic strategies such as
tucking, non-nutritive sucking, breastfeeding,
oral sucrose/glucose solutions and sensorial stim-
ulation have been proven to be very effective in
decreasing periprocedural pain (Pediatrics 2016).
4.9 Pre-operative Management
in the Older Child
Many of the pre-operative management strategies
used in the neonate are equally applicable to the
older child. Pre-operative management needs to
be tailored according to a number of factors,
including the urgency of the surgery, the age of
the child and any associated medical conditions
affecting the patient. In the setting of a sick
patient requiring an urgent operation, aggressive
resuscitation according to Advanced Paediatric
Life Support protocols may be required. In the
elective setting there is usually more time for a
meticulous pre-operative course.
Pre-operative evaluation in older children
should involve the children themselves under-
standing and consenting to the planned proce-
dure. The patient may demonstrate sufficient
knowledge of the proposed surgery to give or
withhold consent and the issue of “Gillick com-
petence” will require evaluation when obtaining
informed consent (Wheeler 2006). The level of
maturity possessed by the patient may well dic-
tate the provision of in-depth explanation of the
planned surgery, including play therapy, model-
ling, operating room tour and coping skills for
the parents. Authors have stressed the utility of
these facilities and the importance of providing
leaflets to support spoken information in order to
optimize the psychological pre-operative prepa-
ration of the child and their parents (Kumar et al.
2019; Landier et al. 2018). The unfortunate alter-
native that is still common place involves the
physical restraint of an extremely anxious, non-­
compliant child in the operating theatre.
46
Associated medical conditions in paediatric
surgical practice may have significant bearing on
pre-operative management of these patients.
There has been a significant shift in favour of day
case paediatric surgery in recent years. The inci-
dence of unplanned admission following day
case surgery remains low at approximately 2.5%,
with the most frequent reasons for admission
being unexpected surgical complexity, pain, post-
operative nausea and vomiting and duration of
procedure (Green et  al. 2020). Close liaison
between paediatric surgeons and their anaesthetic
colleagues pre-operatively should minimize post-­
operative complications such as unplanned
admission following day surgery. In an ideal situ-
ation the anaesthetist would review the child in
the outpatient setting in partnership with the sur-
gical team.
Although the majority of children undergo-
ing anaesthesia are healthy, it is crucial to
detect any underlying risk factor that may lead
to an unexpected adverse event in the peri-
operative period. However, pre-operative
assessment should not involve unnecessary
tests that create a stressful environment for the
child and the family prior to surgery. Especially
for ambulatory patients, a focussed pre-opera-
tive clinic that could be nurse-­led may be of
benefit.
Unfortunately, accurate pre-operative assess-
ment of paediatric patients for peri-operative sur-
gical and anaesthetic risk remains difficult. One
of the most utilized tools for risk assessment, the
American Society of Anaesthesiologists (ASA)
grade, has been shown to have a significant
degree of interoperator variability. There is a
need for a more refined tool for pre-operative
physical status grading in the paediatric popula-
tion (Aplin et al. 2007).
As in the neonatal population, adequate pre-­
operative preparation may involve investigations
such as routine bloodwork, radiological studies
and identification and correction of underlying
medical conditions that might contribute to peri-­
operative morbidity. If blood loss is anticipated,
the availability of cross-matched blood should be
ensured. Also, it is imperative that all relevant
radiological investigations should be present in
A. Muntean et al.
the operating room at the time of surgery. This is
particularly crucial if unexpected operative find-
ings arise and especially when operating on
paired organs.
4.10 Vascular Access
The paediatric population is formed of patients
across a range of ages and a variety of conditions.
A significant percent of the inpatients requires
either a short term vascular access device (VAD)
in the setting of an acute pathology or a long term
VAD in case of a chronic condition. Appropriate
VAD selection and insertion have an impact on
the patient’s management and outcome and the
device selection must be based on a number of
criteria: age, therapy duration and frequency,
infusate characteristics (Ullman et al. 2020). The
venous access can be obtained via a peripheral
line or a central line.
The types of peripheral lines are represented by:
• peripheral intravenous cannula (PIVC)  - this
is the simplest method of venous access and it
is used for short term, usually less than 7 days,
administration of peripherally compatible
therapy such as antibiotics and intravenous
fluids. The most frequently used sites are the
veins along the dorsal side of the hand, ante-
cubital fossae, veins on the dorsal aspect of
the foot and the scalp veins in neonates. With
the improvements of techniques, equipment
and the availability of 22–24 gauge plastic
cannulas, percutaneous cannulation of veins
has become possible even in small premature
infants and at present it is rarely necessary to
perform a “cut-down”. In older children,
larger bore cannulas, such as 20–18 gauge, are
frequently more appropriate.
• midline catheter - is a line that is longer than a
peripheral cannula but shorter than a peripher-
ally inserted central catheter (PICC), its tip is
placed in a larger portion of the vein with a
greater blood flow. This device is indicated in
patients who require peripherally compatible
therapy for a duration that exceeds 6  days
(Moureau and Chopra 2016),
4  Pre-operative Management and Vascular Access 47

• intraosseous access (IO)—is an excellent tempo- central line as they can be placed at bed side
rary vascular access alternative in an emergency. without general anaesthesia and can be used for
The most common site for intraosseous infusion a longer duration, usually between 4 weeks and
is the proximal tibia, with secondary sites includ- six months (Chesshyre et al. 2015). This type of
ing the distal tibia and distal femur. The duration line is preferred in neonates and in patients that
of use is a few hours and it is important to can receive treatment at home. These catheters
remember that positive pressure is required to can be successfully inserted by dedicated nurs-
infuse intravenous fluids via the intraosseous ing personnel to provide long-term venous
route. It is mandatory to continuously monitor access with a reduced incidence of thromboem-
the limb for any signs of extravasation. bolic complications (Pitts 2013). A variety of
peripheral veins lend themselves to the place-
In the resuscitation scenario, where large vol- ment of PICC lines, the preferred insertion site
umes of fluid may need to be given quickly, it is is the basilic vein and particularly the scalp
important to remember Poiseuille’s equation, in veins in most neonates (Callejas et  al. 2016)
that the flow rate through a cannula is propor- (Fig.  4.1). To minimize thrombotic complica-
tional to the fourth power of the cannula’s radius, tions, it is important to ensure that the catheter
thus emphasising the importance of wide-bore tip resides in a central vein.
cannulas in the trauma setting in the older child • Tunnelled central lines (Broviac or
(Lewis et al. 2013). Hickman)—are the preferred access for
The types of central lines are represented by administration of nonperipherally compatible
the following: therapy such as chemotherapy or total paren-
teral ­nutrition for a long duration, usually
• Umbilical catheters—are represented by cath- months to years (Carraro et  al. 2013). These
eters that are passed via the umbilical vein. It are commonly placed in either the neck or
is an appropriate mean of venous access for groin veins in neonates, and in older children
the first week of life, but the main concern is
the tip positioning due to small size of the
patient and difficult securement. Malposition
within the portal circulation can result in
severe hepatic injury (Ares and Hunter 2017).
• Nontunneled central lines—are the preferred
access for administration of peripherally and
nonperipherally compatible therapy for short
duration, respectively less than 14 days. These
catheters are available in different sizes, lengths
and number of lumens. The primary sites for
insertion are the internal jugular vein, subcla-
vian vein and femoral vein. The risk of infec-
tion is not different between these sites, but the
femoral lines have a higher risk of thrombus
formation (Derderian et al. 2019). This central
line is often mandatory and is a basic monitor-
ing aid for the anaesthetist at the time of opera-
tion, and sometimes can be performed in theatre
immediately before starting the operation.
• Peripherally inserted central catheters (PICC)—
represent a valuable alternative to a nontunneled Fig. 4.1  PICC line insertion in superficial temporal vein
48 A. Muntean et al.

the jugular and subclavian veins are the pre-

ferred sites, either approached via an open
technique or percutaneously (Figs.  4.2, 4.3,
4.4). The line is then tunnelled away from the
insertion site to a skin exit site, where a cuff is
placed under the skin to decrease the risk of
dislodgment and infection (Ares and Hunter
2017). These catheters are available in differ-
ent sizes and number of lumens.
Internal Common carotid
• Totally implantable central lines (ports)—rep- Jugular v. a. and
resent an option in patients that suffer from vagus n.

chronic conditions such cystic fibrosis or

sickle cell disease and only require intermit-
tent venous access over a long period of time
(Ullman et al. 2020; Ares and Hunter 2017).
This device enables the patient to carry out
many day-to-day activities such as swimming.
Fig. 4.4  Open approach to right internal jugular vein
cannulation

Internal
jugular vein

Right
subclavian vein

Right
brachiocephalic vein
Superior vene vava

Fig. 4.2  Central venous access sites on the right side of

the neck

Superior epigastric vein

Superior Fig. 4.5 Implantable devices for long-term vascular

circumflex access
iliac vein

Lateral It is our practice to insert the reservoir subcu-

femoral
cutaneous taneously in the axilla, where it is readily
vein accessible but cosmetically inconspicuous. It
Saphenous vein
is important that a non-coring “Huber” needle
is used to access the silicone membrane of the
Fig. 4.3  Venous access sites in the right groin port in order to avoid damage (Fig. 4.5).
4  Pre-operative Management and Vascular Access
The goal for all central venous access devices
(CVADs) is to provide safe and reliable vascular
access for treatment, without complications
related to insertion, maintenance or removal
(Ullman et al. 2015).
The tip position of a central venous catheter
should be at the junction of the right atrium and
superior vena cava, parallel to the vein wall and
position should always be confirmed with chest
radiograph before using the catheter (Ares and
Hunter 2017). Catheters that are inserted in small
children and that are likely to be used for many
years have the tip sometimes placed in the right
atrium to allow for further change in position as
the patient grows.
When choosing the catheter, the size and the
number of lumens should be carefully considered
in order minimize the risk of complications. Due
to the risk of thrombosis, catheter-to-vein ratio
should be <45% (Ullman et  al. 2020) and to
decrease the risk of infection a catheter with the
minimum number of lumens or ports necessary
should be selected (Ullman et al. 2020; Chesshyre
et al. 2015) Good fixation of all venous and arte-
rial lines is essential to avoid dislodgement, as
reinsertion of these vascular lines can be very
difficult.
Critically ill patients will require an arterial
line especially at the time of operation, either
because of the surgery, when it is expected to
result in significant fluid shift and haemodynamic
instability, or in a neonate, because of a signifi-
cant underlying cardiopulmonary disease of the
newborn. This arterial line is for monitoring the
haemodynamic and biochemical status, espe-
cially throughout the operative procedure. In the
neonate, right radial artery percutaneous cathe-
terization is preferred because it allows sampling
of preductal blood for measurement of oxygen
tension. If the baby already has an umbilical
artery catheter, it is safer to use it strictly for the
purpose of blood pressure monitoring and blood
sampling and not for the administration of drugs.
In older children, the radial arteries are the pre-
ferred sites, generally followed by the posterior
tibial and femoral arteries, but brachial artery can
be used as well (Schindler et al. 2005). Access to
the relevant artery is generally obtained by a
49
ultrasound guided percutaneous technique but
also may be approached via a formal cut-down
(Aouad-Maroun et al. 2016).
Occasionally obtaining vascular access can be
anticipated to be particularly difficult. This fre-
quently occurs in the setting of a child who has
had multiple previous central venous lines.
Screening modalities such as venous Doppler
ultrasound, magnetic resonance venograms
(MRV) or computer tomography angiogram
(CTA) may give information on the patency of
the remaining veins (Shankar et al. 2002).
Unfortunately, in children with end-stage cen-
tral venous access, sites that are more difficult to
access may have to be considered in order to
obtain adequate vascular access. These can
include transhepatic Broviac insertion via a per-
cutaneous approach, thoracoscopy or formal tho-
racotomy with cannulation of the azygos vein or
superior vena cava or in extreme circumstances
insertion of a central line through the auricle of
the right atrium (Rodrigues et al. 2006).
Overall 25% of paediatric CVADs fail before
completion of treatment, with an incidence rate
of 1.97 per 1000 catheter days (Ullman et  al.
2015). The complications that can occur with
placement of a central venous access device can
be divided into early and long-term. The rate of
immediate complications is reported between
7–18% of CVAD placements and are represented
by bleeding or haematoma, vascular injury from
needle insertion or dilator (arterial puncture, per-
foration of SVC or right atrium), pneumothorax,
range of arrythmias due to catheter tip, air or
thrombotic pulmonary embolism and incorrect
placement (Ullman et  al. 2015). The long-term
risks are central line-associated bloodstream
infections (CLABSI), thrombosis and risk of pul-
monary embolism, blockage, catheter fracture or
migration (Ares and Hunter 2017). A systematic
review showed that, overall 10.3% of paediatric
CVADs developed an associated blood stream
infection with tunnelled CVADs having the high-
est risk of infection or blockage, while the totally
implanted devices had the lowest rate of failure
and complications (Ullman et  al. 2015). Most
CLABSI respond to appropriate antibiotic and
salvage of an infected line should be considered,
50
unless the patient’s condition worsens or if bacte-
raemia persists longer than 72 hours after initia-
tion of appropriate antibiotic treatment (Wolf
et al. 2013).
Some institutions have adopted a specialized
nurse led paediatric vascular access team that
resulted in an improvement in the reduction of
central line-associated blood stream infections, a
reduced need for sedation and general anaesthe-
sia and a better patient experience (Pitts 2013).
References
Adamkin DH, Papile LA, Baley JE, Bhutani VK, Carlo
WA, Kumar P et al (2011) Clinical report – Postnatal
glucose homeostasis in late-preterm and term infants.
Pediatrics 127(3):575–579
American Academy of Pediatrics Subcommittee on
Hyperbilirubinemia (2004 Jul) Management of
Hyperbilirubinemia in the Newborn Infant 35 or
More Weeks of Gestation. Pediatrics [Internet]
114(1):297–316. Available from http://www.citeulike.
org/group/11862/article/5941222
Aouad-Maroun M, Raphael CK, Sayyid SK, Farah F,
Akl EA (2016 Sep) Ultrasound-guided arterial can-
nulation for paediatrics. Cochrane Database Syst
Rev [Internet] (9). Available from http://doi.wiley.
com/10.1002/14651858.CD011364.pub2
Aplin S, Baines D, De Lima J (2007) Use of the ASA
Physical Status Grading System in pediatric practice.
Paediatr Anaesth 17(3):216–222
Ares G, Hunter CJ (2017) Central venous access in chil-
dren: indications, devices, and risks. Curr Opin Pediatr
29(3):340–346
Askin DF (1997a Aug) Interpretation of neonatal blood
gases, Part I: Physiology and acid-base homeostasis.
Neonatal Netw [Internet] 16(5):17–21. Available from
http://www.ncbi.nlm.nih.gov/pubmed/9325869
Askin DF (1997b Sep) Interpretation of neonatal blood
gases, Part II: Disorders of acid-base balance.
Neonatal Netw [Internet] 16(6):23–9. Available from
http://www.ncbi.nlm.nih.gov/pubmed/9325876
Benachi A, Sarnacki S (2014) Prenatal counselling and
the role of the paediatric surgeon. Semin Pediatr Surg
[Internet] 23(5):240–243. https://doi.org/10.1053/j.
sempedsurg.2014.09.002
Berkley EMF, Goens MB, Karr S, Rappaport V (2009 Jul)
Utility of fetal echocardiography in postnatal manage-
ment of infants with prenatally diagnosed congenital
heart disease. Prenat Diagn [Internet] 29(7):654–8.
Available from http://doi.wiley.com/10.1002/pd.2260
Berman L, Jackson J, Miller K, Kowalski R, Kolm P,
Luks FI (2018) Expert surgical consensus for prenatal
counseling using the Delphi method. J Pediatr Surg
A. Muntean et al.
[Internet] 53(8):1592–1599. https://doi.org/10.1016/j.
jpedsurg.2017.11.056
Bhatia J (2006 May) Fluid and electrolyte management
in the very low birth weight neonate. J Perinatol
[Internet] 26(S1):S19–21. Available from http://www.
nature.com/articles/7211466
Bhutani VK, Stark AR, Lazzeroni LC, Poland R, Gourley
GR, Kazmierczak S et al (2013) Predischarge screen-
ing for severe neonatal hyperbilirubinemia identifies
infants who need phototherapy. J Pediatr [Internet]
162(3):477–482.e1. https://doi.org/10.1016/j.
jpeds.2012.08.022
Buisson P, Bach V, Elabbassi EB, Chardon K, Delanaud S,
Canarelli JP et al (2004) Assessment of the efficiency
of warming devices during neonatal surgery. Eur J
Appl Physiol 92(6):694–697
Callejas A, Osiovich H, Ting JY (2016) Use of peripher-
ally inserted central catheters (PICC) via scalp veins in
neonates. J Matern Neonatal Med 29(21):3434–3438
Carraro F, Cicalese MP, Cesaro S, De Santis R, Zanazzo
G, Tornesello A et  al (2013) Guidelines for the use
of long-term central venous catheter in children with
hemato-oncological disorders. on behalf of support-
ive therapy working group of Italian Association of
Pediatric Hematology and Oncology (AIEOP). Ann
Hematol 92(10):1405–1412
Chesshyre E, Goff Z, Bowen A, Carapetis J (2015)
The prevention, diagnosis and management of cen-
tral venous line infections in children. J Infect
[Internet] 71(S1):S59–S75. https://doi.org/10.1016/j.
jinf.2015.04.029
Chien YH, Chen YL, Tsai LY, Mu SC (2020) Impact
of urine osmolality/urine sodium on the timing of
diuretic phase and non-invasive ventilation sup-
port: differences from late preterm to term neonates.
Pediatr Neonatol [Internet] 61(1):25–30. https://doi.
org/10.1016/j.pedneo.2019.04.006
Choudhury S (2020) Fetal malformations and the
pediatric surgeon. J Indian Assoc Pediatr Surg
[Internet] 25(2):65. http://www.jiaps.com/text.
asp?2020/25/2/65/276940
Chu C, Yan Y, Ren Y, Li X, Gui Y (2017 Apr) Prenatal
diagnosis of congenital heart diseases by fetal echo-
cardiography in second trimester: a Chinese multi-
center study. Acta Obstet Gynecol Scand [Internet]
96(4):454–63. Available from http://doi.wiley.
com/10.1111/aogs.13085
Committee Opinion No. 640: Cell-Free DNA Screening
For Fetal Aneuploidy (2015 Sep) Obstet Gynecol
[Internet] 126(3):e31–e37. Available from http://jour-
nals.lww.com/00006250-­201509000-­00051
Cools F, Henderson-Smart DJ, Offringa M, Askie LM
(2015) Elective high frequency oscillatory ventilation
versus conventional ventilation for acute pulmonary
dysfunction in preterm infants. Cochrane Database
Syst Rev (3)
Dalton HJ, Reeder R, Garcia-Filion P, Holubkov R, Berg
RA, Zuppa A et  al (2017) Factors associated with
bleeding and thrombosis in children receiving extra-
4  Pre-operative Management and Vascular Access
corporeal membrane oxygenation. Am J Respir Crit
Care Med 196(6):762–771
Datta P, Aravindan A (2017) Glucose for children dur-
ing surgery: pros, cons, and protocols: a postgradu-
ate educational review. Anesth Essays Res [Internet]
11(3):539. Available from http://www.aeronline.org/
text.asp?2017/11/3/539/208678
De Almeida MFB, Guinsburg R, Sancho GA, Rosa IRM,
Lamy ZC, Martinez FE et al (2014) Hypothermia and
early neonatal mortality in preterm infants. J Pediatr
164(2):271–276
Derderian SC, Good R, Vuille-dit-Bille RN, Carpenter
T, Bensard DD (2019) Central venous lines in criti-
cally ill children: thrombosis but not infection is site
dependent. J Pediatr Surg [Internet] 54(9):1740–1743.
https://doi.org/10.1016/j.jpedsurg.2018.10.109
Eaton S, Coppi P De, Pierro A (2018) Fluid, electorlyte
and nutritional support of the surgical neonate. In:
Rickham’s Neonatal Surgery
Edwards MO, Kotecha SJ, Kotecha S (2013) Respiratory
distress of the term Newborn Infant. Paediatr Respir
Rev [Internet] 14(1):29–37. https://doi.org/10.1016/j.
prrv.2012.02.002
Extracorporeal Life Support Organisation. ECLS Registry
Report [Internet] (2020). Available from: https://www.
elso.org/Registry/Statistics/InternationalSummary.
aspx
Fetal Diagnosis and Treatment Committee. Fetal diag-
nosis and therapy: a reference handbook for pediatric
surgeons. American Pediatric Surgical Association.
2013
Finer N, Barrington KJ, Pennaforte T, Altit G (2017)
Nitric oxide for respiratory failure in infants born at or
near term (Review). Cochrane Libr (1)
Friis-Hansen B (1983 Feb) Water distribution in the
foetus and newborn infant. Acta Paediatr [Internet]
72(s305):7–11. Available from http://doi.wiley.
com/10.1111/j.1651-­2227.1983.tb09852.x
Giordano V, Edobor J, Deindl P, Wildner B, Goeral K,
Steinbauer P et al (2019) Pain and sedation scales for
neonatal and pediatric patients in a preverbal stage
of development: a systematic review. JAMA Pediatr
173(12):1186–1197
Gleason C, Juul S (2017) Avery’s diseases of the Newborn,
10th edn. Elsevier Health Sciences, p 1656
Gournay V (2011) The ductus arteriosus: physiology, reg-
ulation, and functional and congenital anomalies. Arch
Cardiovasc Dis 104(11):578–585
Green Z, Woodman N, McLernon DJ, Engelhardt T (2020)
Incidence of paediatric unplanned day-case admis-
sions in the UK and Ireland: a prospective multicentre
observational study. Br J Anaesth 124(4):463–472
Hung PY, Chen CC, Chen WJ, Lai HS, Hsu WM, Lee PH
et al (2006) Long-term prognosis of patients with bili-
ary atresia: a 25 year summary. J Pediatr Gastroenterol
Nutr 42(2):190–195
Hutson JM, O’Brien M, Beasley SW, Teague WJ,
King SK, eds (2014) Jones’ clinical paediatric sur-
gery [Internet]. Chichester: Wiley. http://doi.wiley.
com/10.1002/9781118777305
51
Iacobelli S, Guignard J-P (2020 Feb) Renal aspects of
metabolic acid–base disorders in neonates. Pediatr
Nephrol [Internet] 35(2):221–8. Available from http://
link.springer.com/10.1007/s00467-­018-­4142-­9
iFIND: intelligent Fetal Imaging and Diagnosis [Internet]
(n.d.). Available from: http://www.ifindproject.com
Kastl JT (2017) Renal function in the fetus and neo-
nate  – the creatinine enigma. Semin Fetal Neonatal
Med [Internet] 22(2):83–89. https://doi.org/10.1016/j.
siny.2016.12.002
Kinsella JP, Truog WE, Walsh WF, Goldberg RN,
Bancalari E, Mayock DE et  al (1997) Randomized,
multicenter trial of inhaled nitric oxide and high-­
frequency oscillatory ventilation in severe, persistent
pulmonary hypertension of the newborn. J Pediatr
131(1I):55–62
Kumar A, Das S, Chauhan S, Kiran U, Satapathy S (2019)
Perioperative anxiety and stress in children undergo-
ing congenital cardiac surgery and their parents: effect
of brief intervention—a randomized control trial. J
Cardiothorac Vasc Anesth [Internet] 33(5):1244–1250.
https://doi.org/10.1053/j.jvca.2018.08.187
Landier M, Villemagne T, Le Touze A, Braïk K, Meignan
P, Cook AR et  al (2018) The position of a written
document in preoperative information for pediatric
surgery: a randomized controlled trial on parental
anxiety, knowledge, and satisfaction. J Pediatr Surg
[Internet]. 53(3):375–380. https://doi.org/10.1016/j.
jpedsurg.2017.04.009
Laptook AR, Bell EF, Shankaran S, Boghossian NS,
Wyckoff MH, Kandefer S et  al (2018) Admission
temperature and associated mortality and morbidity
among moderately and extremely preterm infants. J
Pediatr 192:53–59.e2
Lawrence KM, Monos S, Adams S, Herkert L, Peranteau
WH, Munson DA et  al (2020) Inhaled nitric oxide
is associated with improved oxygenation in a sub-
population of infants with congenital diaphrag-
matic hernia and pulmonary hypertension. J Pediatr
[Internet] 219:167–172. https://doi.org/10.1016/j.
jpeds.2019.09.052
Lewis GC, Crapo SA, Williams JG (2013) Critical skills
and procedures in emergency medicine. Vascular
access skills and procedures. Emerg Med Clin North
Am 31(1):59–86
Lieberman E, Lang J, Richardson DK, Frigoletto
FD, Heffner LJ, Cohen A (2000) Intrapartum
maternal fever and neonatal outcome. Pediatrics
105(1I):8–13
Maisels MJ (2006 Dec) Neonatal Jaundice. Pediatr
Rev [Internet] 27(12):443–454. Available from
http://pedsinreview.aappublications.org/lookup/
doi/10.1542/pir.27-­12-­443
McCabe AJ, Wilcox DT, Holm BA, Glick PL (2000)
Surfactant – a review for pediatric surgeons. J Pediatr
Surg 35(12):1687–1700
Mccall EM, Alderdice F, Halliday HL, Vohra S, Johnston
L (2018) Interventions to prevent hypothermia at birth
in preterm and/or low birth weight infants. Cochrane
Database Syst Rev 2018(2)
52
McNamara DG (1990) Value and limitations of ausculta-
tion in the management of congenital heart disease.
Pediatr Clin North Am [Internet] 37(1):93–113.
https://doi.org/10.1016/S0031-­3955(16)36834-­1
Miller SS, Lee HC, Gould JB (2011) Hypothermia in very
low birth weight infants: Distribution, risk factors and
outcomes. J Perinatol [Internet] 31(SUPPL. 1):S49–
S56. https://doi.org/10.1038/jp.2010.177
Minarik G, Repiska G, Hyblova M, Nagyova E, Soltys
K, Budis J et  al (2015) Utilization of benchtop next
generation sequencing platforms ion torrent PGM and
miseq in noninvasive prenatal testing for chromosome
21 trisomy and testing of impact of in silico and physi-
cal size selection on its analytical performance. PLoS
One 10(12):1–12
Moureau N, Chopra V (2016 Sep) Indications for periph-
eral, midline, and central catheters: summary of the
michigan appropriateness guide for intravenous
catheters recommendations. J Assoc Vasc Access
[Internet] 21(3):140–8. Available from: https://merid-
ian.allenpress.com/java/article/21/3/140/434328/
Indications-­for-­Peripheral-­Midline-­and-­Central
Mugford M, Elbourne D, Field D (2008 Jul) Extracorporeal
membrane oxygenation for severe respiratory fail-
ure in newborn infants. Cochrane Database Syst Rev
[Internet] 11(1):50–1. Available from http://doi.wiley.
com/10.1002/14651858.CD001340.pub2
Muraji T, Tsugawa C, Nishijima E, Tanano H, Matsumoto
Y, Kimura K (1989 Apr) Gastroschisis: a 17-year
experience. J Pediatr Surg [Internet] 24(4):343–5.
Available from https://linkinghub.elsevier.com/
retrieve/pii/S0022346889802659
NICE (2016) Neonatal jaundice. Clinical Guideline
[Internet]. Available from https://www.nice.org.uk/
guidance/cg98/evidence/full-­guideline-­245411821
Oh W (2012) Fluid and electrolyte management of
very low birth weight infants. Pediatr Neonatol
[Internet] 53(6):329–333. https://doi.org/10.1016/j.
pedneo.2012.08.010
Ontaneda A, Annich GM.  Novel surfaces in extracor-
poreal membrane oxygenation circuits. Front Med
2018;5(Nov):1–9
Pandya KA, Puligandla PS (2015) Pulmonary hyperten-
sion management in neonates. Semin Pediatr Surg
[Internet] 24(1):12–16. https://doi.org/10.1053/j.
sempedsurg.2014.11.003
APP Committee on fetus and newborn and section on anes-
thesiology and pain. Prevention and Management of
Procedural Pain in the Neonate: An Update. Pediatrics
[Internet] (2016 Feb) 137(2):e20154271–e20154271.
Available from: http://pediatrics.aappublications.org/
cgi/doi/10.1542/peds.2015-­4271
Penman G, Tavener K, Hickey A (2017) Neonatal feeding:
care and outcomes following gastrointestinal surgery.
Infant [Internet] 13(2):61–64. Available from: http://
search.ebscohost.com/login.aspx?direct=true&db=rzh
&AN=122012659&amp%0Alang=fr&site=ehost-­live
Petroze RT, Puligandla PS (2019) Preoperative cardio-
pulmonary evaluation in specific neonatal surgery.
A. Muntean et al.
Semin Pediatr Surg [Internet] 28(1):3–10. https://doi.
org/10.1053/j.sempedsurg.2019.01.002
Pitts S (2013) Retrospective analysis of a pediatric vas-
cular access program and clinical outcomes. JAVA – J
Assoc Vasc Access [Internet] 18(2):114–120. https://
doi.org/10.1016/j.java.2013.04.001
Porter ML, Dennis BL (2002) Hyperbilirubinemia in the
term newborn. Am Fam Physician 65(4):599–606
Puligandla PS, Grabowski J, Austin M, Hedrick H,
Renaud E, Arnold M et  al (2015) Management of
congenital diaphragmatic hernia: a systematic review
from the APSA outcomes and evidence based practice
committee. J Pediatr Surg [Internet] 50(11):1958–
1970. https://doi.org/10.1016/j.jpedsurg.2015.09.010
Raboei EH (2008) The role of the pediatric surgeon in the
perinatal multidisciplinary team. Eur J Pediatr Surg
18(5):313–317
Ramasauskaite D (2018) Management of pregnancy and
delivery in prenatally diagnosed congenital anomalies.
Congenit Anomalies – From Embryo to Neonate
Rodrigues AF, Van Mourik IDM, Sharif K, Barron DJ, De
Giovanni JV, Bennett J et  al (2006) Management of
end-stage central venous access in children referred
for possible small bowel transplantation. J Pediatr
Gastroenterol Nutr 42(4):427–433
Ruano R (2020) Prenatal regenerative fetoscopic inter-
ventions for congenital anomalies. BMJ 370
Sauer PJJ, Dane HJ, Visser HKA (1984) New standards
for neutral thermal environment of healthy very low
birthweight infants in week one of life. Arch Dis Child
59(1):18–22
Schindler E, Kowald B, Suess H, Niehaus-Borquez
B, Tausch B, Brecher A (2005) Catheterization of
the radial or brachial artery in neonates and infants.
Paediatr Anaesth 15(8):677–682
Shaffer SE, Norman ME (1989) Renal function
and renal failure in the newborn. Clin Perinatol
16(1):199–218
Shankar KR, Abernethy LJ, Das KSV, Roche CJ, Pizer
BL, Lloyd DA et  al (2002) Magnetic resonance
venography in assessing venous patency after multiple
venous catheters. J Pediatr Surg 37(2):175–179
Silverman WA, Sinclair JC (1966 Jan) Temperature regu-
lation in the Newborn Infant. N Engl J Med [Internet]
274(3):146–8. Available from: http://www.nejm.org/
doi/abs/10.1056/NEJM196601202740306
Sinkar P, Iyer S, Kallathiyan K (2020) Prenatal serum
screening  – a summary of our experience with high
risk reporting. J Matern Neonatal Med [Internet]
0(0):1–6. https://doi.org/10.1080/14767058.2020.177
4872
Stanley CA, Rozance PJ, Thornton PS, De Leon DD,
Harris D, Haymond MW, et al (2015 Jun) Re-evaluating
“transitional neonatal hypoglycemia”: mechanism and
implications for management. J Pediatr [Internet]
166(6):1520–1525.e1. Available from https://linking-
hub.elsevier.com/retrieve/pii/S0022347615001651
Statter MB (1992 Aug) Fluids and electrolytes in infants
and children. Semin Pediatr Surg [Internet] 1(3):
4  Pre-operative Management and Vascular Access
208–11. Available from http://www.ncbi.nlm.nih.gov/
pubmed/1345489
Sweet CB, Grayson S, Polak M (2013 Sep)
Management strategies for neonatal hypoglyce-
mia. J Pediatr Pharmacol Ther [Internet] 18(3):
199–208. Available from http://www.jppt.org/doi/
abs/10.5863/1551-­6776-­18.3.199
Sweet DG, Carnielli V, Greisen G, Hallman M, Ozek E,
Te Pas A et al (2019) European consensus guidelines
on the management of respiratory distress syndrome –
2019 update. Neonatology 115(4):432–450
Takayama JI, Teng W, Uyemoto J, Newman TB, Pantell
RH (2000) Body temperature of newborns: what is
normal? Clin Pediatr (Phila) 39(9):503–510
Tennant PW, Pearce MS, Bythell M, Rankin J (2010)
20-Year survival of children born with congenital
anomalies: a population-based study. Lancet [Internet]
375(9715):649–656. https://doi.org/10.1016/
S0140-­6736(09)61922-­X
Thornton PS, Stanley CA, De Leon DD, Harris
D, Haymond MW, Hussain K et  al (2015)
Recommendations from the pediatric endocrine soci-
ety for evaluation and management of persistent hypo-
glycemia in neonates, infants, and children. J Pediatr
[Internet] 167(2):238–245. https://doi.org/10.1016/j.
jpeds.2015.03.057
Ullman AJ, Marsh N, Mihala G, Cooke M, Rickard CM
(2015) Complications of central venous access devices:
a systematic review. Pediatrics 136(5):e1331–e1344
53
Ullman AJ, Bernstein SJ, Brown E, Aiyagari R, Doellman
D, Faustino EVS, et al (2020 Jun) The Michigan appro-
priateness guide for intravenous catheters in pediatrics:
miniMAGIC.  Pediatrics [Internet] 145(Supplement
3):S269–S284. Available from: http://pediatrics.aap-
publications.org/lookup/doi/10.1542/peds.2019-­3474I
Veldman A, Fischer D, Nold MF, Wong FY (2010)
Disseminated intravascular coagulation in term
and preterm neonates. Semin Thromb Hemost
36(4):419–428
Wataganara T, Grunebaum A, Chervenak F, Wielgos M
(2017) Delivery modes in case of fetal malformations.
J Perinat Med 45(3):273–279
Wheeler R (2006) Gillick or Fraser? A plea for consistency
over competence in children. Br Med J 332(7545):807
Wild KT, Rintoul N, Kattan J, Gray B, Engle RW, Keene
S, et  al (2020 May) Extracorporeal Life Support
Organization (ELSO): guidelines for neonatal respi-
ratory failure. ASAIO J [Internet] 66(5):463–70.
Available from https://journals.lww.com/10.1097/
MAT.0000000000001153
Wolf J, Curtis N, Worth LJ, Flynn PM (2013) Central line-­
associated bloodstream infection in children: an update
on treatment. Pediatr Infect Dis J 32(8):905–910
World Health Organization (1997) Thermal protection of
the newborn: a practical guide [Internet]. Available
from: https://www.who.int/maternal_child_adolescent/
documents/ws42097th/en/
 Anaesthesia and Analgesia
Coilin Collins Smyth and Suzanne Crowe
5.1 Introduction
Over the past 70  years or so, the provision of
anaesthesia for the child requiring surgery has
developed from being a relatively haphazard
affair to achieving the status of a recognized sub-
specialty. The improved outcomes seen follow-
ing surgery, where even very young and ill
infants and children are concerned, have been
due in no small part to advances in anaesthetic
management. Equally important has been an
increased appreciation of the need for an effi-
cient smooth-­ working team. The success of
major paediatric surgery depends on the maxi-
mum cooperation between surgeons, anaesthesi-
ologists, paediatricians, nursing and paramedical
personnel. It is therefore appropriate that every-
one involved in the care of the hospitalized child,
whether working inside or outside the operating
theatre, should be familiar with the basic tech-
niques used in maintaining a favourable physio-
We would like to acknowledge Dr. Declan Warde for the
materials we used from his previous chapter while writing
this new and updated version.
C. C. Smyth
Our Ladys Children’s Hospital, Crumlin,
Dublin, Ireland
S. Crowe (*)
Children’s Health Ireland, Crumlin, Dublin,
Republic of Ireland
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_5
5
logic milieu in the face of surgical intrusion,
while at the same time, ensuring adequate
anaesthesia.
5.2 Pre-operative Evaluation
and Preparation
The cornerstones of pre-operative anaesthetic
management are a detailed knowledge of the
child’s personal and family history combined
with a physical examination. Consideration must
also be given to the planned specific surgical pro-
cedure and its implications in terms of potential
blood loss, physiological disruption and the
child’s subsequent monitoring and post-operative
care requirements.
5.2.1 History
Particular attention should be paid to a history of
pre-maturity, respiratory disease, e.g. broncho-
pulmonary dysplasia or asthma, congenital heart
disease, any suggestion of neuromuscular disease
(patients with conditions like Duchenne’s muscu-
lar dystrophy may exhibit severe adverse reac-
tions following exposure to certain anaesthetic
agents), recent upper respiratory tract infection or
vaccination (see below), and any family history
of anaesthesia-related problems. Allergies and
current or recent drug therapy should be recorded.
55
56 C. C. Smyth and S. Crowe

5.2.2 Physical Examination
The anaesthetist should make a brief appraisal of
the child’s overall condition and follow this with
a careful physical examination, paying particular
attention to the respiratory and cardiovascular
systems. Airway anatomy should be carefully
assessed in order to identify potential difficulties
with endotracheal intubation prospectively.
operative assessment. This can range from dis-
charge home on the day of surgery to admission
to the critical care ward. This must be discussed
with the parents, as social factors can play a role
in the decisions along with clinical factors.
5.3 Operating Theatre
and Anaesthetic Equipment

The primary objectives of anaesthesia are the

5.2.3 Laboratory Investigations
Most older children presenting for minor surgery do
not require any pre-operative laboratory workup.
Minimum laboratory data required for infants and
children undergoing major surgery include full
blood count, blood urea and serum electrolytes,
blood glucose and calcium, coagulation profile and
urine specific gravity. Additional investigations may
be required in specific circumstances.
provision of sleep, analgesia, life support, inten-
sive surveillance and appropriate operating con-
ditions for the patient (irrespective of age)
requiring surgery. In order for these to be
achieved, it is imperative that both operating the-
atre environmental conditions and anaesthetic
equipment be appropriate for infants and chil-
dren. Appropriate measures should be taken to
minimize the risk of heat loss, especially where
surgery on infants is concerned.

5.2.4 Premedication 5.3.1 Breathing Systems

The decision to premedicate a child should be An appropriate anaesthetic circuit for use in
infants and children needs to be light, have mini-
made in the pre-operative assessment. Risk fac-
mal flow resistance and dead space, allow for
tors for poor behavioural compliance include age
warming and humidifying of inspired gases and
>8 months and <4 years, brief pre-operative time
and baseline temperament. Recent positive or be adaptable to spontaneous, assisted or con-
trolled ventilation. The most widely used system
negative experiences should also be investigated.
continues to be the T-piece, designed by Philip
A discussion with the caregiver is vital to ascer-
tain the need for premedication. Sedative pre-Ayre and later modified by Rees. Connectors and
medication is not used in neonates or in othertubes should also offer minimal flow resistance
and dead space. Knowledge of the probable
situations where it may cause increased risk, e.g.
children with airway compromise, sleep apnoea diameter and length of the endotracheal tube
appropriate for any given child is essential. Use
or obesity. Its use in other children appears to
of an endotracheal tube (ET) of too large a diam-
have declined in recent years—this may be due in
eter may result in tracheal wall damage, while
part to the fact that increased parental involve-
excess length leads to endobronchial intubation.
ment in perioperative care serves to reduce the
incidence of separation anxiety. Midazolam andThe presence of a cuff limits the diameter of the
tube that can be used, with consequent increased
diazepam (for older children) are the most com-
monly used premedicant drugs. resistance to airflow. This limitation is becoming
less significant with the advent of the micro-cuff
ET. Face masks are generally used for only brief
5.2.5 Post-operative Planning periods in infants and children, but should pro-
vide a good fit and have a low dead space. Use of
The post-operative care of each child must be the laryngeal mask airway (LMA) has become a
considered and planned for during the pre-­ standard practice in day care surgery.
5  Anaesthesia and Analgesia
5.3.2 Laryngoscopes
The infant’s airway has anatomical differences to
the older child, which makes endotracheal intu-
bation more challenging. These include a larger
relative tongue size and a more floppy epiglottis.
Because of these anatomical variants, most
anaesthetists prefer to use a laryngoscope with a
straight blade in this age group, lifting the epi-
glottis forward from behind to facilitate endotra-
cheal intubation. Standard curved blades,
available in various sizes, are appropriate for
older children. In recent years, video laryngo-
scopes have become available for regular use.
5.3.3 Ventilators
Most infants and children can be ventilated using
standard adult ventilators, provided the ventilator is
of low internal compliance and equipped with pae-
diatric breathing tubes. The ventilator should be
capable of delivering small tidal volumes and rapid
respiratory rates. It should also have an adjustable
inspiratory flow rate and inspiratory to expiratory
ratio so that peak airway pressure is kept as low as
possible. Ventilation should be delivered so as to
minimise risk of barotrauma or volutrauma. This
can be achieved via pressure control or volume
control. Tidal volume should be approximately
6–8 ml/kg with an increasing respiratory rate with
decreasing age. A suitable temperature-controlled
humidifier should be incorporated in the inspira-
tory side of the ventilator circuit. The ability to
deliver air and oxygen mixtures through the venti-
lator or the anaesthetic circuit should be available.
5.3.4 Monitoring Equipment
A complete range of monitoring equipment suit-
able for paediatric use is required.
5.4 Induction of Anaesthesia
The anaesthetic agents employed in infants and
children are identical to those used in adults. The
choice of induction technique depends on (a) the
57
age, size and physical status of the child, (b) the
relative hazard of regurgitation, and (c) the per-
sonal preference of the anaesthetist. Where the
older child is concerned, his or her preference
may also be taken into consideration. The intro-
duction of local anaesthetic creams has helped
reduce the “fear factor” for those children need-
ing intravenous (i.v.) cannulation prior to induc-
tion of anaesthesia, especially those requiring
multiple anaesthetics. Many are also helped by
having a parent remain with them until they are
asleep—this was almost unknown 30  years ago
but is now commonplace. Inhalational induction
is still preferred, especially in younger children,
by many paediatric anaesthetists, and is particu-
larly appropriate when difficulty with venous
access is anticipated, where a slower induction is
desired because of concern regarding loss of air-
way control, or when the child specifically
requests it.
5.4.1 Parental Presence
There has been a significant change in the advice
regarding parental presence. Previously, it was
unusual for a caregiver to be present for the dura-
tion of the induction of anaesthesia. In current
practice, the vast majority of paediatric cases are
induced with a caregiver present. The benefits
include decreased anxiety for the child. Most
caregivers also prefer to be present. A 2017 study
supports this (Sadeghi et  al. 2017). It is also
thought to decrease the need for premedication.
These benefits are dependent on the caregiver not
projecting their anxiety to the child. This needs to
be tactfully explained to the caregiver prior to
arrival in the theatre.
5.5 Intravenous Agents
For many years, sodium thiopentone was the
drug of choice for i.v. induction of anaesthesia in
patients of all ages. In most centres, it has now
been replaced by propofol, which provides an
equally rapid and smooth induction combined
with a more rapid and complete recovery, and
also has an antiemetic effect. It also has increased
58
depression of the laryngeal reflexes. Pain on
injection can be prevented in most instances by
the prior addition of lignocaine to the propofol
solution. Ketamine is associated with greater car-
diovascular stability than many other anaesthetic
drugs and is a potent analgesic. It has a slower
onset of action than propofol. It may be used with
beneficial effect when it is necessary to induce
anaesthesia in the shocked child.
5.6 Inhalational Agents
5.6.1 Halothane
For decades, halothane was the most popular
volatile anaesthetic for inhalational induction in
infants and young children. This is largely
because it is usually associated with a smooth
induction without irritant effects on the airway. In
adult practice, its use has been superseded by
newer agents to such an extent that it has become
difficult to obtain, much to the chagrin of many
paediatric anaesthetists who still consider that it
has a useful role. Its decreased popularity in adult
anaesthesia was related to its potential for hepatic
toxicity. Halothane has also been associated with
increased dysrhythmias and negative inotropy in
the paediatric population compared to sevoflu-
rane (Lerman 2002; Friesen et al. 2000).
5.6.2 Isoflurane
Inhalational induction of anaesthesia with isoflu-
rane is not as rapid or as smooth as with halo-
thane. Indeed, this agent has been shown to be
associated with a significant incidence of hypoxic
episodes during induction of anaesthesia in older
children. It has considerable potentiating effects
on non-depolarizing muscle relaxants, so that
lower doses of the latter can be used. It is an
excellent agent for maintenance of anaesthesia.
5.6.3 Enflurane
This agent is not widely used in paediatric anaes-
thesia because its irritant effects render it rela-
C. C. Smyth and S. Crowe
tively unsatisfactory for inhalational induction
while it offers few particular advantages during
the maintenance period.
5.6.4 Desflurane
Airway irritant effects also render desflurane
unsuitable for inhalational induction. However,
recovery times in infants are shorter than those
following other volatile anaesthetics. The agent
has been recommended for maintenance of
anaesthesia in the ex-premature infant prone to
apnoea and ventilatory depression. Desflurane is
also the most environmentally toxic agent.
5.6.5 Sevoflurane
In older children, induction time with sevoflurane
is shorter than with halothane. However, this
does not appear to be the case where infants are
concerned. The agent has been reported to cause
more respiratory depression than halothane in
infants and young children, but perhaps not to a
degree that is clinically significant. It has become
the most widely used volatile agent for inhala-
tional induction and is also popular for mainte-
nance of anaesthesia.
5.6.6 Nitrous Oxide
This gas does not provide adequate anaesthe-
sia when used alone with oxygen. It is most
often employed as a carrier that supplements
potent volatile anaesthetics, thereby reducing
the concentration required and minimizing
­cardiovascular depressant effects. One limita-
tion of its use is the fact that it is many times
more soluble in blood than nitrogen. As a result,
the inhalation and subsequent diffusion of the
gas causes an increase in the volume of compli-
ant spaces. It follows that the agent should not
be used in patients with congenital diaphrag-
matic hernia, lobar emphysema or bowel
obstruction. It is also important to bear in mind
the negative environmental impact of nitrous
oxide (Baum VC, 2007).
5  Anaesthesia and Analgesia
5.7 Neuromuscular Blocking
Agents
5.7.1 Succinylcholine
Because of the number of side-effects, including
bradycardia, hyperkalaemia and triggering of
malignant hyperpyrexia reactions associated with
this agent, its use has declined dramatically in
recent years. However, it remains pre-eminent in
rapidly providing optimum conditions for endo-
tracheal intubation, and therefore retains a lim-
ited place in emergency paediatric anaesthetic
practice. For elective surgery, intubation is now
more frequently facilitated by either deep inhala-
tional anaesthesia or use of a non-depolarizing
muscle relaxant.
5.7.2 Atracurium and Vecuronium
Atracurium and vecuronium were originally
introduced because their duration of action was
intermediate between that of succinylcholine and
older agents such as pancuronium, and because
they offered increased cardiovascular stability. In
addition, atracurium is attractive in that its metab-
olism is independent of hepatic and renal func-
tion (plasma cholinesterase). Because of their
pharmacokinetic profiles, both drugs are suitable
for use by continuous intravenous infusion,
although atracurium infusion requirements show
marked individual variation.
5.7.3 Mivacurium
Mivacurium is a short-acting, non-depolarizing,
neuromuscular agent that is rapidly hydrolysed
by plasma pseudocholinesterase. The duration of
block produced by the drug is more rapid in
younger paediatric patients. Satisfactory intubat-
ing conditions are not achieved as quickly as with
succinylcholine but serious side-effects occur
less frequently.
59
5.7.4 Pancuronium
Pancuronium is used in neonates and infants. It
has a slow onset of action. It causes an increase in
heart rate, which is why it is useful in the neonate
who is at increased risk of bradycardia and whose
cardiac output is more dependent on heart rate
than the older child or adult.
5.7.5 Rocuronium
Rocuronium has a dose-dependent onset time,
which can be as low as 45 s at a dose of 1 mg/kg.
It has an intermediate duration of action in most
children, although neuromuscular blockade may
be prolonged in young infants. It has become
very common in anaesthesia practice due to its
fast onset and immediate reversibility with
sugammadex.
5.8 Maintenance of Anaesthesia
The maintenance of anaesthesia and the quantifi-
cation of the depth of anaesthesia is vital to pre-
vent awareness and maintain haemodynamic
stability. This can be achieved via inhalation or
intravenous methods.
5.8.1 Inhalational Maintenance
of Anaesthesia
Because of the vulnerability of the infant’s respi-
ratory system, spontaneous respiration is not
used for long periods in the anaesthetized neo-
nate. Mechanical ventilation helps ensure ade-
quate gas exchange and also leaves the hands of
the anaesthetist free to perform other tasks.
Manual ventilation allows rapid detection of air-
way obstruction or disconnection, and is particu-
larly useful during thoracic surgery.
The most widely used agents for maintenance
of anaesthesia in the paediatric population are
60
sevoflurane isoflurane and desflurane, usually com-
bined with 50% oxygen in nitrous oxide.
Consideration should be given to the use of air and
oxygen mixtures in preterm neonates. Older chil-
dren may be allowed to breathe spontaneously for
longer periods; the use of mechanical ventilation is
often dictated by the nature of the surgical proce-
dure. In recent years, consideration has also been
given to the environmental impact of volatile
anaesthetic gases with nitrous oxide and desflurane
being the most detrimental. While the optimal
anaesthetic for the individual child is still para-
mount, some consideration can be given to the
environmental impact (Campbell and Pierce 2015).
5.8.2 Total Intravenous Anaesthesia
(TIVA)
The alternative to inhalational maintenance is
TIVA. This involves the continuous IV infusion of
an anaesthetic agent, most commonly propofol, to
maintain adequate depth of anaesthesia. Mathe-
matical models (Kataria) have been designed to
determine how much of each drug has to be
injected per minute to maintain the correct con-
centration in the intravascular compartment. Other
commonly co-administered drugs are remifent-
anil, fentanyl and dexmedetomidine. Logistical
considerations include an I.V. cannula that is func-
tioning well and is normally kept in clear view
during the procedure to prevent inadvertent dis-
connection/leaking. The benefits of TIVA include
a decreased risk of nausea and vomiting and
decreased airway responsiveness. It can also be
very useful to ensure adequate supply of anesthe-
sia when the airway is shared with the surgeon.
The use of an additional monitor of anaesthetic
depth (in addition to haemodynamic monitoring)
is advised when using TIVA. This is usually in the
form of a processed EEG like entropy or bispectral
index (BIS) (Gaynor and Ansermino 2016).
5.9 Reversal and Extubation
If a volatile agent or an intravenous agent has
been used for the maintenance of anaesthesia, it
should be discontinued shortly prior to the end of
C. C. Smyth and S. Crowe
surgery. When muscle relaxants have been used in
intubated patients, any residual relaxation must be
reversed. This is now most commonly achieved
via the use of sugammadex (2–4  mg/kg), which
effectively reverses both rocuronium and
vecuronium. Neostigmine (0.06 mg/kg) or edro-
phonium (1  mg/kg), combined with either atro-
pine (0.02–0.03  mg/kg) or glycopyrrolate
(0.01 mg/kg), can also be used. Controlled venti-
lation is continued with 80–100% oxygen in air
until spontaneous respiration has returned. Infants
should not be extubated until fully awake and
breathing adequately. In most cases, reversal of
neuromuscular blockade and resumption of spon-
taneous respiration occurs rapidly. If difficulty is
encountered, this may be due to hypothermia, aci-
dosis, hypocalcaemia, or the fact that an incre-
mental dose of relaxant has been given too close
to the end of surgery and not adequately reversed.
5.10 Recovery from Anaesthesia
Initial recovery from anaesthesia and surgery
should be in a fully equipped recovery area with a
one-to-one ratio of personnel trained in paediatric
nursing. Monitoring of vital signs, adequacy of
protective airway reflexes, and correct positioning
to prevent airway obstruction, regurgitation and
aspiration are the priorities during the immediate
post-operative period. The recovery room nurse
also monitors the wound site for bleeding, checks
the security of dressings and the adequacy of pain
relief. The main factors influencing the rate of
recovery in children include the use of premedi-
cant drugs, the induction and maintenance tech-
niques, the age of the child, and the duration of
surgery. Parental involvement in the early recovery
phase is now encouraged in many institutions.
5.11 Post-operative Care
Most children may be offered a drink or light
snack as soon as they have recovered conscious-
ness. Delays in oral intake can be appropriate fol-
lowing endotracheal intubation, some dental or
oral procedures, and when local anaesthetic drugs
have been used on the upper or lower airway. In
5  Anaesthesia and Analgesia 61

addition to post-discharge guidance provided by 5.13 Cardiovascular Monitoring

the surgical team, parents or guardians should
also be given both verbal and written instructions 5.13.1 Precordial and Oesophageal
regarding post-operative pain management and Stethoscope
what to do if problems arise.
Although its use in paediatric anaesthetic prac-
tice has declined, the stethoscope is particularly
5.11.1 Post-operative Intensive Care valuable, allowing continuous monitoring of
Management heart and breath sounds cheaply and simply. In
the neonate, the intensity of the heart sounds var-
Paediatric critical care units are specialized facil- ies with the stroke volume so that an indication of
ities within paediatric hospitals to facilitate high cardiac output is provided.
levels of monitoring and organ support to criti-
cally unwell children. This level of care can be
required in the pre-operative and/or post-­ 5.13.2 ECG
operative period. It is usually an elective admis-
sion when this level of care is predicted due to As myocardial ischaemia is uncommon in chil-
specific surgery and/or the pre-operative comor- dren, the principal function of the ECG is to
bidities that could be exacerbated as a result of monitor heart rate and detect arrhythmias, espe-
surgery. Unplanned post-operative admissions cially bradycardia. Primary arrhythmias are
also occur where an unexpected event intra-­ uncommon except in those with congenital heart
operatively leads to the child requiring critical disease, but causes of secondary arrhythmias
care in the post-operative period. The support include hypoxia, hypercarbia and surgical stimu-
required includes, but is not exclusive to, mechan- lation, e.g. strabismus surgery. ECG monitoring
ical ventilation, inotropic or vasopressor support, may also detect inadvertent i.v. injection of local
renal replacement therapy and parenteral nutri- anaesthetics.
tion. Clear planning and communication is
required between the anaesthesiologist, the surgi-
cal team and the critical care team. 5.13.3 Blood Pressure

Routine non-invasive monitoring of blood pres-

5.12 Monitoring
The clinical condition of the anaesthetized child
can deteriorate more rapidly and with less warn-
ing than that of patients in any other age group. It
follows that careful and continuous monitoring is
essential. No piece of monitoring equipment will
adequately replace the attentive anaesthetist
(Moscuzza, 2003). However, a number of devices
providing helpful information that cannot be
ascertained by clinical means alone are available.
The monitoring employed in any particular case
depends on the physical status of the child and
the surgical procedure to be undertaken.
sure during anaesthesia and surgery is carried out
with automated devices using oscillometry. The
appropriate cuff size must be used in order to
obtain accurate measurements. Direct intra-­
arterial monitoring is the most accurate measure-
ment of blood pressure and provides a “beat to
beat” assessment. Its use is generally restricted to
very ill children or those undergoing major sur-
gery. All the proximal and distal arteries of the
arms and legs may be used. While it is generally
a safe procedure, risks include ischaemia, infec-
tion and haematoma. Where possible, the radial
artery is the optimal choice as the distal arm has
dual arterial supply via the ulnar and radial.
62
5.13.4 Central Venous Pressure
Central venous pressure monitoring is useful in
infants and children undergoing major surgery
with anticipated large fluid shifts, if significant
blood loss (and replacement) is expected, and
during surgery for congenital heart disease. The
right internal jugular vein is usually the simplest
to cannulate. Monitoring of left atrial pressure
and pulmonary capillary wedge pressure is rarely
indicated in infants or children.
5.14 Respiratory Monitoring
5.14.1 Pulse Oximetry
Hypoxia is the most common critical incident in
paediatric anaesthesia. As detection of cyanosis
in infants and young children is difficult, the rou-
tine use of pulse oximetry is now mandatory.
Thermal injury and pressure necrosis have been
reported when sensor probes have been applied
too tightly.
5.14.2 Capnography
The measurement of pCO2 in inspired and expired
gases is also mandatory. Capnography not only is
a monitor of the adequacy of ventilation but also
gives warning of disruption in gas supply, inade-
quate fresh gas flow and oesophageal intubation.
It also gives you an indication of perfusion to the
lungs.
5.14.3 Ventilator Pressure/Volume
Monitoring
All ventilators will monitor for age-specific,
breath-by-breath volumes and pressures. This
decreases the risk of barotrauma and volutrauma
as well as ensuring adequate ventilation and oxy-
genation are being delivered to the child. The
ventilator will also monitor for leaks in the cir-
cuit, gas flow adequacy and delivery of anaes-
thetic gases.
C. C. Smyth and S. Crowe
5.15 Temperature Monitoring
Monitoring temperature is important in paediat-
ric anaesthesia because of the increased risk of
both hypo- and hyperthermia. Heat loss in the
neonate and infant is of paramount importance.
Common sites for temperature probes used pre-­
operatively include the pharynx, oesophagus and
rectum. Temperature should be measured prior to
induction of anaesthesia and normothermia
maintained throughout, unless hypothermia is
indicated (off pump cardiac surgery).
5.16 Neuromuscular Blockade
Monitoring
Monitoring neuromuscular blockade using a
peripheral nerve stimulator is routine practice
when non-depolarizing muscle relaxants have
been administered.
5.17 Other Monitoring
The release of new, novel monitoring devices is
not infrequent. However, few are shown to
improve outcomes and therefore do not become
part of routine anesthesia monitoring. Processed
EEG monitoring has become common during
TIVA to monitor depth of anaesthesia in the
absence of end-tidal gas monitoring. Near Infra-­
red Spectroscopy (NIRS) attempts to measure
cerebral oxygen delivery. It is becoming increas-
ingly used in major paediatric surgeries, most
commonly when cardiac bypass is being used.
5.18 Fluid Balance
Healthy children undergoing minor operations
can reasonably be expected to tolerate oral fluids
a short time after completion of surgery and do
not require intraoperative i.v. fluids. The goal of
intraoperative fluid management in those who are
dehydrated pre-operatively or who are undergo-
ing major surgery is to sustain homeostasis by
providing the appropriate amount of parenteral
5  Anaesthesia and Analgesia
fluid to maintain adequate intravascular volume,
cardiac output, and ultimately, oxygen delivery to
tissues at a time when normal physiological func-
tions are altered by surgical stress and anaesthetic
agents. The composition of the administered fluid
will vary according to the maturity of the child
and pre-operative electrolyte and glucose levels.
Because of the problems associated with hyper-
glycaemic states in infancy, care should be taken
with the use of 10% dextrose infusions. Blood
and fluid loss can be extensive and very difficult
to measure during neonatal surgery. The former
is best estimated by the use of small volume suc-
tion traps, by weighing small numbers of surgical
swabs before they dry out, and by serial haemato-
crit measurements. During lengthy surgery,
serum electrolytes and blood glucose should be
measured at regular intervals. Urine output may
be monitored by the use of adhesive collecting
bags or bladder catheterization. Estimated third
space loss may be replaced by continuous admin-
istration of lactated Ringer’s solution at 3–5 ml/
kg/h. The adequacy of volume replacement can
be assessed by monitoring blood pressure, central
venous pressure, peripheral circulatory state and
urine output.
5.19 Special Considerations
for the Premature Infant
Congenital defects occur more commonly in pre-
term infants, so surgery is frequently required.
Organs and enzyme systems are very immature
in this cohort and meticulous attention to detail
during anaesthetic and surgical management is
imperative if survival rates are to be high. The
large body surface area and lack of subcutaneous
fat make maintenance of body temperature very
difficult, so a high neutral thermal environment is
essential. Respiratory fatigue occurs very easily
and may be exacerbated by residual lung damage
following mechanical ventilation, persistent fetal
circulation and oxygen dependency. The response
to exogenous vitamin K is less satisfactory than
in term infants and there is an increased risk of
bleeding. In addition, anaemia is common
because of reduced erythropoiesis, a short eryth-
63
rocyte life span and iatrogenic causes such as fre-
quent blood sampling. Fluid and electrolyte
management can be difficult—insensitive losses
are high and hypoglycaemia and hypocalcaemia
occur easily, while renal function and the ability
of the cardiovascular system to tolerate fluid
loads are reduced.
5.20 Anaesthesia for Specific
Surgical Conditions
5.20.1 Oesophageal Atresia
Once a diagnosis of oesophageal atresia (with or
without fistula) has been made, the blind upper
pouch should be continuously aspirated using a
Replogle or similar tube. In general, the opera-
tion may be safely delayed pending improvement
of any aspiration pneumonia that has developed.
Pre-operative echocardiography is recommended
to diagnose any anomaly that could affect anes-
thesia and surgical management. Pre-thoracotomy
bronchoscopy is practised in some centres and
may influence subsequent management.
Anaesthesia is similar to that for other neonatal
procedures, but special care must be taken with
positioning of the endotracheal tube, the tip of
which should be located above the carina but
below any fistula present. Surgical retraction dur-
ing the operation may compromise either respira-
tory or cardiac function, so that close monitoring
is essential. If serious contamination has not
occurred and unless the surgeon deems the anas-
tomosis to be especially tight, extubation is usu-
ally possible shortly after the conclusion of
surgery (Al-Rawi and Booker 2007).
5.20.2 Congenital Diaphragmatic
Hernia
This condition was formerly regarded as one of
the great emergencies of paediatric surgical prac-
tice, but it is now considered that the timing of
repair should be based on the optimization of
clinical parameters rather than a specific time
period post-delivery. The key features are lung
64 C. C. Smyth and S. Crowe

hypoplasia and pulmonary hypertension. Pre-­ 5.20.4 Exomphalos and Gastroschisis

operative ventilatory and haemodynamic support
along with correction of metabolic disturbance Anaesthetic concerns include heat and fluid loss
are almost invariably required. Inhaled nitric from the exposed bowel and the fact that primary
oxide, high frequency oscillation ventilation, liq- closure of the abdominal wall defect may push the
uid ventilation and extracorporeal membrane diaphragm cephalad, thus compromising respira-
oxygenation may also be used. Positive pressure tory function. Special care must be taken to keep
ventilation using bag and mask should be avoided heat loss to a minimum. Fluid requirements are
prior to endotracheal intubation, as expansion of much greater than in normal neonates. To maintain
the viscera contained within the hernia will cause plasma oncotic pressure, at least 25% of fluid
further lung compression. Nitrous oxide should intake should be given as colloid. The extent of
be avoided for the same reason. 14% have a car- respiratory compromise can assist the anaesthetist
diac anomaly so echocardiography is essential. A in advising the surgeon whether or not primary
reasonable anaesthetic technique includes con- closure is feasible (Puri and Hollworth, 2006). A
trolled ventilation with fentanyl 0.01–0.02  mg/ proportion of infants, especially after repair of gas-
kg, an intermediate-acting muscle relaxant and troschisis, require postoperative ventilation. The
100% oxygen or oxygen in air as required. Great introduction of staged closure of gastroschisis
caution should be exercised in the use of volatile using preformed silos has simplified anaesthetic
anaesthetic agents. Airway pressures should be and paediatric intensive care unit management.
kept as low as possible. Should advanced ventila-
tory techniques such as the use high frequency
oscillation be required in order to achieve pre-­ 5.20.5 Myelomeningocele, Shunt
operative stabilisation, these may be safely con- (and Revision Shunt) for Spina
tinued during surgery. Most infants will require Bifida
mechanical ventilation in the post-operative
period. Surgical manipulation can cause periods Surgery for myelomeningocele is carried out
of profound cardiovascular instability requiring with the infant in the prone position and the chest
invasive monitoring and support. and pelvis should be supported with pads so that
the abdomen remains free from external pressure.
If the defect is large, heat and fluid loss during
5.20.3 Intestinal Obstruction surgery can pose problems and should be moni-
tored as closely as possible. Endotracheal intuba-
The various forms of neonatal intestinal obstruc- tion may be difficult in the presence of
tion account for approximately 35% of all surgi- hydrocephalus. Children who have had repeated
cal procedures in the newborn. The major shunt surgery and regular bladder catheterization
anaesthetic problems are those of fluid and elec- may develop latex allergy and a latex-free anaes-
trolyte imbalance (which must be corrected pre-­ thesia technique should be considered.
operatively), abdominal distension (causing
diaphragmatic splinting) and the risk of regurgi-
tation and aspiration of gastric contents into the 5.20.6 Muscle Biopsy
lungs. Following decompression of the stomach,
a rapid-sequence induction incorporating pre-­ This operation represents an example of a proce-
oxygenation, propofol and a fast acting muscle dure that, although relatively minor and innocu-
relaxant with gentle cricoid pressure is advised. ous from a surgical viewpoint, may pose
Anaesthesia is then continued in the usual way. considerable problems for the anaesthetist. It is
The same principles apply to the management of most frequently performed to either confirm or
those older infants with a diagnosis of intussus- exclude a diagnosis of a possible neuromuscular
ception who require operative reduction. disease. Children presenting for muscle biopsy
5  Anaesthesia and Analgesia
may have decreased respiratory and cardiac
reserve, and be at increased risk of perioperative
aspiration. They may also have metabolic
derangements and can be prone to developing
hypoglycaemia. Use of succinylcholine is abso-
lutely contraindicated, as it has been associated
with lethal hyperkalaemia in this patient popula-
tion. There may also be increased sensitivity to
nondepolarizing muscle relaxants. Many paediat-
ric anaesthetists also prefer to avoid volatile
agents due to the potential for malignant hyper-
pyrexia, thus necessitating the use of a continu-
ous infusion anaesthetic technique, most
commonly with propofol.
5.20.7 Herniotomy
in the Ex-Premature Infant
Improved survival rates in premature and low
birth weight infants have led to increased num-
bers of presentations for inguinal hernia repair.
While the surgical procedure may be relatively
straightforward, these infants represent a consid-
erable challenge for the anaesthetist. They must
be managed by anaesthetists and surgeons with
adequate training and ongoing experience in hos-
pitals with appropriate facilities and personnel.
Ex-premature infants up to 60  weeks post-­
conceptual age are at risk of life-threatening
apnoea after anaesthesia and surgery. They
should have respiratory monitoring for at least
12 h post-operatively and should not be managed
as day-cases. Intravenous caffeine 5 mg/kg given
i.v. at induction has been extrapolated from man-
agement of preterm neonatal apnoea in neonatal
ICU, but direct evidence is lacking. Respiratory
monitoring is still required. Regional techniques
also reduce, but do not eliminate, the risk of post-
operative apnoea and may require supplementary
sedation or light general anaesthesia.
5.21 Post-operative Analgesia
in Children
There have been significant improvements in
pain relief following surgery in children. In the
past, this was usually achieved by intramuscular
65
injection of narcotics—many children suffered in
silence, believing “the cure to be worse than the
disease.” In current practice, more effective and
humane routes of administration along with a
wider range of types of analgesics are employed.
Optimal pain relief must be sought for every pae-
diatric surgery. Prevention of pain whenever pos-
sible using multi-modal analgesia can be adapted
for day-cases, major surgery, the critically ill
child and the very young. Most paediatric acute
pain services use techniques of co-analgesia
based on five classes of drugs, namely local
anaesthetics, opioids, non-steroidal anti-infl
ammatory drugs (NSAIDs), paracetamol (acet-
aminophen) and adjuncts such as alpha antago-
nists (clonidine). For many day-case procedures,
opioids can and should be omitted altogether
because combinations of the other three classes
usually provide excellent pain control. Limiting
opioid use will decrease nausea, decrease consti-
pation and decrease dependence. In children’s
hospitals or other centres where significant num-
bers of children undergo anaesthesia and surgery,
the establishment of a dedicated paediatric pain
service is the desirable standard of care.
5.21.1 Local and Regional
Anaesthesia
Regional anaesthesia is becoming an increasing
part of routine anaesthesia. In children, it is pre-
dominantly used as an adjunct to general anes-
thesia. In this form, the benefits include a
decreased requirement for inhalational or intra-
venous anaesthesia. It is associated with a
decreased acute stress response to surgery. In the
postoperative period, it offers effective non-­
sedative analgesia. The incidence of major com-
plications is extremely low even when central
blocks, e.g. epidural analgesia, are used. In its
simplest form, local anaesthetic can be infiltrated
directly to the surgical site, Alternatively, the
anaesthetist may perform a targeted nerve block
or a fascial plane block. An example of a fascial
plane block commonly used is a fascia iliaca
block. The most common central regional tech-
nique performed in children is the caudal epi-
dural injection (Llewellyn and Moriarty, 2007).
66
Numerous other local, regional and central anaes-
thetic blocks, often performed with ultrasound
guidance, are used on an increasingly regular
basis for virtually all types of surgery, including
open cardiac surgery and in many instances their
efficacy in the postoperative period can be pro-
longed for as long as necessary through the use of
continuous infusion techniques.
Until recently, the most widely used local
anaesthetic agent for regional blockade was race-
mic bupivacaine. This has now largely been
replaced by either ropivacaine or levobupiva-
caine, both of which appear to offer greater
safety. However, it remains essential to adhere to
published maximum dosage guidelines (2 mg/kg
in infants, 2.5 mg/kg in children for single bolus
injection with either drug). Use of some adjunc-
tive agents, e.g. clonidine, ketamine with single
dose or continuous epidural blockade, is increas-
ingly popular as both the effectiveness and dura-
tion of blockade appear to be enhanced. Recent
studies have demonstrated IV dexamethasone
can increase the duration of regional analgesia.
5.21.2 Opioids
Morphine remains the most widely used opioid
for intra- and post-operative analgesia in infants
and children. Bolus injections of 0.1–0.2 mg/kg
or infusions between 0.01 and 0.03 mg/kg/h pro-
vide adequate analgesia with an acceptable level
of side-effects when administered with an appro-
priate level of monitoring. It should be noted that
morphine’s elimination half-life is prolonged in
the newborn when compared with older infants
and children. Patients in this age-group are also
more susceptible to the drug’s respiratory depres-
sant effects, Patient-controlled analgesia (PCA)
is now widely used in children as young as
5 years and compares favourably with continuous
infusion. Nurse-controlled analgesia (NCA) is
useful in younger children or in those without the
physical or mental capacity to use PCA success-
fully. Oral, sublingual, transdermal, intranasal
and rectal routes of opioid administration have all
been described and may have a role to play in
C. C. Smyth and S. Crowe
specific cases. Use of the intramuscular route is
no longer considered appropriate. Tramadol,
oxycodone and pethidine may have some appli-
cability as alternatives to morphine in the periop-
erative period. Fentanyl, alfentanil and
remifentanil may have a role in intensive care
practice after major surgery. A 2019 study sug-
gested that opioid prescriptions are not required
for many paediatric surgeries and this leads to
increased opioid use and misuse (Harbaugh et al.
2019).
5.21.3 Non-steroidal Anti-­
inflammatory Drugs (NSAIDs)
These drugs are important in the prevention and
treatment of mild to moderate pain in children.
They are highly effective when used in combina-
tion with local anaesthetic or regional nerve
blocks. They may also be used in combination
with opioids, leading to a significant “opioid-­
sparing” effect, which not only reduces the opioid
dose requirement but also lessens the incidence of
opioid-related side-effects, e.g. ileus, urinary
retention and dependence. NSAIDs, e.g. diclofe-
nac in combination with paracetamol, produce
better analgesia than either alone. They should be
avoided in infants aged less than 6 months, chil-
dren with aspirin or NSAID allergy, those with
dehydration or hypovolaemia, children with renal
or hepatic failure, coagulation disorders, peptic
ulcer disease, or in those who are at significant risk
of haemorrhage. Concurrent administration of
NSAIDs with anticoagulants, steroids or nephro-
toxic agents is not recommended. NSAIDs may
provoke bronchospasm in some asthmatic patients.
However, it is useful to check for past exposure to
these drugs as many asthmatic children can toler-
ate them with no adverse effects. There have been
suggestions that they impair bone healing. The
beneficial effects of their short-term use in most
children undergoing orthopaedic surgery probably
outweigh this possible risk but caution is recom-
mended following some major orthopaedic sur-
gery, e.g. spinal fusion, limb-lengthening
procedures (Nuelle, 2020).
5  Anaesthesia and Analgesia
5.21.4 Paracetamol (Acetaminophen)
Paracetamol has both analgesic and antipyretic
effects. Its analgesic potency is relatively low. On
its own, it can be used to treat most mild and
some moderate pain. In combination with
NSAIDs or a mild opioid such as codeine, it can
be used to treat or prevent most moderate pain.
Oral formulations are widely available.
Absorption from the rectum is slow and incom-
plete, except in neonates. It is important to realise
that the time to peak analgesia even after i.v.
administration is 1–2  h. In younger infants and
sick children, considerable downward dose
adjustments are needed.
5.22 Day-Case Anaesthesia
and Surgery
In many units, up to 75% of surgery in children is
carried out on a day-case basis. Day-case surgery
benefits the child, the caregiver and the health-
care system. The same standards of care apply
whether surgery is carried out on in-patients or
day-cases. Staff should have been trained in the
care of children, the environment should be child-­
friendly and child-safe, and there should be free
parental access to the conscious child. While
many operations can be carried out as day-cases,
a number of procedures are not suitable. These
include any operation following which there is a
significant risk of postoperative haemorrhage,
other urgent complications or where there is a
likelihood of postoperative pain requiring sophis-
ticated control. Other factors that need to be con-
sidered include the age and maturity of the child,
his or her overall medical condition, the presence
of anaesthetic risk factors, and the family’s social
circumstances.
5.22.1 Preparation of Child
and Parents
The child and caregiver should be provided with
a clear verbal explanation (age appropriate) of
what will happen on the day of surgery and this
should be reinforced by written information and
67
guidance regarding pre-operative fasting. Pre-­
admission programmes, e.g. “Saturday Clubs”
may be helpful.
5.22.2 Premedication
Pre-operative anxiety is associated with adverse
outcomes. These include an increased analgesic/
anaesthetic requirement during the surgery. Post-­
operatively, there is an increased risk of delirium
and behavioural changes. As discussed above,
premedication with anxiolytics can combat this
but are also associated with slower recovery time
post-operatively. In many centres, the majority of
children presenting for day-case surgery are not
routinely premedicated. Non pharmacological
methods such as play therapy are adequate.
Midazolam has for some years been the drug of
choice but it has recently been suggested that
clonidine may be preferable. The need for pre-
medication should be assessed during the pre-­
operative assessment.
5.22.3 Anaesthetic Technique
The anaesthetic technique employed should be as
simple and non-invasive as possible. Both inhala-
tional and i.v. induction and maintenance tech-
niques may be used.
It is preferable to maintain spontaneous respi-
ration during the maintenance period, although
use of non-depolarizing muscle relaxants is not a
contraindication to day-case surgery. If possible,
airway control should be achieved with either a
face mask or laryngeal mask airway, with endo-
tracheal tubes being reserved for cases where
they are specifically required, e.g. upper gastroin-
testinal endoscopy. Endotracheal intubation is
not, however, a contraindication to day-case
surgery.
5.22.4 Analgesia for Day-Cases
Good pain control is critical to the success of
paediatric day-case surgery. Overall, a multi-
modal, non-opioid-based analgesic plan should
68
be made for the immediate post-operative period
and the post-discharge period. Opioids are asso-
ciated with a higher incidence of post-operative
nausea and vomiting (PONV) and should be
avoided where possible. Regional analgesia is
effective and safe for day-case surgery, and
should be used as part of the analgesic regime in
all cases where it is practical to do so (Lönnqvist
and Morton, 2005). Single-­injection techniques
like penile block, ilioinguinal-­ iliohypogastric
block and great auricular nerve block are exam-
ples of effective options. A single-dose caudal
epidural block can be appropriate for paediatric
day-case surgery although inadvertant motor
block may be an obstacle to timely discharge.
NSAIDs and paracetamol should also be used
routinely unless specifically contraindicated.
Oral analgesics are the mainstay of continuing
pain relief at home after day surgery and it is vital
to encourage parents to give analgesics preemp-
tively for 24–48  h and before any local anaes-
thetic has worn off. They should also be advised
on whom to contact if pain control problems
arise.
5.22.5 Post-operative Nausea
and Vomiting
Post-operative nausea and vomiting is a funda-
mentally unpleasant experience for the child. It
also causes delays in discharge. Many day-case
surgeries in paediatrics carry an increased risk
including ENT surgery, ophthalmic surgery and
orchidopexy. Prevention is more effective than
treatment and minimising the use of opioids, con-
sidering TIVA and prophylactic use of antiemet-
ics in high risk patients can decrease the
incidence. Ondansetron followed by dexametha-
sone are the most commonly used intraoperative
antiemetic agents.
5.22.6 Discharge Criteria
Prior to discharge, the child should be fully con-
scious, pain-free and able to move normally. Vital
signs should be normal. There should be no respi-
C. C. Smyth and S. Crowe
ratory distress or stridor. The ability to drink and
tolerate clear fluids is desirable, but adequately
hydrated children can be allowed home prior to
drinking if they meet other discharge criteria.
5.22.7 Reasons for Hospital
Admission
Approximately 1% of children undergoing day-­
case surgery ultimately require overnight hospi-
tal admission. The reasons include persistently
abnormal vital signs or level of consciousness,
persistent nausea and vomiting, surgical or anaes-
thetic problems (unexpectedly prolonged or dif-
ficult surgery, regurgitation, aspiration, allergic
reactions), bleeding and difficulties with pain
control.
5.22.8 Transport Home
The child should travel home in a private car or
taxi and should be accompanied by a responsible
adult. Use of buses or trains should be avoided
and the total travelling time should not be
­excessively long. They should be accompanied
overnight by a responsible caregiver with the
ability to contact emergency services if required.
5.23 Some Topics of Current
Interest to Both
Anaesthetists and Surgeons
5.23.1 Fasting Prior to Anaesthesia
and Surgery
Pulmonary aspiration of acidic gastric contents
has long been recognized as a cause of morbidity
and mortality in patients undergoing anaesthesia
and surgery. While the precise incidence of this
dreaded complication is unknown, there is evi-
dence that children are affected more frequently
than adults, with as many as 26% of deaths asso-
ciated with paediatric anaesthesia being attrib-
uted to aspiration in one series. Pre-operative
fasting has long been the mainstay in preventing
5  Anaesthesia and Analgesia
Table 5.1  Fasting guidelines prior to anaesthesia and
surgery
Children scheduled for elective anaesthesia
        Clear fluids: 1 h minimum
Breast milk: 4 h minimum
Other fluids and all solids: 6 h minimum
Children scheduled for emergency anaesthesia
        All fluids and solids: 6 h minimum
aspiration. For decades, children were “fasted
from midnight the night before”. If surgery was
delayed, fasting times became excessively long.
This caused unnecessary discomfort for the child.
There was also the risk of hypoglycaemia and
dehydration—there are obvious dangers involved
in administering potent anaesthetic drugs to
potentially hypovolaemic young children.
Furthermore, fasting leads to a state of physio-
logical stress in the child inducing increased cat-
echolamines, cortisol and insulin resistance. The
child becoming hungry, thirsty and emotionally
upset has been shown to make inhalational induc-
tion of anaesthesia more difficult. It follows that
the minimum “starve time” that will not signifi-
cantly increase the risk of regurgitation and aspi-
ration of gastric contents should be used in
institutions caring for the child undergoing sur-
gery. A recent consensus guideline has decreased
the time for clear fluid from 2 hours to 1 hour pre-­
induction of anaesthesia. It also stated that the
primary issue with fasting times is the lack of
parent education that longer fasting is not safer.
Studies have shown that most children fast for
much longer than the required time prior to elec-
tive surgery. Table  5.1 outlines some current
recommendations.
5.23.2 Upper Respiratory Tract
Infection
In the past, almost all children with evidence of
upper respiratory tract infection (URTI) had their
surgery postponed. While most studies agree that
children with active or recent URTI are at
increased risk of perioperative complications, for
the most part, these are manageable and without
long-term sequelae (Tait and Malviya, 2005).
69
Most paediatric anaesthetists now agree that chil-
dren with mild uncomplicated URTIs undergoing
procedures that do not involve instrumentation of
the airway can be safely anaesthetized without
any significant increase in risk. Most also agree
that any child with severe symptoms should have
surgery deferred for at least 4 weeks.
5.23.3 Anaesthesia
and Immunization
Anaesthetists are often faced with a child who
has recently been immunized presenting for
either elective or emergency surgery. Concerns
over whether or not the anaesthesia or surgery
will affect the response of the child to the vaccine
in achieving seroconversion and as a result
immunity. Secondly, and more seriously, whether
the vaccine might cause major adverse conse-
quences to perioperative outcome. There is lim-
ited high level evidence on this topic. A recent
international survey carried out to ascertain the
attitudes and practices of paediatric anaesthetists
regarding anaesthesia in the child who had
recently been immunized or who was scheduled
for immunization in the near future revealed little
consensus of opinion (Short, 2006). It does seem
prudent, however, to adopt a cautious approach
where the timing of elective surgery is concerned,
and guidelines issued as a result of this survey are
summarized in Table 5.2.
Table 5.2  Guidelines regarding anaesthesia, surgery and
immunization
1. Postpone all elective procedures requiring
anaesthesia rather than immunization, especially
in infants.
2. Opportunistic immunization during anaesthesia is
inadvisable.
3. Postpone anaesthesia for 1 week after vaccination
with inactive vaccines: diphtheria, tetanus,
pertussis, inactive polio, Hib, Meningitis C.
4. Postpone anaesthesia for 3 weeks after vaccination
with live attenuated vaccines: measles, mumps,
rubella, oral polio vaccine and BCG.
5. Delay immunization for 1 week after surgery has
taken place.
70
5.24 Enhanced Recovery After
Surgery (ERAS)
in the Paediatric Population
The use of an ERAS programme is commonplace
in adult anaesthesia and surgery. The aim is to
optimise the patient’s progress through a resource
efficient programme, benefiting both the patient
and the system. The key pillars of ERAS include
preoperative assessment to avoid day-of-surgery
delays or cancellations, limited fasting times,
multimodal analgesia, early mobilisation and
feeding and early removal of drains. This is in
combination with the least invasive surgical
option. ERAS in the paediatric population is
becoming more popular. The most adaptable
components include the preoperative assessment
and discussion with the patient and their family
regarding expectations. Intraoperatively, regional
anaesthesia and a multimodal approach to anal-
gesia will all decrease the need for the over use of
opioid medications. However, the blanket appli-
cation of ERAS on the inherently heterogeneous
paediatric population requires more individuali-
sation than their adult counterparts. There will
need to be age- and disease-appropriate ERAS
protocols as the infant cannot be treated accord-
ing to the same protocol as a 6-year-old or a
16-year-old.
5.25 Conclusion
This chapter summarises the key elements of
paediatric anaesthesia. The administration of a
safe and appropriate anaesthetic is now accepted
as a key component of paediatric surgery
(Stoddart and Lauder, 2004). We highlight the
safe journey through the preoperative, intraoper-
ative, and postoperative periods and how each
stage is managed. We have shown the key consid-
erations in specific patient cohorts going for cer-
tain surgical procedures. This basic anaesthetic
knowledge is important for the surgeon, allowing
them to function effectively as a central member
of the perioperative team. This inter-speciality
education optimises the efficiency of the whole
C. C. Smyth and S. Crowe
team and therefore, positively impacts patient
outcomes.
References
Al-Rawi O, Booker PD (2007) Oesophageal atresia and
tracheo-oesophageal fistula. Continuing Education in
Anaesthesia Critical Care & Pain 7(1), February 2007
Baum VC (2007) When nitrous oxide is no laughing mat-
ter: nitrous oxide and pediatric anesthesia. Pediatr
Anesth 17:824–830
Campbell M, Pierce J (2015) Atmospheric science,
anaesthesia, and the environment. BJA Education
15:173–179
Friesen RH, Wurl JL, Charlton GA (2000) Haemodynamic
depression by halothane is age-related in paediatric
patients. Paediatr Anaesth 10:267–272
Gaynor J, Ansermino JM (2016) Paediatric total intrave-
nous anaesthesia. BJA Education 16(11):369–373
Harbaugh CM, Vargas G, Streur CS, et  al (2019)
Eliminating unnecessary opioid exposure after com-
mon children’s surgeries [published online September
4, 2019]. JAMA Surg. https://doi.org/10.1001/
jamasurg.2019.2529
Lerman J (2002) Inhalational agents. In: Bissonnette B,
Dalens BJ (eds) Pediatric anesthesia principles and
practice. McGraw-Hill, New York, pp 215–236
Llewellyn N, Moriarty A (2007) The national pediatric
epidural audit. Pediatr Anesth 17:520–533
Lönnqvist P-A, Morton NS (2005) Postoperative analge-
sia in infants and children. Br J Anaesth 95:59–68
Moscuzza F (2003) Monitoring during paediatric anaes-
thesia. Anaesth Intens Care Med 4:19–21
Nuelle JAV, Maj MC, Coe KM, Oliver HA, Cook JL,
Hoernschemeyer DG, Gupta SK (September 2020)
Effect of NSAID use on bone healing in pediatric frac-
tures: a preliminary, prospective, randomized, blinded
study. J Pediatr Orthop 40(8):e683–e689. https://doi.
org/10.1097/BPO.0000000000001603
Puri P, Höllwarth ME (eds) (2006) Pediatric surgery.
Springer, Berlin, Heidelberg
Sadeghi A, Khaleghnejad Tabari A, Mahdavi A, Salarian
S, Razavi SS (2017) Impact of parental presence dur-
ing induction of anesthesia on anxiety level among
pediatric patients and their parents: a randomized clin-
ical trial. Neuropsychiatr Dis Treat 12, 3237–3241.
Published 2017 Feb 20. https://doi.org/10.2147/NDT.
S119208
Short JA, Van der Walt JH, Zoanetti DC (2006) Immuni-
zation and anesthesia: an international survey. Pediatr
Anesth 16:514–522
Stoddart PA, Lauder GL (2004) Problems in anaesthesia:
paediatric anaesthesia. Taylor & Francis, London
Tait AR, Malviya S (2005) Anesthesia for the child with
an upper respiratory tract infection: still a dilemma?
Anesth Analg 100:59–65
 Respiratory Management
of the Surgical Patient
Gregory Nolan and Suzanne Crowe
6.1 Introduction
The need for post-operative respiratory support
in the paediatric population is multifactorial.
Pain, the depressive effect of anaesthesia and
analgesic agents, and surgery that results in com-
promised respiratory mechanics are all common
indications for the need for a variable period of
post-operative ventilation. Neonatal patients, due
to their slower metabolism and elimination path-
ways, are especially vulnerable to a prolongation
of the depressive effects of anaesthesia, necessi-
tating post-operative respiratory support.
Compromised respiratory mechanics can
result either directly from thoracic surgery or
indirectly from abdominal surgery that causes an
increase in intra-abdominal pressure. Surgical
procedures, like gastroschisis reduction or con-
genital diaphragmatic hernia (CDH) repair,
which result in upward displacement of the dia-
phragm, can decrease functional residual capac-
ity (FRC) and reduce thoracic compliance. In
infants and younger children, tidal breathing
occurs in the range of closing capacity, i.e. the
volume of the lung at which dependent airways
begin to close. This places them at increased risk
of atelectasis, lobar collapse, and hypoxaemia.
G. Nolan · S. Crowe (*)
Children’s Health Ireland, Crumlin,
Dublin, Republic of Ireland
e-mail: suzanne.crowe@olchc.ie
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_6
6
Modern critical care ventilators are predomi-
nantly turbine-driven. A user-friendly interface
allows the clinician to select a mode that delivers
positive pressure ventilation to patients at the
extremes of prematurity through to adulthood.
The microprocessor controls an inspiratory valve
that opens to deliver warmed, humidified gas to
the patient through the inspiratory limb of a cir-
cuit. An expiratory valve then allows passive
expiration through the expiratory limb of the cir-
cuit, which is filtered, and then vented to the
atmosphere. It is this expiratory valve that also
controls the level of positive end-expiratory pres-
sure (PEEP) delivered to the patient.
The detailed characteristics of ventilators used
in intensive care are complex and beyond the
scope of this chapter. For the purpose of a general
discussion on post-operative ventilation, it is suf-
ficient to know that all mandatory modes deliver
positive pressure to the lung, the ventilator can be
set to deliver gas either according to a preset
pressure limit (pressure-control ventilation) or a
preset tidal volume (volume-control ventilation),
and the inspiratory and expiratory phases are
time-cycled. The number of breaths delivered by
the ventilator can either be set by the operator or
patient-determined when in a supported ventila-
tory mode. The duration of the inspiration and
expiration in each ventilator cycle can be altered
by adjusting the inspiratory time. The machine
cycles from inspiration to expiration when either
the preset pressure or volume is reached.
71
72
6.2 Respiratory Physiology
Paediatric respiratory physiology differs from
that of their adult counterparts. Infants and young
children are at a mechanical disadvantage
because of the high compliance and low elastic
recoil of their chest wall. The smaller paediatric
airways have a higher resistance, and more work
is required to achieve the same tidal volume (per
kilogram) than in adults. A proportion of the
work of breathing is lost in the distortion of the
ribcage. These factors place an infant or younger
child at a greater risk of atelectasis and lung col-
lapse—and certain patients will require a period
of post-operative ventilation.
6.2.1 Gas Exchange
In the lungs, the uptake of oxygen and the elimi-
nation of carbon dioxide between air and blood
occurs by simple diffusion.
6.2.1.1 Oxygen Uptake
In room air, the diffusion gradient for oxygen
allows it to travel from high pressure in the alve-
oli (PAO2 105  mm Hg) to low pressure in the
pulmonary artery (40 mm Hg). While the gradi-
ent is favourable, the alveoli must remain open
for gas exchange to occur. Hysteresis is a prop-
erty of the normal lung used to describe the dif-
ference between inspiratory and expiratory
compliance. It explains the alveoli opening dur-
ing inspiration and remaining open during most
of the expiration and only ceasing gas exchange
at end-expiration.
Positive end-expiratory pressure (PEEP)
delivered by a ventilator maintains the patency of
the alveoli through the respiratory cycle. This
facilitates further gas exchange and improves
oxygenation. In oedematous or atelectatic lung,
the clinician can increase the PEEP, and thereby
the mean airway pressure (Paw), to maintain the
patency of otherwise unstable alveoli that would
otherwise collapse at low lung volumes.
The fraction of inspired oxygen (FiO2) can
also be increased to compensate for a large alveo-
lar to arterial oxygen gradient.
G. Nolan and S. Crowe
6.2.1.2 Carbon Dioxide Elimination
CO2 homeostasis is achieved by balancing CO2
production (metabolic rate) with its elimination
via alveolar ventilation (alveolar ventila-
tion = tidal volume – dead space). In spontaneous
respiration, the child compensates for any
increases in CO2 production, as seen in fever, rig-
ors, or sepsis, by increasing alveolar ventilation,
primarily by increasing their respiratory rate. In
the same way, any loss of lung volume as a result
of an intra-thoracic lesion will result in an
increase in the respiratory rate rather than the
tidal volume to achieve the same alveolar
ventilation.
In mechanical ventilation, the expired minute
ventilation (MVe) can be augmented by adjusting
either the ventilator tidal volume or rate
(MVe  =  VT  ×  rate). To reduce the incidence of
ventilator-induced lung injury, the tidal volume
should not exceed the physiological range of
6–8 mL/kg. At these physiological tidal volumes,
increasing the ventilator rate is usually the most
effective way of increasing CO2 clearance.
6.3 Mechanical Ventilation
Intermittent mandatory ventilation (IMV) deliv-
ers a preset number of clinician-controlled
breaths, which can either be pressure or volume
(flow) limited. In its most commonly used form,
synchronised intermittent mandatory ventilation
(SIMV), these mandatory breaths are synchro-
nised to coincide with patient effort. While the
clinician can set the ventilator rate, in SIMV, the
breaths are delivered at the patient’s intrinsic
respiratory rate rather than evenly spaced over
the minute.
6.3.1 Pressure-Control Ventilation
In a pressure-control mode, for example
SIMV-PC, the clinician is able to set a peak
inspiratory pressure (PIP), which then, depend-
ing on the patient’s lung compliance and airway
resistance, delivers a tidal volume. The positive
end-­expiratory pressure (PEEP), inspiratory
6  Respiratory Management of the Surgical Patient
time, and number of mandatory breaths are also
clinician-set.
In an uncomplicated post-operative patient,
adequate tidal volumes (6–8  mL/kg) can be
achieved with a PIP of 15–20 cm H2O. Due to the
fixed pressure setting, any changes to airway
resistance or lung compliance, however small,
will have an effect on the tidal volume delivered
to the patient. An increase in airway resistance
will require an increase in set ventilator pressures
to achieve the same tidal volumes and maintain
the same minute ventilation.
6.3.2 Volume-Control Ventilation
In contrast with pressure-control ventilation,
volume-­control ventilation is where the clinician
is able to set a tidal volume appropriate for the
patient’s weight (mL/kg). The ventilator will
compensate for any change in patient conditions;
i.e. if there is a fall in compliance or a rise in air-
way resistance, minute ventilation will remain
the same, as the ventilator increases the inspira-
tory pressure to deliver a constant tidal volume.
To avoid pressure-related lung injury (baro-
trauma), the clinician can also set a limit to the
pressure delivered, usually 5 cm H2O above the
PIP required for the set tidal volume. The clini-
cian also sets a rate of mandatory breaths, as well
as PEEP and inspiratory time.
Historically, the use of uncuffed endotracheal
tubes in children, with often large airway leaks,
made volume-control ventilation a less reliable
mode of ventilation. However, with the develop-
ment of low-profile cuffed tubes in paediatric air-
way management, the volume-control mode has
added utility, and SIMV-VC is now the preferred
mode of ventilation in many paediatric intensive
care units.
6.3.3 Support Modes
In both volume- and pressure-control synchro-
nised modes of ventilation, any breaths that the
patient triggers above the mandatory rate will
also be pressure-supported. While this feature
73
exists as an adjunct to the mandatory modes
described above, it is also a standalone mode,
where the pressure-supported breaths are deliv-
ered in conjunction with a continuous positive
airway pressure (CPAP, synonymous with PEEP).
Pressure-support ventilation becomes an
important concept when weaning a patient from
the ventilator. As the patient’s clinical condition
improves, the number of mandatory breaths
delivered by the ventilator can be slowly reduced,
allowing the patient to breathe up over the venti-
lator. Once a minimum mandatory rate is achieved
with satisfactory gas exchange, the clinician can
next reduce the PEEP and inspiratory pressures
by improving lung compliance before attempting
extubation.
6.3.4 Triggering the Ventilator
‘Triggering’ in mechanical ventilation refers to
the process that results in the opening of the
inspiratory valve and the initiation of inspira-
tion. This is either machine-initiated in a man-
datory mode or patient-initiated in a support
mode. A patient can initiate a breath by over-
coming either a flow or pressure trigger set by
the clinician. In the more novel neurally adjusted
ventilatory assist (NAVA) mode, breath delivery
is timed with the electrical activity of the dia-
phragm by means of a nasogastric catheter with
embedded electrodes (Liet et al. 2016). Patient-
triggered modes of ventilation are useful in
assisting weaning from ventilation.
6.3.5 Setting Ventilator Parameters
The goals of ventilation are to maintain normal
physiological parameters with respect to PaCO2
and PaO2, while minimising inspiratory pressures
(and tidal volumes) to avoid lung injury. The ven-
tilator settings will largely be determined by the
age and weight of the child. Having said that, an
otherwise well child would commonly require a
PEEP of 5  cm H2O and peak inspiratory pres-
sures of 15–20 cm H2O to achieve physiological,
non-injurious tidal volumes of 6–8 mL/kg. Peak
74
inspiratory pressures exceeding 30  cm H2O are
potentially injurious to the lungs (Kneyber et al.
2017).
The inspiratory to expiratory ratio (I:E ratio)
is usually fixed at 1:2, with an inspiratory time
that can be extrapolated from an age-appropriate
respiratory rate. The inspiratory time (Ti) can be
increased to facilitate a longer period of gas
exchange and reduce the required PIP to achieve
a desired tidal volume in VC mode. This however
is at the expense of a shortened expiratory time,
which can lead to breath-stacking—the initiation
of a new breath prior to the complete expiration
of the previous breath (Bersten 2009).
Measurement of ventilatory sufficiency, i.e.
respiratory monitoring, is discussed below.
6.4 Alternative Modes
of Ventilation
6.4.1 High-Frequency Oscillatory
Ventilation (HFOV)
In a patient who is difficult to either oxygenate or
ventilate, HFOV represents an escalation in respira-
tory support. The ventilator generates a high-fre-
quency sine wave that accelerates diffusion and
elimination of CO2 from the lung with minimal bulk
gas flow. Oscillations typically occur at between 8
and 12 Hz (frequency per second). Oxygenation is
achieved by a high fresh gas flow into the circuit
and, by adjusting this, the device is capable of gen-
erating mean airway pressures in excess of 30 cm
H2O, while minimising the shearing forces associ-
ated with ventilating at high pressures in a conven-
tional mode. The clinician is able to adjust the Paw
and FiO2 to improve oxygenation and can either
increase the amplitude or decrease the frequency
(Hz) to aid CO2 elimination. This mode of ventila-
tion has proven efficacy in acute respiratory distress
syndrome (ARDS) and congenital diaphragmatic
hernia (CDH) where CO2 elimination needs to
occur at low airway pressures to avoid pulmonary
hypertensive crises (Wong et al. 2020).
G. Nolan and S. Crowe
6.4.2 Non-invasive Ventilation (NIV)
NIV has utility as a step-down therapy in the
extubated patient or as an escalation of support in
the self-ventilating patient with respiratory fail-
ure. PEEP and pressure-support ventilation can
be delivered through either a face or nasal mask.
Airway pressure remains positive throughout
both the inspiratory and expiratory cycles.
Functional residual capacity (FRC) is main-
tained, decreasing the work of breathing and pre-
venting lung collapse. There is an additional
potential advantage in that avoiding invasion of
the airway reduces the risk of ventilator-­
associated pneumonia (VAP).
6.5 Endotracheal Tube (ETT) Size
and Positioning
The useful formula for determining the appro-
priate internal diameter of the ETT as a func-
tion of age (ETT (mm) = (Age (years)/4) + 4)
has been largely superseded by the introduc-
tion of low-­profile cuffed tubes for which there
are manufacturer-­ recommended age and
weight guidelines. In general, a child will
require a cuffed tube with an internal diameter
of 0.5  mm less than that calculated with the
above formula. Nasal fixation, rather than oral,
for the ETT is preferred in neonates and small
children. This provides for greater patient
comfort and acceptability, and the tube may be
more securely fastened to the face and upper
lip.
Following insertion of the ETT, the operator
confirms the position of the tube by directly
visualising its markings at the level of the glot-
tis, auscultation of the chest, and waveform cap-
nography—which is the gold standard for
differentiating between endotracheal and
oesophageal intubation. A radiographic plain
film of the chest ordered after intubation should
demonstrate the tip of the tube at the level of the
clavicles.
6  Respiratory Management of the Surgical Patient
6.6 Care of the Intubated Patient
6.6.1 Humidification
Humidification is one of the least emphasised but
most important aspects of respiratory care.
Appropriate humidification of the inspired gases
not only helps to prevent complications from the
drying of the airway mucosa but also improves
lung compliance and assists in avoiding the dan-
gerous complication of endotracheal tube block-
age from inspissated secretions.
Optimal humidity is the delivery of fully satu-
rated gases (44  mg/L H2O) at a temperature of
37  °C to the lower airways. This is most effi-
ciently achieved with the heated-bath type
humidifiers, which are commonplace on most
modern ventilators. A heated electric coil in the
inspiratory limb of the ventilator circuit helps to
prevent the condensation of water vapour from
forming in the tubing. Temperature sensors at the
water bath and at the patient’s airway maintain a
consistent temperature through the inspiratory
limb of the ventilator circuit. As a result, there is
less cooling of gases in the ventilator tubing and
the inspired gas is delivered to the ET tube fully
saturated at 37 °C.
Heat and moisture exchangers (HMEs) or
Swedish noses are also capable of supplying
moisture and preventing heat loss from the respi-
ratory tract, though far less efficiently than the
water-bath type humidifiers. HMEs have their
utility in patient transport and anaesthetic cir-
cuits, where ventilation is not prolonged and a
water-bath humidifier is not practical. As the con-
denser–humidifier becomes increasingly satu-
rated, the airway resistance tends to rise, and they
must be changed every 24 hours.
6.6.2 Endotracheal Tube Suctioning
Routine suctioning to maintain the patency of
the endotracheal tube and prevent atelectasis is
of utmost importance in the ventilated patient.
To prevent nosocomial infection, it should be
75
performed as a sterile procedure with surgical
gloves and a clean catheter. The patient should
be pre-­oxygenated to avoid hypoxaemia, desatu-
ration, and bradycardia. Depending on individ-
ual unit protocols, suctioning can be performed
either in-­line or by detaching the patient from
the ventilator and suctioning directly down the
ETT.  With the latter method, due caution must
be given to the aerosolisation of the patient’s air-
way secretions.
6.7 Respiratory Monitoring
Monitoring respiratory function in the post-­
operative patient is crucial. In the ventilated
patient, it is essential to monitor the main func-
tions of the respiratory system—the facilitation
of oxygen transport and the removal of carbon
dioxide—in order to properly assess the patient’s
status, progress, and their requirement for
increasing respiratory support while mitigating
the injurious effects of over-ventilation (Cheifetz
2016).
6.7.1 Invasive Monitoring
The most accurate assessment of gas exchange is
via arterial blood sampling. Invasive arterial lines
can be commonly found in the radial and brachial
arteries in the upper limb, and dorsalis pedis and
posterior tibial arteries in the lower limb. In the
neonatal intensive care unit, an umbilical arterial
line is sometimes transduced for invasive blood
pressure monitoring and gas sampling. The par-
tial pressure of oxygen (PaO2), carbon dioxide
(PaCO2), and acid–base (pH) can be directly
measured from these samples.
PEEP and inspired oxygen are titrated to
maintain a PaO2 greater than 60–70 mm Hg. In
the uncomplicated post-operative patient, minute
ventilation can be adjusted to maintain a PaCO2
of between 35 and 45  mm Hg. With modern
blood gas machines, samples need only be 0.1–
0.2  mL for a full acid–base profile, as well as
76 G. Nolan and S. Crowe

electrolytes (including sodium, potassium, chlo- 6.7.2.3 Transcutaneous Carbon

ride, and ionised calcium). These small samples Dioxide (TcPCO2)
prove particularly useful in low birth weight and Transcutaneous measurement of carbon diox-
premature infants in whom frequent blood sam- ide has specific clinical utility in the oscillated
pling may necessitate “top-up” transfusions. patient, in whom end-tidal measurements of
CO2 are not possible. Transcutaneous CO2
(TcPCO2) monitoring uses the modified
6.7.2 Non-invasive Monitoring Severinghaus probe with a heating element,
which heats the skin to 41–44  °C.  This has
6.7.2.1 Pulse Oximetry (SpO2) been shown to provide a reliable proxy for
Pulse oximetry provides a continuous, non-­ PaCO2 in children and small infants.
invasive measure of oxygen saturation of arterial Transcutaneous monitoring is most useful in
haemoglobin. A light-emitting diode, positioned the observation of trends; the absolute values
on a narrow area of the body, such as a finger or will need to be correlated at intervals with an
toe, allows light to pass through a pulsating arterial sample. The risk of burns from the
­capillary bed where it is sensed by a photodetec- heating element requires that the probe site be
tor on the other side. Measurements of light changed every 4–6 hours.
absorption at wavelengths of 660  nm (red) and
960 nm (infrared) during the peak of a transmit-
ted pulse are used to derive a surrogate for arte- 6.8 Conclusions
rial oxygen saturation (SpO2).
Pulse oximetry gives pulse-to-pulse variation, Respiratory management of paediatric surgical
allowing for rapid response to patient deteriora- patients is uniquely challenging. Paediatric respi-
tion. There are however limitations to pulse ratory physiology differs from that of their adult
oximetry, with measurement inaccuracies at high counterparts. Infants and young children are at a
and low extremes of saturation. Motion artefact, mechanical disadvantage because of the high
low perfusion states, and dyshaemoglobinaemias compliance and low elastic recoil of their chest
are also contributors to potential inaccuracies wall. The smaller paediatric airways have a
with pulse oximetry. higher resistance, and more work is required to
achieve the same tidal volume (per kilogram) as
6.7.2.2 End-Tidal Carbon Dioxide in adults. A proportion of the work of breathing is
(PETCO2) lost in the distortion of the ribcage. These factors
Capnography is presented as a graphic waveform place an infant or younger child at a greater risk
of the concentration of expired CO2 over time. It of atelectasis and lung collapse. Optimal respira-
forms a part of the standard monitoring of the tory management of the post-operative paediatric
intubated patient and is the gold standard for con- surgical patient requires a thorough understand-
firming endotracheal intubation. In the normal ing of patient anatomy, physiology, and clinical
capnogram, CO2 is zero during inspiration and at condition.
the beginning of expiration as anatomical dead
space gas is washed out. During expiration, the
CO2 concentration rises as alveolar gas reaches References
the sensor. The peak CO2 value during expiration
approximates alveolar CO2 (PACO2), and there- Bersten A (2009) Mechanical ventilation. In: Bersten A,
fore, because of its diffusion characteristics, arte- Soni N (eds) Oh’s intensive care manual, 6th edn.
Butterworth-Heinemann, New York
rial CO2 (PaCO2). In patients with normal lungs, Cheifetz I, Lee J, Venkataraman S (2016) Respiratory
the PETCO2 is usually 5 mm Hg less than the mea- monitoring. In: Nichols D, Shafner D (eds) Rogers’
sured PaCO2. In the presence of lung disease, the textbook of pediatric intensive care, 5th edn. Woltere
PaCO2-PETCO2 gradient widens (West 2015). Kluwer, New York
6  Respiratory Management of the Surgical Patient
Kneyber MCJ, de Luca D, Calderini E et  al (2017)
Recommendations for mechanical ventilation of
critically ill children from the Paediatric Mechanical
Ventilation Consensus Conference (PEMVECC).
Intensive Care Med 43(12):1764–1780
Liet JM, Barrière F, Gaillard-Le Roux B, Bourgoin P,
Legrand A, Joram N (2016) Physiological effects of
invasive ventilation with neurally adjusted ventila-
77
tory assist (NAVA) in a crossover study. BMC Pediatr
16(1):180
West J, Luks A (2015) West’s respiratory physiology, 10th
edn. Wolter Kluwer, New York
Wong JJ, Liu S, Dang H et al (2020) The impact of high
frequency oscillatory ventilation on mortality in pae-
diatric acute respiratory distress syndrome. Crit Care
24(1):31
 Fluid Management
S. O’Sullivan and Suzanne Crowe
7.1 Paediatric Fluid Haemostasis
Paediatric physiology presents unique challenges
to achieving normal fluid homeostasis. Input
must balance output to ensure an adequate circu-
lating volume and therefore maintain tissue per-
fusion (Bohn 2009).
7.2 Intake
Intake varies hugely across a paediatric popula-
tion. First, the normal thirst and subsequent
behavioural response to seek water do not exist in
infants and young children. Second, it may be
limited by medical co-morbidities that obviate
the capacity for normal enteral intake—this may
be structural (i.e. duodenal atresia) or functional
(i.e. neurological disease). Lastly, certain vulner-
able populations, such as those with underlying
metabolic disorders or pre-term neonates with
low birth weights, may have relatively restricted
or increased nutritional requirements and, as will
S. O’Sullivan
Paediatric Intensive Care Unit, Our Lady’s Children’s
Hospital Crumlin, Dublin, Ireland
e-mail: osullisd@tcd.ie
S. Crowe (*)
Children’s Health Ireland, Crumlin,
Dublin, Republic of Ireland
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_7
7
be addressed later, metabolic and fluid require-
ments appear to correlate quite closely.
7.3 Output
Output regulation is also more challenging in the
paediatric population. Glomerular filtration rate
(GFR) is lower at birth and increases thereafter, as
does the capacity for restriction of sodium excre-
tion. It is inversely proportional to age in the pre-
term neonate. This renders the patients more
susceptible to disproportionate sodium loss, as
well as volume overload. Moreover, infants strug-
gle at the extremes of urine concentration, both
high and low, and coupled with the relatively high
solute load found in formula feed renders them
less able to compensate for states of water loss
(i.e. gastroenteritis). Urine is also the predomi-
nant physiological source of electrolyte loss in the
absence of a co-existent GI pathology. In criti-
cally ill or post-op surgical patients, neurohumor-
ally mediated losses can increase dramatically
above the baseline level of 2–3  mmol/kg/day of
sodium and 1–2 mmol/kg/day of potassium.
Younger patients also have a relatively
increased skin surface area to body weight ratio,
allowing them to lose both heat and water more
easily. (Due to the relationship between water
evaporation from skin and the latent heat of
vaporisation required for evaporation, the two
components are intimately linked.) Thermal
79
80
stress (i.e. pyrexia) can increase these losses dra-
matically. In the absence of pathology, such as
cystic fibrosis, sweat predominantly contains
water with a small amount of sodium. Insensible
losses from the respiratory tract (estimated at
15 ml/100 kcal/day) are the other large contribu-
tor to fluid balance under physiological condi-
tions. Stool fluid loss is slight unless some
pathology is superimposed.
Therefore, there are a variety of consider-
ations to be balanced when considering fluid
therapy for a paediatric patient.
7.4 Fluid Therapy
7.4.1 Historical Basis: Composition
and Rate
Paediatric fluid therapy owes much to the semi-
nal paper by Holliday and Segar (1957) (Holliday
and Segar 1957). This established the central
tenet of weight-based maintenance hypotonic
(lower osmolality than plasma) fluid therapy to
meet ongoing requirements in fasting children,
with added glucose to stave off the risk of unde-
tected and potentially disabling hypoglycae-
mia—especially in neonates who were felt to be
particularly vulnerable given their reduced gly-
cogen stores and the non-specific and predomi-
nantly somnolent effects of hypoglycaemia.
Although glucose does contribute to calcu-
lated osmolality in IV fluids, it is rapidly taken up
and consumed during cellular metabolism and
therefore does not actually contribute to their
in vivo osmolality. Consequently, a solution such
as 5% dextrose is, in practice, a means of admin-
istering free water and will redistribute equally
amongst fluid compartments.
This article suggested that based upon the
electrolyte content of both human’s and cow’s
milk, the electrolyte requirements for 24 h were
Na 3 mEq/100 kcal and K 2 mEq/100 kcal (which
correlates closely with the observed urinary
losses noted above).
This resulted in years of hypotonic fluid
administration to fasting children, most usually
formulated as 0.45% normal saline with 5% dex-
trose, at a rate determined by the 4-2-1 rule. This
S. O’Sullivan and S. Crowe
postulated that 1  ml of water was required for
every 1 kcal of energy metabolised.
Thus, based on the calculated expenditure of
the average hospital patient, it was derived that
the hourly rate of fluid administration is 4  ml/
kg/h. for the first 10 kg bodyweight, 2 ml/kg/h.
for the second 10  kg, and 1  ml/kg/h. thereafter.
As a convenient rule of thumb for patients greater
than 20  kg, hourly fluid administration
rate = weight (in kg) + 40 ml/h.
These calculated fluid requirements do not
apply to neonates as full maintenance fluid
requirements for the non-ventilated term new-
born commence at 60 ml/kg/24 h in the first day
of life and increase thereafter to 100–150  ml/
kg/24  hours by day 5 of life (Green and Lillie
2017). In the ventilated newborn, however, these
requirements are decreased to 70% of baseline
due to the reduced loss of both water and heat in
the warmed, humidified ventilator circuit.
Importantly, Holliday and Segar make specific
mention of the fact that this formula is intended
as maintenance replacement only. It does not
account for pre-existing deficits OR ongoing
losses (i.e. bleeding, vomiting, diarrhoea).
Misinterpretation of this intention has occasion-
ally led to the prescribing of ‘multiples’ of main-
tenance fluid rates in order to correct losses,
leading to a potentially dangerous over-­
administration of hypotonic solutions, and there-
fore excess free water (Oh and Sutherland 2016).
At this point, it is worth noting that the best
fluid therapy for a child in fact involves restoring
enteral intake as completely and as early as pos-
sible. This obviates the potential complications
of fluid therapy that will be discussed later in the
chapter and takes advantage of native homeo-
static mechanisms governing electrolyte and
mineral absorption and subsequent plasma con-
centrations. As such, recent advances, particu-
larly in the anaesthetic management of paediatric
patients, have capitalised on revised fasting
guidelines from the AABGI—among other pro-
fessional societies—to minimise fasting times
(Thomas et al. 2018; Sümpelmann et al. 2019).
For children who cannot be fed for whatever
reason—usually in the peri-operative period—
there are a number of strategies for resuming
feeding. However, the details of TPN and specific
7  Fluid Management 81

enteral feed composition are far beyond the scope
of this chapter and are best approached with
expert allied health input.
More recently, this ‘classical’ fluid recipe,
particularly in the post-operative paediatric surgi-
cal patient, has been called into question.
7.4.2 Hyponatraemia and ADH
Release
Surgery, as well as a multitude of other stimuli
including pain, nausea, stress, opioids and posi-
tive pressure ventilation, has been demonstrated
to increase anti-diuretic hormone (ADH) secre-
tion (Bailey et al. 2010). This is distinct from the
syndrome of inappropriate anti-diuretic hormone
(SIADH), which is due to a pathology—most
often lung or CNS—rather than as an appropriate
response to an external stimulus.
ADH is responsible for water retention by pro-
moting aquaporin translocation into the membrane
of the kidney collecting tubule, thereby promoting
re-absorption of free water from the urine, and is
usually released in response to either a drop in
effective circulating volume or secondarily due to
a change in plasma osmolality (Cuzzo et al. 2020).
Under the circumstances described above, a post-
op patient is primed for a non-­physiological secre-
tion of ADH and subsequent free water retention
(without the associated electrolytes usually found
in plasma). This increases plasma volume and
causes dilutional hyponatraemia.
This post-op drop in sodium has been demon-
strated in a number of surgical populations and
has been associated with adverse outcomes due
to rapid sodium shifts and osmotic complications
(Oh and Sutherland 2016; Bailey et  al. 2010;
Sümpelmann et  al. 2011a, b, 2017). Hypotonic
fluid administration may have exacerbated these
shifts as the maintenance fluids prescribed con-
tained an excess of free water, which was then
not adequately excreted.
7.4.3 Isotonic Crystalloids
These issues with hyponatraemia have prompted
a switch to isotonic fluids for maintenance ther-
apy, of which 0.9% normal saline (NaCl) and
compound sodium lactate (Hartmann’s solution
or Ringer’s lactate) are among the two most com-
mon, though other solutions such as Ringer’s
acetate and Plasma-Lyte also exist. See Table 7.1
for compositions of some commonly used fluids.
These fluids have an osmolality approaching
that of plasma (288 mOsmol/kg H2O) and there-
fore do not easily redistribute between fluid com-
partments the way free water (administered as
dextrose 5%) does. This keeps them within the
intravascular space for longer and thereby makes
them more appropriate resuscitation fluids for an
existing deficit, maintenance therapy and for ongo-
ing losses, though bleeding patients are best resus-
citated with a judicious, balanced blood transfusion
rather than with large volumes of crystalloid.
However, their use does present its own prob-
lems. NaCl (0.9%) has a chloride ion concentra-
tion far above that of plasma (Cl 154 mmol/L),
and even modest administration volumes can pre-

Table 7.1  Composition of commonly used intravenous fluids

Na+ Cl− Osmolality Electrolyte free
Fluid type (mmol/l) (mmol/l) (mOsm/kg) Buffer (mmol/L) pH water/l
0.9 NaCl 154 154 308 5.5 0
0.9 NaCl 5% dex 154 154 560 4 0
5% dex 0.45 NaCl 77 77 406 4 500
5% dex 0 0 252 4 1000
Ringer’s lactate 130 109 272 Lactate 28 6.5 114
Ringer’s lactate 5% 130 109 525 Lactate 28 6.5 114
Plasma-Lyte 140 98 294 Acetate 27, 7.4
Gluconate 23
Plasma 142 103 288 HC03 24 0
82
cipitate a hyperchloraemic metabolic acidosis (a
type of normal anion gap metabolic acidosis) and
subsequent acute kidney injury (AKI). There is
also the concern that 0.9% NS may induce hyper-
natraemia, as the Na concentration is also higher
than that of plasma, although this is rarely appre-
ciated clinically unless there is a defect of sodium
excretion or excessive free water loss.
More balanced solutions, such as CSL, over-
come these problems with electrolyte concentra-
tions closer to that of plasma. It is important to
note that the potassium (K 4  mmol/L) ion con-
centration found in CSL is negligible in compari-
son to total body potassium levels (being a
predominantly intracellular ion) and is therefore
unlikely to contribute significantly to clinically
relevant hyperkalaemia. Moreover, the lactate
content (28  mmol/L) is rapidly metabolised to
bicarbonate in the liver and therefore aids in buff-
ering a metabolic acidosis. In fact, this solution
was developed precisely to resuscitate paediatric
metabolic acidosis (Awad et al. 2008). It will not
contribute to lactataemia, except in patients with
severely compromised liver function.
Despite their more physiologically balanced
electrolyte concentrations, these solutions still
have the potential for post-operative hyponatrae-
mia in a process called ‘desalination’ (Bailey
et al. 2010). As noted above, urinary electrolyte
losses can increase dramatically post-operatively,
rising as high as 350  mmol/L Na, and despite
administering balanced solutions, this loss,
combined with excessive ADH secretion, can
­ Only pre-term neonates and those with very high
result in a detectable, though less marked, fall in
serum sodium concentration (Bohn 2009).
7.5 Colloids
Colloid solutions (solutions containing large
molecules with an osmotic effect) have gained
popularity in recent years, particularly in the
adult population, for their perceived benefit of
remaining within the intravascular compartment
for longer than their crystalloid counterparts.
Studies have struggled to identify a mortality
S. O’Sullivan and S. Crowe
benefit from any single solution, and safety con-
cerns have been raised regarding several prepara-
tions, such as coagulopathy, pruritis and renal
failure with starch solutions, and anaphylaxis and
extra capillary leak with gelatins. Moreover,
these large molecules are administered suspended
in solutions (usually 0.9% saline) and therefore
suffer from all the limitations of their base solu-
tions, too (Bailey et al. 2010).
Albumin is the most well-recognised colloid
solution in use in the paediatric population and
appears to be relatively safe, although concerns
have been raised about its use in traumatic brain
injury. At present, there is little evidence to favour
its use in general paediatric patients, although it
may play a more beneficial role in patients after
cardiac surgery and in neonates, the two patient
populations that are particularly vulnerable to
fluid overload (Bailey et al. 2010).
7.6 Hypoglycaemia
Although hypoglycaemia was, quite rightly,
feared (particularly in the unconscious intra-­
operative patient), its importance in contributing
to peri-operative morbidity may have been over-­
estimated. New evidence suggests that the vast
majority of children can appropriately mobilise
glycogen stores under the stress of surgery, par-
ticularly if enteral intake is restricted as little as
possible before and after the operative period.
pre-op glucose requirements appear to need
maintenance dextrose infusions, and even then,
they involve far lower concentrations than previ-
ously prescribed—1–2% dextrose seems to be
sufficient (Sümpelmann et al. 2011a). There was
also a growing concern that intra- and peri-­
operative hyperglycaemia may have been inad-
vertently contributing to increased osmotic
diuresis and subsequent electrolyte loss, rebound
hypoglycaemia when the fluid intake was
stopped, poor wound healing, and potentially
worsening neurological injury, particularly in the
presence of ischaemia or hypoxia.
7  Fluid Management
7.7 The ‘Ideal’ Fluid
Recently, there has been significant interest in
combining all of the above elements into a putative
‘ideal’ fluid, a balanced electrolyte solution that is
isotonic with plasma and has a low d­ extrose con-
centration. This has been supported by several
consensus guidelines and subsequent granting of
European marketing authorisation for the develop-
ment of such a fluid (Sümpelmann et  al. 2011b;
Berleur et al. 2003). As yet, there is no commer-
cially available preparation that fits the require-
ments listed above. Therefore, in deviating from
standard preparations, the onus is on the prescrib-
ing and administering clinician to ensure that cus-
tom mixtures are safe and that the level of
complexity and risk involved usually outweigh the
benefit gained over judicious appreciation for the
limitations of readily available fluid therapies.
7.8 Conclusion
Intravenous fluid therapy remains a somewhat
contentious topic in paediatrics. Evidence is
often limited or disproportionately confined to an
easily studied population, such as post-op surgi-
cal patients. Nonetheless, fluids are drugs like
any other, and an understanding of their uses and
limitations are important for all clinicians
involved in paediatric care.
Most importantly, it is crucial to remember
that ongoing therapy requires repeated assess-
ment and adjustment. This should include the
measurement of electrolytes, fluid balance, blood
sugar level and haemodynamic parameters, as
well as an awareness of the indication for said
fluids and appreciation of the relative urgency to
discontinue therapy when it is no longer indi-
cated. Rules of thumb, dogma and calculations
based upon historic assumptions cannot be a sub-
stitute for real-time adaptation.
83
References
Awad S, Allison SP, Lobo DN (2008 Apr) The history of
0.9% saline. Clin Nutr 27(2):179–188
Bailey AG, McNaull PP, Jooste E, Tuchman JB (2010
Feb) Perioperative crystalloid and colloid fluid man-
agement in children: where are we and how did we get
here? Anesth Analg 110(2):375–390
Berleur M-P, Dahan A, Murat I, Hazebroucq G (2003 Feb)
Perioperative infusions in paediatric patients: rationale
for using Ringer-lactate solution with low dextrose
concentration. J Clin Pharm Ther 28(1):31–40
Bohn D (2009) Fluid, electrolyte and respiratory manage-
ment. In: Puri P, Höllwarth M (eds) Pediatric surgery
[Internet]. Springer, Berlin, Heidelberg, pp  53–63.
https://doi.org/10.1007/978-­3-­540-­69560-­8_6
Cuzzo B, Padala SA, Lappin SL (2020) Vasopressin
(Antidiuretic Hormone, ADH). In Treasure Island
(FL)
Green J, Lillie J (2017) Intravenous fluid therapy in chil-
dren and young people in hospital N29. Arch Dis
Child Educ Pract Ed 102(6):327–331
Holliday MA, Segar WE (1957 May) The maintenance
need for water in parenteral fluid therapy. Pediatrics
19(5):823–832
Oh GJ, Sutherland SM (2016 Jan) Perioperative fluid
management and postoperative hyponatremia in chil-
dren. Pediatr Nephrol 31(1):53–60
Sümpelmann R, Mader T, Dennhardt N, Witt L, Eich C,
Osthaus WA (2011a Nov) A novel isotonic balanced
electrolyte solution with 1% glucose for intraoperative
fluid therapy in neonates: results of a prospective mul-
ticentre observational postauthorisation safety study
(PASS). Paediatr Anaesth 21(11):1114–1118
Sümpelmann R, Becke K, Crean P, Jöhr M, Lönnqvist
P-A, Strauss JM et al (2011b Sep) European consen-
sus statement for intraoperative fluid therapy in chil-
dren. Eur J Anaesthesiol 28(9):637–639
Sümpelmann R, Becke K, Brenner S, Breschan C, Eich
C, Höhne C et  al (2017 Jan) Perioperative intrave-
nous fluid therapy in children: guidelines from the
Association of the Scientific Medical Societies in
Germany. Paediatr Anaesth 27(1):10–18
Sümpelmann R, Becke K, Zander R, Witt L (2019
Jun) Perioperative fluid management in children:
can we sum it all up now? Curr Opin Anaesthesiol
32(3):384–391
Thomas M, Morrison C, Newton R, Schindler E (2018)
Consensus statement on clear fluids fasting for elec-
tive pediatric general anesthesia. Pediatr Anesth
[Internet] 28(5):411–414. https://onlinelibrary.wiley.
com/doi/abs/10.1111/pan.13370
 Sepsis
Lexie H. Vaughn and Jeffrey S. Upperman
8.1 Introduction Including
Definition and Incidence
Sepsis is a major cause of morbidity and mortal-
ity in the United States, accounting for over
720,000 hospitalizations annually. Despite recent
advances in neonatal and pediatric critical care
medicine, the number of children suffering from
sepsis continues to rise, and sepsis remains the
leading cause of death in children worldwide. A
retrospective review of observational cohort data-
sets from 1995, 2000, and 2005 demonstrated an
81% increase in cases of severe sepsis in patients
aged 19  years or younger over the ten-year
period. Notably, the prevalence of sepsis is sig-
nificantly higher in infants and newborns than in
older children.
The most common pathogens identified in the
pediatric and adult populations include Esche-
richia coli, Pseudomonas aeruginosa, Klebsiella,
and Bacteroides species. The most common organ-
isms found in neonates within the first 72 hours of
life are group B streptococci and E. coli.
L. H. Vaughn
Department of General Surgery, Vanderbilt University
Medical Center, Nashville, CA, USA
e-mail: lexie.h.vaughn.1@vumc.org
J. S. Upperman (*)
Department of Pediatric Surgery, Monroe Carell Jr.
Children’s Hospital at Vanderbilt,
Nashville, CA, USA
e-mail: jeffrey.upperman@vumc.org
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_8
8
In the adult population, sepsis is defined as
life-threatening organ dysfunction in a patient
with a suspected infection. Former criteria for
defining the clinical signs and symptoms of the
systemic inflammatory response syndrome
(SIRS) focused on the inflammatory response of
the host and defined a continuum of clinical pro-
gression from SIRS to sepsis to shock. This
definition has fallen out of favor since 2016,
with the publication of Third International
Consensus Definitions for Sepsis and Septic
Shock (Sepsis-3). SIRS criteria have now been
replaced in the clinical setting by the Sequential
[Sepsis-­Related] Organ Failure Assessment
(SOFA) as a metric of organ dysfunction. This
new scoring system accompanies a new defini-
tion of sepsis, which emphasizes the dysregu-
lated host response to infection as the primary
cause of sepsis and organ dysfunction in a patient
with a suspected infection. The predictive valid-
ity of in-hospital mortality for adult ICU patients
has been demonstrated to be significantly greater
for SOFA than the former SIRS criteria. Despite
these reassuring results, the SOFA score and
updated definition of sepsis were developed
based on adult patients only and do not take into
account differences in baseline physiology and
responses to insult in children and neonates.
In children and neonates, sepsis is still defined
based on SIRS criteria as a triad of fever, tachy-
cardia, and vasodilation with a change in mental
status or prolonged capillary refill >2  seconds
85
86
(Alaedeen et al. 2006). There has not yet been a
generalized transition to scoring systems for
organ dysfunction in the clinical setting as seen
in the adult population; however, some predictive
scoring systems do exist for the pediatric popula-
tion. Leclerc et  al. reported excellent predictive
validity of in-hospital mortality in pediatric ICU
patients using the Pediatric Logistic Organ
Dysfunction-2 (PELOD-2). This tool was origi-
nally proposed in 2005 and updated in 2013, and
it incorporates Glasgow coma score (GCS),
­systolic blood pressure (SBP), mean arterial pres-
sure (MAP), and heart rate (HR). While there has
been no generalized consensus regarding the use
of this tool, much like the SOFA score in adults,
it shows promise as a diagnostic tool for pediatric
sepsis.
Children with sepsis and progressive septic
shock present a unique set of challenges for cli-
nicians. The diagnosis and management, spe-
cifically the initial fluid resuscitation and
subsequent hemodynamic support, are different
than in the adult population due to differences in
baseline physiology and changes that occur as
children age.
8.2 Risk Factors
8.2.1 Barriers to Infection
The human body is colonized by a variety of non-
pathogenic microorganisms. These normal florae
adhere to the epithelial lining and prevent the
attachment of other pathogenic microbes.
Additional protective mechanisms, such as intes-
tinal peristalsis, gastric acid secretion, and immu-
noglobulins, help to limit pathologic microbial
invasion (Table  8.1). Oropharyngeal, nasopha-
ryngeal, tracheobronchial, and gastrointestinal
secretions are rich in immunoglobulins, which
help to prevent bacterial attachment to the epithe-
lium. Specifically, immunoglobulin A (IgA)
binds microorganisms at the epithelial surface,
thereby impairing attachment to the epithelial
lining—a critical step in the establishment of
infections.
L. H. Vaughn and J. S. Upperman
Table 8.1 Defense mechanisms against microbial
invasion
Host defense Actions
Gastric acid Lower pH promotes a hostile
environment for bacterial growth
Peristalsis Coordinated movements sweep the
bacteria downstream, limiting
attachment to the mucosa
Local flora Indigenous microbial flora prevents
the overgrowth of pathogenic
Gram-negative bacteria
Immunoglobulin IgA coats and aggregates bacteria
and prevents their attachment to
intestinal mucosa
Mucus Intestinal mucus forms a thick
barrier, which prevents bacterial
attachment
Any breach in the mucosal barrier permits
bacteria or viruses to infiltrate the epithelial lin-
ing and elicit an inflammatory response. For
instance, trauma, surgery, malnutrition, burns,
immunosuppression, shock, and reperfusion
injury following an ischemic event can cause gut
barrier failure. Following reperfusion injury,
inflammatory cells elaborate toxic reactive oxy-
gen species, such as superoxide (O2−) and hydro-
gen peroxide (H2O2), which damage the epithelial
lining and permit the translocation and internal-
ization of microbes.
8.3 Pathophysiology of Sepsis
Figure 8.1 summarizes the pathogenesis of the
systemic inflammatory response syndrome
(SIRS) and the immune components involved.
8.3.1 Bacterial Virulence
The most common pathogens identified in chil-
dren are bacterial in origin. Establishment of a
clinical bacterial infection in the host begins with
the attachment of the microbe to the epithelial
surface of the host, followed by subsequent inter-
nalization. The internalized microbe must then
evade the local cellular and humoral host defense
mechanisms in order to cause infection, multiply,
8 Sepsis 87

Pathogenesis of systemic inflammatory response syndrome (SIRS)

Barrier Failure

Bacterial Invasion

LPS release

LPS-LBP

Cellular
activation

Neutrophil Macrophage Endothelial cell

Inflammatory IL-1, IL-6, TNFα IL-1, IL-6, Il-8, IL-1, IL-8, PAF
cytokines TNFα, PAF

Adhesion L-Selectin B1, B2 integrin E, P-Selectin

molecules B2 integrin ICAM-1, PECAM-1

Cellular effect Rolling, Adhesion, Chemotaxis Adhesion, Clotting

Transmigration, cascade, Activation
Chemotaxis

ROI

TLR4

CD14 Bacterial killing

Distant organ injury
SIRS

Fig. 8.1 Pathogenesis of systemic inflammatory
response syndrome (SIRS). Bacterial invasion secondary
to barrier failure leads to the local release of lipopolysac-
charide (LPS), with consequent formation of an
LPS–lipopolysaccharide-binding protein (LBP)–CD14–
Toll-like receptor 4 (TLR4) complex on neutrophils,
macrophages, and endothelial cells, resulting in cellular
activation. Inflammatory cytokines are released, upregu-
late adhesion molecules, and promote chemotaxis of neu-
trophils and macrophages. The activated cells release
microbicidal agents typically designed for bacterial kill-
ing, but they may be injurious and promote distant organ
injury and SIRS if the inflammatory process is “uncon-
trolled.” ICAM intercellular adhesion molecule, IL inter-
leukin, MCP monocyte chemotactic protein, MIP
macrophage inflammatory protein, NO nitric oxide, PAF
platelet-activating factor, PECAM platelet-­endothelial
cell adhesion molecule, ROI reactive oxygen intermedi-
ate (or species), TNF tumor necrosis

damage local tissue, and elicit an inflammatory
response. This is a process that depends largely
on microbial virulence factors.
The process of bacterial adherence requires
interaction between specific cell surface recep-
tors on the host and key molecules on the patho-
gen, called adhesins. Bacterial fimbriae or pili are
known to promote bacterial adherence to muco-
sal surfaces. E. coli expresses different types of
fimbriae that permit their attachment to the
D-mannose receptor on epithelial cells. Some
microbes also display adhesins that facilitate
entry into the host. Indirectly, the host secretes
proteins that have a common peptide sequence
Arg-Gly-Asp, such as fibronectin, laminin, col-
lagen, and vitronectin, which enhance bacterial
attachment to the host.
Once the microbe has attached to the cell sur-
face, the organism may gain entry into the cell
through a process called internalization. This
requires high-affinity binding between the
microbe’s pili and cell surface receptors. The cell
88 L. H. Vaughn and J. S. Upperman

surface contains a receptor called integrin, which organ dysfunction, specifically in the neonatal
binds the bacterial pili. The affinity of the pili for population. Several mechanisms for this
this receptor determines whether the microbe increased risk have been proposed, including
attaches to the cell and becomes internalized. selective proliferation of pathologic bacteria sec-
Bacterial internalization takes place through ondary to antibiotic use, proinflammatory host
phagocytosis. The internalized bacteria are trans- immune response, and decreased production of
ported in intracellular vesicles known as endo- short-chain fatty acids. Changes in the microbi-
somes or phagosomes. ome during critical illness may also impact the
Once the bacteria have evaded the initial host host response, increase the risk of end-organ dys-
defense mechanisms and entered the cell, they function, and ultimately impact the clinical
must survive within the intracellular milieu to course of a patient with sepsis. With increasing
establish an infection. Fusion of the cell’s lyso- accessibility of culture-independent methods for
some with the phagosome leads to acidification microbial identification, there are potential thera-
of the phagolysosome complex and neutraliza- peutic molecular targets for patients with sepsis;
tion of the internalized bacteria by specific toxins however, prospective data in humans are cur-
such as hyaluronidase, collagenase, proteinase, rently limited, and additional investigation is
deoxyribonuclease, and lecithinase. The bacteria needed.
may counterattack by secreting exotoxins to help
neutralize the host defense mechanisms. For
instance, S. aureus produces catalase, which neu- 8.3.3 Neutrophils
tralizes hydrogen peroxide. Streptolysin, a strep-
tococcal exotoxin, can inhibit neutrophil Neutrophils are terminally differentiated effector
migration and impair phagocyte cytotoxicity. cells that constitute the first line of defense in
One of the most potent bacterial toxins is called response to infection or tissue injury. The neutro-
lipopolysaccharide (LPS) (or endotoxin). It con- phil contains proteolytic enzymes and reactive
tains an O-specific side chain, a core polysaccha- oxygen species that can cause local tissue dam-
ride, and an inner lipid A region. The lipid A age when released into the extracellular matrix.
region is a highly potent stimulator of the inflam- After a 14-day development in the bone marrow,
matory response. This molecule may initiate sep- the neutrophils circulate in the bloodstream for
tic shock by stimulating the release of 6–14 h. Nearly 50% of the circulating neutrophils
inflammatory mediators such as arachidonic acid attach or adhere to the vascular endothelium—a
and leukotrienes, or through complement activa- process known as margination. If there are no
tion. Endotoxin, alone, is sufficient to induce detectable infections, the neutrophils undergo
shock when given experimentally to laboratory apoptosis, or programmed cell death, in the liver
animals or to human volunteers. or the bone marrow. The neutrophils that adhere
to the vascular endothelium must leave the blood-
stream through a process known as diapedesis to
8.3.2 Microbiome reach the tissues. There, they can survive for
another 48  h performing critical functions such
The human microbiome and specifically the as phagocytosis and microbial killing. Adhesion
native microorganisms found in the human intes- molecules, such as selectins, integrins, and the
tine play a significant role in immune regulation immunoglobulin superfamily, govern the adher-
and in the pathogenesis of some diseases. The ence of neutrophils to the vascular endothelium.
intestine is sterile in utero but becomes colonized (L)-selectin (CD62L) on the neutrophil surface
at birth and diversifies quickly and significantly binds to endothelium (E)-selectin and platelet
in the neonatal period. Recent evidence has sug- (P)-selectin, which is upregulated when the endo-
gested that disruption in the microbiome may thelial cells are activated by injury, infectious
increase the risk of sepsis and progression to end-­ agents, or inflammatory mediators. Migration of
8 Sepsis
the neutrophil to the site of injury is regulated by
a class of molecules known as integrins, which
are expressed on the neutrophil surface.
Specifically, binding of β2 integrin to intercellular
adhesion molecule 1 (ICAM-1) on the endothe-
lial cell directs neutrophil traffic.
LPS can affect neutrophil adhesion and migra-
tion by stimulating the release of tumor necrosis
factor-α (TNF-α), interleukin-1 (IL-1), and
interferon-γ (IFN-γ), which are known to upregu-
late ICAM-1 and E-selectin. Lipopolysaccharide
binding protein (LBP), a 58-kDa acute phase
reactant that is synthesized in the liver, enhances
the sensitivity of monocytes and granulocytes to
LPS by facilitating binding of LPS to the CD14
cell membrane molecule and to Toll-like receptor
4 (TLR-4) on the surface of neutrophils and
monocytes. This interaction upregulates β2 integ-
rin CD11b/CD18 and enhances the neutrophil–
endothelial interaction. Clinically, patients with
leukocyte adhesion deficiency are susceptible to
recurrent bacterial infections due to the lack of β2
integrin receptor CD11b/CD18, which results in
the inability of the neutrophil to adhere to the
endothelium and effect bacterial killing.
Migration of the neutrophil to the site of tissue
injury is governed, in part, by platelet endothelial
cell adhesion molecule 1 (PECAM-1), CD99,
and other adhesion molecules. These molecules
are expressed on the surface of blood vessels and
maintain the vascular permeability barrier.
Evidence suggests that an antibody to PECAM-1
inhibits neutrophil transmigration and endotoxin-­
induced leukocyte sequestration in the lung, liver,
and muscle. Neutrophil egress requires a chemo-
tactic gradient through the extracellular matrix.
Important chemotactic peptides include mono-
cyte chemotactic protein 1 (MCP-1), platelet-­
activating factor (PAF), leukotriene B4, and
interleukin 8. Only small amounts of chemotaxis
are required for the neutrophil to become respon-
sive. A cascade of intracellular signaling path-
ways is activated when the neutrophil binds to the
endothelium. These events eventually lead to
conformational changes in the cytoskeleton of
the neutrophil and permit transendothelial egress
and rapid movement along the chemotactic
gradient.
89
The neutrophil’s primary objective is to
destroy the microorganism, which is achieved
through phagocytosis followed by intracellular
killing. Specific immunoglobulins, such as IgG,
enhance the phagocytic activity of the neutrophil
and stimulate complement activation. The fusion
of the phagosome with the lysosome, which con-
tains powerful antimicrobial agents, aids in the
killing of the microbe.
Release of reactive oxygen intermediates,
formed by the enzyme NADPH, is the principal
oxygen-dependent mechanism involved in the
killing of microbes in the lysosome. In the neu-
trophil, the respiratory burst catalyzes the reduc-
tion of molecular oxygen (O2) to superoxide
(O 2̄), which is subsequently converted to hydro-
gen peroxide (H2O2) by superoxide dismutase.
Hydrogen peroxide can form a hydroxyl radical
in the presence of iron or other metals and can
also form hypochlorous acid (HOCl) in the pres-
ence of myeloperoxidase. HOCl is the chemical
that accounts for the cytotoxicity of the neutro-
phil in the presence of nitrogen-containing com-
pounds. Enzymes such as lysozyme, elastase,
lactoferrin, cathepsin, and defensins within the
phagolysosome act synergistically to promote
microbial killing.
8.3.4 Monocytes–Macrophages
The monocyte–macrophage also plays an impor-
tant role in the response to microbial infection.
There are many similarities between the neutro-
phil and monocyte–macrophage complex. Both
phagocytose and use lysosomes to kill the patho-
gen. Both produce reactive oxygen intermediates
on stimulation by LPS and IFN-γ. The monocyte
evolves from a precursor (promonocyte) in the
bone marrow, which undergoes maturation by
acquiring specific granules. The monocyte then
migrates to various tissues and organs where it
further differentiates into macrophages. Tissue
macrophages are the principal effectors in the
defense against intracellular pathogens. They can
phagocytose and destroy many common bacteria,
but with less efficiency than the neutrophil.
Macrophages express adhesion molecules such
90
as L-selectin and β1 and β2 integrins. This distinc-
tion is important since macrophages can still
migrate to the site of inflammation in patients
with leukocyte adhesion deficiency (lack of β2
integrin). An important difference between the
neutrophil and the macrophage is that the macro-
phage, after engulfing the bacteria, can present
the antigenic fragments to the T lymphocytes in
the context of major histocompatibility complex
(MHC) class II molecules. This enhances the
release of inflammatory cytokines and the micro-
bicidal activity of the macrophage.
Like neutrophils, macrophages produce reac-
tive oxygen species; however, they also produce
nitric oxide (NO), which has diverse biological
properties. NO is the product of the conversion of
arginine to citrulline by nitric oxide synthase
(NOS). Three isoforms of NOS exist: neuronal
NOS (NOS-1) and endothelial NOS (NOS-3) are
expressed constitutively. Inducible NOS (NOS-2,
or iNOS) found in the macrophage is activated in
response to inflammatory mediators. NO is rela-
tively innocuous but can react with reactive oxy-
gen species to form cytotoxic molecules. For
instance, peroxynitrite is an important reactive
nitrogen intermediate that is formed by the reac-
tion of NO with O 2̄ in inflammatory lesions
in  vivo and is responsible for the cytopathic
effects of NO. NO may also react with metallo-
proteins to form S-nitrosothiols. Sustained over-
production of these compounds may lead to
cellular injury and multisystem organ
dysfunction.
8.3.5 Lymphocytes
Lymphocytes and natural killer cells are the pre-
dominant effector cells against intracellular
organisms. Lymphocytes originate from the bone
marrow; however, some leave the bone marrow
to undergo maturation in the thymus. Once
mature, T lymphocytes migrate to peripheral
lymphoid organs such as the spleen, lymph
nodes, and the Peyer’s patches in the intestine,
where they establish residence. Other lympho-
cytes mature in the bone marrow and become B
L. H. Vaughn and J. S. Upperman
cells, which produce immunoglobulins. The main
job of T lymphocytes is to regulate cell-mediated
immunity against intracellular pathogens. This
requires recognition of the inciting antigen by
MHC class II proteins, cellular activation, clonal
expansion, and targeted killing. The MHC pro-
teins on cell surfaces govern antigen presenta-
tion. Macrophages, dendritic cells, and
B-lymphocytes can act as antigen-presenting
cells. These cells phagocytose the microbe and
digest it into smaller fragments or peptides that
are then bound to the MHC class II proteins and
then presented to T helper cells. In addition, any
cell that is infected can present microbial anti-
gens on its cell surface using MHC class I mole-
cules. CD8+ cytotoxic T lymphocytes then target
these cells and release serine proteases to induce
apoptosis.
8.3.6 Immunoglobulins
Immunoglobulins, or antibodies, represent a
class of proteins that are synthesized from mature
B-lymphocytes or plasma cells. The primary role
of antibodies is to prevent microbial attachment
to, or invasion of, the host epithelium. There are
five major classes of immunoglobulins: IgA, IgG,
IgM, IgD, and IgE.  The predominant immuno-
globulins are IgG, IgM, and IgA.
IgM, with its short half-life of 5 days, initiates
the first response to an infection in the blood-
stream. The levels of IgM then start to decrease
while the levels IgG begin to increase. IgG, which
is directed against bacteria and viruses, consti-
tutes 85% of serum immunoglobulins found in
the intravascular and extravascular compart-
ments. The biologic potency of this protein
resides in its ability to opsonize bacteria by bind-
ing the antigen to the neutrophil, monocyte, or
macrophage.
Mucosal immunity is governed by IgA,
which is synthesized by plasma cells within
lymphoid tissue adjacent to the epithelial sur-
face. Once secreted, IgA binds pathogenic
microbes and prevents their attachment to the
epithelial surface.
8 Sepsis 91

8.3.7 Cytokines cytotoxicity due to the lack of prior antigenic

exposure and a deficiency in cytokine produc-
Cytokines are glycoproteins that orchestrate the tion. Immunoglobulin M is more abundant in
interactions of immune cells with bacteria. Most neonatal secretions, and the differentiation of B
of the immune cells secrete these proteins in cells into IgA- or IgG-producing plasma cells
response to an inflammatory or antigenic stimu- does not occur for months after birth. The term
lus. The proinflammatory cytokines include neonate relies on the maternal transfer of IgG
TNF-α, IL-1, IL-6, IL-8, IL-11, IL-17, and IL-18. across the placenta during the third trimester and
The earliest inflammatory cytokine to arrive at on the mother’s breast milk, which is rich in IgG
the site of injury is TNF-α. The principal anti-­ and IgA. By the fourth month of life, the neonate
inflammatory cytokines are IL-10, IL-27, IL-33, begins to increase production of IgG and the
and transforming growth factor-β (ΤGF-β), which maternal IgG dissipates. As a result, neonates
help neutralize or modulate the production of exhibit increased susceptibility to infections dur-
inflammatory products from monocytes–macro- ing the first four months of life.
phages. Recent data suggest that IL-17, IL-27, In addition to a deficiency in cell-mediated
and IL-33 may play a novel and synergistic role and humoral immunity, the neonate is also at risk
in immune dysfunction and host response to sep- for infection due to immaturity of the antibody-­
sis and may be reasonable biomarkers for sepsis. independent complement system. The levels of
key components of the complement system are
decreased, which leads to a diminished capacity
8.4 Neonates to fight off Gram-negative microbes (C9),
decreased production of chemotactic factor C5a,
In the United States, approximately half of the and a decrease in functional opsonins due to a
cases of sepsis in the pediatric population occur in lack of efficient cross-linking (C3b).
neonates, specifically those born at low birth-
weights. Death or serious disability occurs in 40%
of neonates with sepsis, versus a 10–20% mortal- 8.5 Clinical Features
ity rate in older children. Neonates are predisposed and Diagnosis
to bacterial infections secondary to an immature
cellular and humoral (antibody-­mediated) immune The clinical presentation of sepsis in children and
system. The pool of neutrophils in the neonate rep- neonates may be subtle, and the diagnosis is
resents only 20–30% of the total adult pool and nuanced in this population. While in the adult
consists of 60% circulating leukocytes and 15% population, defining sepsis has moved toward
immature bands. Neonates also have a limited clinical models for end-organ dysfunction (e.g.,
ability to increase the pool of circulating neutro- SOFA), according to the 2005 International
phils in response to infections. This predisposes Consensus Conference on Pediatric Sepsis, pedi-
the neonate to severe neutropenia because it takes atric sepsis is still defined based on the presence
at least 5–7  days to increase the formation of of modified SIRS criteria. These criteria consider
myeloid progenitor stem cells (precursors of neu- the specific physiologic changes that occur as
trophils) in response to infections. Neutrophils in children age and incorporate guidelines based on
the neonate also demonstrate decreased adhesion six age groups: newborns (0–7  days), neonates
to activated endothelium, less efficient phagocyto- (7  days–1  month), infants (1–12  months), tod-
sis secondary to deficiency in opsonins, and dlers (2–5 years), children (6–12 years), and ado-
decreased ability to kill phagocytosed microbes by lescents (12–18  years). Infection may be a
oxygen-dependent mechanisms. diagnosis of pediatric sepsis that requires pres-
T-cell-mediated immunity is also different in ence or suspicion of an infection, secondary to
the neonate compared to older children and viral, bacterial, fungal, or rickettsial pathogens,
adults. There is a decrease in T-cell-mediated and at least two of the following criteria:
92
1. Fever (core temperature > 38.5 °C) or hypo-
thermia (<36 °C).
2. Tachycardia (heart rate [HR]  >  2 standard
deviations above normal for age OR bradycar-
dia (HR < tenth percentile for age) in patients
1 year or greater.
3. Tachypnea (mean respiratory rate [RR]  >  2
standard deviations above normal for age) OR
mechanical ventilation (MV) for an acute pro-
cess unrelated to neuromuscular disease or
general anesthesia.
4. Leukocytosis or leukopenia OR 10% imma-
ture neutrophils.
Severe sepsis is defined as sepsis with cardio-
vascular dysfunction, acute respiratory failure, or
dysfunction in two or more other organs. Septic
shock occurs when the cardiovascular dysfunc-
tion is nonresponsive to isotonic fluid administra-
tion in the first hour. Notably, this patient
population often maintains normotension despite
progressive or worsening sepsis and thus hypo-
tension is not required for the diagnosis of septic
shock in children and neonates. Alternatively,
tachycardia and signs of hypoperfusion, includ-
ing increased capillary refill >2 seconds, altered
mental status, decreased peripheral pulses, and
decreased urine output, are more often diagnostic
of shock in this population.
There are several significant changes that
occur in the baseline physiology and response to
injury or infection as children age. The modified
SIRS criteria incorporate these changes and help
guide the physician in diagnosis. For example,
the diagnosis of pediatric SIRS requires the pres-
ence of either fever or leukocytosis but not both.
Less than 50% of neonates with white blood cell
counts greater than 20,000/mm3 or less than
4000/mm3 are ultimately diagnosed with an
infection, and the presence of fever or hypother-
mia may be a more sensitive finding. In addition
to the criteria listed above, the ratio of immature
to total neutrophils may be useful in determining
the likelihood of infection in the neonate. A ratio
greater than 0.2 is a sensitive indicator of infec-
tion. Thrombocytopenia is a nonspecific and late
predictor of neonatal sepsis, but trends in platelet
count are important for the clinician to address.
L. H. Vaughn and J. S. Upperman
Similarly, children may maintain normotension
until they have progressed to significant cardio-
vascular collapse. Thus, the modified SIRS crite-
ria allow for the diagnosis of SIRS, sepsis, and
shock in the presence of normotension. In the
neonatal population specifically, hypoxemia, pul-
monary hypertension, and cardiac collapse sec-
ondary to increased pulmonary vascular
resistance during the transition from fetal to neo-
natal circulation further complicate the diagnosis
and management of sepsis.
The most prominent clinical or physiologic
change that occurs in SIRS, sepsis, and shock is
reduced systemic vascular resistance (SVR),
resulting in decreased peripheral extraction of
oxygen. In response to this so-called “warm
shock,” the pediatric patient must increase car-
diac output and increase minute ventilation to
achieve a higher delivery of oxygen to ischemic
tissues. An inadequate increase in oxygen deliv-
ery causes marked anaerobic metabolism and
production of excess lactate. Increased lactate on
hospital admission has been associated with
increased risk for ICU admission, end-organ dys-
function, and mortality.
Notably, up to 50% of pediatric patients pres-
ent with increased SVR or “cold shock,” a phe-
nomenon that is not observed in the adult
population. A higher resting heart rate may lead
to inadequate diastolic filling and resultant
decreased cardiac output (CO). To compensate,
children will initially have an increase in periph-
eral vasoconstriction, and hypotension will not
occur until much later in the clinical course.
Similarly, production of excess lactate may also
be a rather late sign of sepsis and indicate a worse
prognosis in children.
The 2001 International Sepsis Definitions
Conference proposed a purely biochemical or
immunological, rather than clinical, criteria to
identify the inflammatory response in children.
C-reactive protein (CRP), interleukin 6 (IL-6),
and procalcitonin (PCT) were mentioned as
potential markers for use in neonates and pediat-
ric patients. PCT is secreted during Gram-­
negative sepsis from an unknown extrathyroidal
source and may be used as a guide for antibiotic
therapy. Maja Pavcnik-Arnol et  al. performed a
8 Sepsis
prospective observational study investigating
potential biochemical markers that could diag-
nose bacterial sepsis in the neonate. In critically
ill neonates less than 48  hours old, LBP was a
better marker of sepsis on the first day of sus-
pected infection than IL-6 and PCT. In critically
ill neonates older than 48  hours and older chil-
dren, LBP was a better marker than IL-6 and
CRP. In culture-confirmed sepsis, LBP had 91%
sensitivity, 98% negative predictive value, 85%
specificity, and 52% positive predictive value.
When there is suspected infection, cultures
should be obtained from peripheral blood as well
as any indwelling intravenous lines. Peripheral
cultures should be obtained from at least two dif-
ferent venipuncture sites if possible. Once blood-
stream infection is identified, repeat or follow-up
cultures are not necessary in most cases.
Subsequent blood cultures may be justified in
patients who deteriorate clinically or who fail to
improve despite appropriate antibiotic therapy. In
some cases, bacteremia may be prolonged, neces-
sitating further blood cultures during treatment.
Urine cultures as well as lumbar puncture should
be performed during the initial workup.
8.6 Management
Pediatric sepsis and septic shock management
follow closely the guidelines of Pediatric
Advanced Life Support (PALS). Upon recogni-
tion of end-organ dysfunction, the physician
should initiate high flow O2 support and establish
intravenous or intraosseous access. The most
prominent feature of pediatric SIRS, sepsis, and
potential progression to septic shock is the
increase in oxygen demand by end organs and a
decrease in peripheral vascular resistance, which
is manifested as a low blood pressure. In the
treatment of sepsis, the goal is to increase oxygen
delivery via aggressive fluid resuscitation, car-
diovascular and respiratory support, and optimiz-
ing electrolytes, as well as hematologic, renal,
metabolic, and nutritional needs. Initial resuscita-
tion should include boluses of 20 mL/kg of iso-
tonic saline up to 60 mL/kg in the first hour. Fluid
boluses are continued until perfusion is improved.
93
The physician should monitor for hepatomegaly,
rales or crackles on auscultation, or other signs of
volume overload during resuscitation. Fluid
resuscitation of more than 40 mL/kg in the first
hour following emergency department presenta-
tion is associated with improved survival and
decreased occurrence of persistent hypovolemia
and does not increase the risk of cardiogenic pul-
monary edema. Delay in resuscitation as little as
30 min is associated with increased mortality in
children over 2 years old.
If there is persistent hypoperfusion after iso-
tonic fluid resuscitation, vasopressors should be
initiated. The goal of vasopressor support is to
maintain cardiac index between 3.3 and 6 L/min/
m2 and SVC oxygen saturation at 70%. Dopamine,
dobutamine, and epinephrine are options for first-­
line vasopressor support. A 2015 double-blind,
prospective randomized control trial demon-
strated increased mortality risk with the use of
dopamine when compared to epinephrine.
Norepinephrine is used specifically in the pres-
ence of “warm shock” or hypotension with
decreased SVR.  Hydrocortisone administration
can be considered in patients with vasoactive-­
resistant shock; however, prospective data are
limited on this therapy.
Fluid resuscitation should continue, and the
physician should monitor the urine output as a
guide to end-organ perfusion and adjust fluid
management accordingly (goal UOP 1–2  ml/
kg/h). In addition, the use of mechanical ventila-
tion may help relieve failing respiratory muscles.
Adjustment to the respiratory rate or to the oxy-
gen concentration on the ventilator is governed
by arterial blood gas.
Source control with antimicrobial therapy is
the cornerstone of the treatment of presumed sep-
sis or SIRS.  Broad-spectrum antibiotics should
be started within 1 hour of the first signs of sep-
sis, and source control should occur rapidly
thereafter. Empiric antibiotic therapy should
include coverage of both Gram-negative and
Gram-positive organisms. Most children with
sepsis or SIRS are diagnosed with Gram-negative
bacterial infections from E. coli, P. aeruginosa,
Klebsiella, and Bacteroides species. In neonates
or term infants, the most common organisms
94
encountered are group B streptococcus, E. coli,
and L. monocytogenes. Although culture is not
required for the diagnosis of SIRS, sepsis, or sep-
tic shock, antibiotics should be tailored based on
available culture data, and consultation with
expert pediatric infectious disease specialists
may be indicated to guide narrowing therapy or
for multidrug-resistant infections. Empiric anti-
fungal therapy may be indicated in patients with
persistent fever or leukocytosis despite empiric
antibiotics and unknown sources.
During systemic Gram-negative and Gram-­
positive bacterial infections, activation of the
coagulation cascade is mediated by the extrinsic
tissue factor pathway. Activated protein C is an
endogenous regulator of coagulation and
inflammation and is a promising therapeutic
target in patients with severe sepsis. The
PROWESS (Protein C Worldwide Evaluation in
Severe Sepsis) study was a large multicenter
randomized, double-blind, placebo-controlled
trial in adult patients with severe sepsis. The
trial demonstrated a decrease in the 28-day
mortality from all causes in adults with sepsis
treated with recombinant human activated pro-
tein C. Additional prospective studies are needed
to determine if activated protein C is a viable
option in the pediatric population.
Sepsis and SIRS cause a release of inflamma-
tory cytokines and hormones that lead to hyper-
glycemia. Increased peripheral insulin resistance
is caused by a release of cortisol, TNF-α, and
IL-1. In addition, there is an increase in hepatic
glucose production, which causes hyperglyce-
mia. Van den Berghe and colleagues demon-
strated that tight glycemic control with a blood
glucose level of 80–110  mg/dl decreases in-­
hospital mortality by 34% in a mixed medical-­
surgical ICU (predominantly adult cardiac
surgery patients) compared to patients with a tar-
geted blood glucose level of 180–200 mg/dl. In
septic neonates and children, hyperglycemia cor-
relates with prolonged ventilator dependency and
increased hospital length of stay.
Nutritional support during sepsis is para-
mount. The advantages of enteral feeds when
compared to parenteral nutrition include buffer-
L. H. Vaughn and J. S. Upperman
ing gastric pH, avoiding the use of centrally
placed catheters, preserving of gut mucosa, limit-
ing the introduction of bacteria and toxins from
the gastrointestinal tract into the circulation, and
preserving a more physiologic pattern of enteric
hormone secretion.
Transfer to a pediatric tertiary care center
should be initiated for any pediatric patient
requiring vasoactive support. However, initial
resuscitation should not be delayed to accommo-
date transfer.
8.7 Outcomes
Overall mortality from sepsis has decreased sig-
nificantly in the pediatric population in the past
40 years, with currently reported mortality rang-
ing from 10% to 20%. However, sepsis remains
the leading cause of death in children world-
wide, and the incidence continues to increase
due to better diagnostic tools and understand-
ing. Outcomes are impacted by time to resusci-
tation and clinical characteristics at the time of
presentation. For every hour without adequate
resuscitation, mortality increases by 40%. Goal-
directed fluid resuscitation and vasoactive sup-
port in the first 72 hours reduce overall mortality.
Furthermore, if hypotension is adequately
reversed in the emergency room, mortality is
reduced twofold.
8.8 Conclusion
Despite advances in the diagnosis and manage-
ment of sepsis, it remains a major cause of death
in children worldwide. Most commonly observed
secondary to bacterial infection, the development
of sepsis is influenced by the host defense and
immune system as well as virulence of the patho-
gen. In children, sepsis is diagnosed based on a
modified SIRS criteria, which includes fever or
leukocytosis with tachycardia or tachypnea.
Management of sepsis follows PALS guidelines
for resuscitation and prioritizes delivery of oxy-
gen therapy as well as fluid resuscitation and
8 Sepsis
early administration of broad-spectrum antibiot-
ics. Outcomes in children with sepsis are
impacted greatly by time to appropriate fluid
resuscitation and antibiotic administration, and
overall mortality ranges from 10% to 20% in this
population.
Reference
Alaedeen DI, Walsh MC, Chwals WJ (2006) Total par-
enteral nutrition-associated hyperglycemia correlates
with prolonged mechanical ventilation and hospital
stay in septic infants. J Pediatr Surg 41(1):239–244,
discussion 239–244
Further Reading
Adelman MW, Woodworth MH, Langelier C et al (2020)
The gut microbiome’s role in the development, main-
tenance, and outcomes of sepsis. Crit Care 24(1):278.
(In eng). https://doi.org/10.1186/s13054-­020-­02989-­1
Bernard GR, Vincent JL, Laterre PF et  al (2001)
Efficacy and safety of recombinant human acti-
vated protein C for severe sepsis. N Engl J Med
344(10):699–709. (In eng). https://doi.org/10.1056/
NEJM200103083441001
Brown KA, Brain SD, Pearson JD, Edgeworth JD, Lewis
SM, Treacher DF (2006) Neutrophils in develop-
ment of multiple organ failure in sepsis. Lancet
368(9530):157–169
Dahmer MK, Randolph A, Vitali S, Quasney MW (2005)
Genetic polymorphisms in sepsis. Pediatr Crit Care
Med 6(3 Suppl):S61–S73
Emr BM, Alcamo AM, Carcillo JA, Aneja RK, Mollen
KP (2018) Pediatric sepsis update: how are children
95
different? Surg Infect 19(2):176–183. (In eng). https://
doi.org/10.1089/sur.2017.316
Hartman ME, Linde-Zwirble WT, Angus DC, Watson RS
(2013) Trends in the epidemiology of pediatric severe
sepsis*. Pediatr Crit Care Med 14(7):686–693. (In
eng). https://doi.org/10.1097/PCC.0b013e3182917fad
Leclerc F, Duhamel A, Deken V, Grandbastien B,
Leteurtre S, (GFRUP) GFdReUP (2017) Can the pedi-
atric logistic organ dysfunction-2 score on day 1 be
used in clinical criteria for sepsis in children? Pediatr
Crit Care Med 18(8):758–763. (In eng). https://doi.
org/10.1097/PCC.0000000000001182
Martin GS, Mannino DM, Eaton S, Moss M (2003) The
epidemiology of sepsis in the United States from 1979
through 2000. N Engl J Med 348(16):1546–1554
Morrow KN, Coopersmith CM, Ford ML (2019) IL-17,
IL-27, and IL-33: a novel axis linked to immunologi-
cal dysfunction during sepsis. Front Immunol 10:1982.
(In eng). https://doi.org/10.3389/fimmu.2019.01982
Pavcnik-Arnol M, Hojker S, Derganc M (2004)
Lipopolysaccharide-binding protein in critically
ill neonates and children with suspected infec-
tion: comparison with pro-calcitonin, interleukin-
6, and C-reactive protein. Intensive Care Med
­
30(7):1454–1460
Seymour CW, Liu VX, Iwashyna TJ et  al (2016)
Assessment of clinical criteria for sepsis: for the third
international consensus definitions for sepsis and sep-
tic shock (sepsis-3). JAMA 315(8):762–774. (In eng).
https://doi.org/10.1001/jama.2016.0288
Ventura AM, Shieh HH, Bousso A, et al (2015) Double-­
blind prospective randomized controlled trial of
dopamine versus epinephrine as first-line vasoac-
tive drugs in pediatric septic shock. Crit Care Med
43(11):2292–2302. (In eng). https://doi.org/10.1097/
CCM.0000000000001260
Watson RS, Carcillo JA, Linde-Zwirble WT, Clermont
G, Lidicker J, Angus DC (2003) The epidemiology of
severe sepsis in children in the United States. Am J
Respir Crit Care Med 167(5):695–701
 Nutrition
Agostino Pierro and Simon Eaton
9.1 Introduction
The nutrition of surgical infants and older chil-
dren has significantly improved subsequent to the
recognition of the importance of appropriate and
tailored nutritional support. The newborn infant
is in a ‘critical epoch’ of development, not only
for the organism as a whole but also for the
individual organs and, most significantly, for
the brain. Adequate nutrition in the neonatal
period is necessary to avoid the adverse effects
of malnutrition on morbidity and mortality and
to minimise the future menace of stunted neuro-
developmental and physical development. The
survival rate of newborn infants affected by iso-
lated congenital gastrointestinal abnormalities is
now very high, with much of the mortality due to
co-morbidities, such as cardiac malformations.
The introduction of parenteral nutrition (PN) and
advancement in nutritional management are cer-
tainly among the main factors responsible for
this improvement, but now the focus has shifted
A. Pierro (*)
Division of General and Thoracic Surgery,
Translational Medicine Program, The Hospital for
Sick Children, University of Toronto,
Toronto, ON, Canada
e-mail: agostino.pierro@sickkids.ca
S. Eaton
Great Ormond Street Hospital and Institute of Child
Health, London, UK
e-mail: s.eaton@ucl.ac.uk
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_9
9
towards long-term outcomes, such as neurode-
velopment, body composition, in which nutrition
plays a key role. Similarly, in older children,
there is much more attention given to longer-
term outcomes in addition to direct surgical
management.
9.2 Historical Background
PN has progressed from numerous historical
anecdotes in the 1930s from the first subcutane-
ous infusion of saline or water to the first suc-
cessful intravenous (IV) infusion of protein
hydrolysates in humans, followed by the first
report of successful total parenteral nutrition
(TPN) in an infant in 1944, and has been given a
huge boost by the first placement of a catheter in
the superior vena cava to deliver nutrients for a
prolonged period. Using this system, Dudrick
and Wilmore showed that adequate growth and
development could be achieved in beagle puppies
and in a surgical neonate (Dudrick and Wilmore
1968; Dudrick et  al. 1967; Hartmann 1985).
Following these initial reports, Filler and co-­
authors reported the first series of surgical neo-
nates with gastrointestinal malformations treated
with long-term TPN (Morgan et al. 1970). During
the 1970s and 1980s, significant improvements
were made in the technique itself and in the
reduction of complications. The last 20  years
have seen considerable changes in the nutritional
97
98 A. Pierro and S. Eaton

management of surgical neonates. Various inves- 9.4 Energy Metabolism

tigators have highlighted the importance of intro-
ducing enteral nutrition as soon as possible after Newborn infants have a significantly higher met-
surgical procedures in adults and children. abolic rate and energy requirement per unit body
weight than children and adults (Fig.  9.1). The
total energy requirement for an extremely-low
9.3 Body Composition birthweight (i.e. <1000  g) preterm infant fed
enterally is 130–150  kcal/kg/day, and that of a
Newborn infants grow very rapidly, have lower term infant is 100–120 kcal/kg/day, compared to
caloric reserves than adults and therefore do not 60–80  kcal/kg/day for a 10-year old and
tolerate prolonged periods of starvation. The 30–40  kcal/kg/day for a 20-year old individual.
body composition of newborn infants is mark- Of the 100–120  kcal/kg/d required by the term
edly different from that of adults. The total body infant, approximately 40–70  kcal/kg/day is
water varies from 86% of body weight at needed for maintenance metabolism, 50–70 kcal/
28 weeks of gestation to 69% at 40 weeks of ges- kg/day for growth (tissue synthesis and energy
tation and 60% in adulthood (Denne et al. 2006). stored), and up to 20 kcal/kg/day to cover energy
This decline in body water also reflects an losses in excreta. Newborn infants receiving TPN
increase in the energy content of the body. The require fewer calories (90–120 kcal/kg/day for a
ratio between the minimal metabolic rate (i.e. preterm infant and 75–85 kcal/kg/day for a term
how much energy is needed on a daily basis) to infant) (Mesotten et al. 2018), due to the absence
non-protein energy reserve is only 1:2 at 28 weeks of energy losses in excreta and the fact that
of gestation. It decreases to 1:29 for term infants energy is not required for thermoregulation when
and 1:100 for adults, which explains the urgent the infant is in an incubator. These requirements
need for adequate caloric intake in very-low-­ should be reduced even further in intensive care
birthweight infants after birth. Full-term neonates units. It should also be stressed that energy
have a higher content of endogenous fat (approx- requirements vary greatly. Although adults show
imately 600  g) and therefore can tolerate a few large increases in resting energy expenditure fol-
days of undernutrition. Nevertheless, adequate lowing trauma, surgery, burns or during severe
nutrition in excess of basic requirements, i.e. infection (Hill and Hill 1998), this does not
enough to support growth, should be instituted as appear to be true for infants (Koletzko et  al.
soon as practicable. 2005), although there are few studies in this area.

Fig. 9.1  Partition of Storage / growth

energy needs in infants 160
and children
Maintenance / energy expenditure
120
kcal/kg/day

Losses (excreta etc.)

80

40

0
preterm term 10 year 20 year
neonate neonate
9 Nutrition
Resting energy expenditure is directly propor-
tional to the growth rate in healthy infants, and
growth is retarded during acute metabolic stress.
Studies in adult surgical patients have shown that
operative stress causes marked changes in protein
metabolism characterised by a post-operative
increase in protein degradation, negative nitrogen
balance, and a decrease in muscle protein synthe-
sis. However, changes in whole body protein
flux, protein synthesis, amino acid oxidation or
protein degradation do not seem to occur in
infants and young children undergoing major
operations, trauma or sepsis, so that infants and
children may divert protein and energy from
growth to tissue repair, resulting in a lack of
hypermetabolism, but also in impaired growth.
Nutritional problems in infants and children
requiring surgery are not unusual. The real
nutritional challenge is not represented by the
operation per se but by the clinical condition of
the patient. Examples include intrauterine
growth retardation in small for gestational age
preterm infants, infants who have suffered mas-
sive intestinal resection for necrotising entero-
colitis, and infants with motility disorders of the
intestine following surgery for atresia, malrota-
tion and midgut volvulus, meconium ileus or
gastroschisis. Nutritional integrity, particularly
in the neonatal period, should be maintained
regardless of the severity of the illness or organ
failure due to the limited energy and protein
stores in neonates.
9.5 Parenteral Nutrition
9.5.1 Indications
PN should be utilised when enteral feeding is
impossible, inadequate or hazardous, but should
be given for the shortest period of time possible
and the proportion of nutrition given enterally
increased as tolerated. Energy reserves are such
that stable term infants can tolerate 3–4  days
without enteral feeds before starting PN if it is
anticipated that enteral nutrition may be resumed
within this time. Premature neonates have smaller
energy reserves and the time before introducing
99
PN is much shorter. The most frequent indica-
tions in paediatric surgery are intestinal obstruc-
tion due to congenital anomalies, although
acquired conditions may require PN for variable
lengths of time. Although infants with some neo-
natal surgical conditions, such as gastroschisis,
are all likely to receive PN, there are some other
congenital anomalies where the use of PN is
more controversial, for example duodenal atre-
sia, in which many surgeons would routinely ini-
tiate PN, whereas some surgeons preferentially
manage patients without PN, either by early ini-
tiation of enteral feeds with (Hall et al. 2011) or
without (Bishay et  al. 2013) the use of trans-­
anastomotic tubes. In addition to congenital
bowel obstruction, PN may also be used in cases
of post-operative ileus, necrotising enterocolitis,
short-bowel syndrome, gastroenterological indi-
cations, and respiratory co-morbidity.
9.5.2 Components of Parenteral
Nutrition
The PN formulation includes carbohydrates, fat,
protein, electrolytes, vitamins, trace elements
and water. The caloric needs for PN are provided
by carbohydrates and lipids. Protein is not used
as a source of calories since the catabolism of
protein to produce energy is an uneconomic met-
abolic process compared to the oxidation of car-
bohydrate and fat, which produces more energy
at a lower metabolic cost. The ideal PN regimen,
therefore, should provide enough amino acids for
protein turnover and tissue growth and sufficient
calories to minimise protein oxidation for energy.
9.5.2.1 Fluid Requirements
Any newborn infant deprived of oral fluids will
lose body fluids and electrolytes in urine, stools,
sweat and evaporative losses from the lungs and
the skin. The insensible water losses from the
skin are particularly high (up to 80–100  ml/kg/
day) in low-birthweight infants. This is due to the
very large surface area relative to body weight, to
the very thin and permeable epidermis, to reduced
subcutaneous fat and to the large proportion of
total body water and extracellular water. The pre-
100
Fig. 9.2 Body 100
composition changes
with age. Percentage
body fat is not shown for 75
adults because of the
extremely wide range
% body
50
weight
25
0
24-25 27
weeks gestation
term infant requires larger amounts of fluid to
replace the high obligatory renal water excretion
due to the limited ability to concentrate urine
(Jochum et al. 2018). In surgical newborns, it is
not unusual to have significant water losses from
gastric drainage and gastrointestinal stoma. In
order to reduce the water losses, it is important to
use double-walled incubators, to place the infant
in relatively high humidity, to use warm humidi-
fied air via the endotracheal tube and in prema-
ture babies to cover the body surface with an
impermeable sheet. However, overhydration is
potentially a problem, leading to complications
such as pulmonary oedema. The proportion of
body weight that is water decreases with postna-
tal age (Fig. 9.2). In addition, the proportion of
total body water, which is extracellular, also
decreases, from 65% at 26  weeks gestation to
40% at term and 20% in childhood. This contrib-
utes to an expected weight loss in the first days of
life. A ‘one-size-fits-all’ prescription is inappro-
priate, and frequent monitoring of weight, urine
output and urea and electrolytes, with reassess-
ment of fluid prescription, should be mandatory
for surgical infants and children.
9.5.2.2 Energy Sources
Carbohydrates and fat provide the main energy
sources in the diet, and this is reflected by their
importance as a source of calories in PN.
Glucose is a main energy source for body cells
and is the primary energy substrate in PN
(Mesotten et  al. 2018). The amount of glucose
that can be infused safely depends on the clinical
A. Pierro and S. Eaton
water
intracellular
extracellular
fat
33 term 1 adult
month
condition and maturity of the infant or child. The
ability of neonates to metabolise glucose may be
impaired by prematurity and low birthweight.
Conversion of carbohydrate to fat (lipogenesis)
occurs when glucose intake exceeds metabolic
needs. The risks associated with this process are
twofold: accumulation of the newly synthesised
fat in the liver and aggravation of respiratory aci-
dosis resulting from increased CO2 production,
particularly in patients with compromised pul-
monary function. In addition, hyperglycaemia
can potentially result from excessive rates of glu-
cose administration.
Since the 1960s, safe commercial intravenous
fat emulsions have become widely used
(Lapillonne et al. 2018). These preparations have
a high caloric value (9 kcal/g of fat) and prevent
essential fatty acid deficiency and are isotonic,
allowing adequate calories to be given via a
peripheral vein. A number of studies in both
adults and infants have shown that the combined
infusion of glucose and lipids confers metabolic
advantages over glucose because it lowers the
metabolic rate and increases the efficiency of
energy utilisation.
In surgical infants receiving PN, there is a
negative linear relationship between glucose
intake and fat utilisation (oxidation and fat syn-
thesis). Net fat synthesis (and increased CO2 pro-
duction) from glucose occurs when the glucose
intake is greater than 18 g/kg/day (i.e. in excess
of energy expenditure) (Fig.  9.3) (Pierro et  al.
1993). The ability to oxidise fat is established
over a few days, so infusion rates are slowly
9 Nutrition
Fig. 9.3 Linear
relationship between
glucose intake and fat
oxidation
utilisation in surgical
infants on TPN
(r = −0.9; p < 0.0001).
Lipogenesis is
fat utlllzation
(g/kg/day)
significant when glucose
intake exceeds 18 g/kg/
day (Pierro et al. 1993)
synthesis
increased; fat tolerance should be monitored by
analysis of plasma triglyceride levels.
Historically, the first and most commonly
used fat emulsions for PN in paediatrics were
based on soybean oil, in which the lipid is present
as long-chain triglycerides (LCTs) (Raman et al.
2017). Medium-chain triglycerides (MCTs) can
increase net fat oxidation without increasing met-
abolic rate when used to partially replace LCTs.
MCT/LCT mixtures may improve essential fatty
acid status (by protecting essential fatty acids
from oxidation). In the last 20 years, several other
lipid emulsions have been introduced, based on
varying proportions of medium-chain triglycer-
ides, mono-unsaturated triglycerides (i.e. olive
oil), or omega-3-polyunsaturated triglycerides
(i.e. fish oil). These have been utilised in several
small-scale studies, both in premature infants and
surgical infants (Kotiya et al. 2016). The poten-
tial routine use of these novel lipids is compli-
cated by differences in licensing between
different geographical areas and by the maximal
rate of administration within the license. The
main aim of the use of novel lipids, however, is to
prevent or treat cholestasis; this will be consid-
ered below.
9.5.2.3 Amino Acids
In contrast to healthy adults who exist in a state
of neutral nitrogen balance, infants and children
need to be in a positive nitrogen balance in order
101
3
2
1
0
-1
-2
-3
10 14 18 22 26
glucose intake (g/kg/day)
to achieve satisfactory growth and development.
Infants are efficient at retaining nitrogen and can
retain up to 80% of the metabolizable protein
intake on both oral and intravenous diets. Protein
metabolism is dependent on both protein and
energy intake. The influence of dietary protein is
well established. An increased protein intake has
been shown to enhance protein synthesis, reduce
endogenous protein breakdown and thus enhance
net protein retention. The protein requirements of
newborn infants are between 2.5 and 3.0  g/kg/
day (Iacone et  al. 2020; van Goudoever et  al.
2018). The nitrogen source of PN is provided as
a mixture of amino acids. The commercially
available solutions contain the eight known
essential amino acids and histidine, which is
known to be essential in children. Complications,
such as azotaemia, hyperammonaemia and meta-
bolic acidosis, have been described in patients
receiving high levels of intravenous amino acids
but are rarely seen with amino acid intake of
2–3  g/kg/day. In patients with severe malnutri-
tion or with additional losses (i.e. jejunostomy,
ileostomy), protein requirements are higher.
Protein retention can be enhanced by giving car-
bohydrates or fat, which are said to be protein spar-
ing. Although some studies have suggested that the
protein-sparing effect of carbohydrates is greater
than that of fat, others have suggested that the pro-
tein-sparing effect of fat may be either equivalent
to, or greater than, that of carbohydrates. The addi-
102
tion of fat calories to the intravenous diet of surgi-
cal newborn infants reduces protein oxidation and
protein contribution to the energy expenditure and
increases protein retention.
The ideal quantitative composition of amino
acid solutions is still controversial, and as there is
significant metabolism and conversion between
enteral amino acids by the intestine and liver, par-
enteral amino acid requirements cannot be sim-
ply transposed from enteral requirements. Amino
acid solutions for infants and children are
designed to approximate plasma amino acid lev-
els, and the absolute parenteral requirement is
only known for a few amino acids. As well as the
amino acids usually considered essential for
adult humans, histidine is considered essential
for infants, and the following amino acids have
all been considered ‘conditionally essential’ for
neonates: arginine, cysteine, glutamine, taurine
and tyrosine. Of these, there has been much inter-
est in glutamine supplementation.
9.5.2.4 Vitamins and Trace Elements
Minerals, vitamins and trace elements are impor-
tant structurally, as cofactors, or as components
of enzymes, and provision of adequate supplies is
important for the growing neonate. Fe, Ca, P and
Mg should all be provided in adequate amounts
for growth and development, but conversely, they
can cause problems if provided in excess of needs
or if their metabolism is impaired (Mihatsch et al.
2018). In addition, the administration of adequate
amounts can be problematic because of a lack of
stability in solution or lack of compatibility with
other components. Consequently, iron is often
only supplemented in longer-term PN, whereas
calcium and phosphate supply depends on solu-
bility in PN mixtures (Domellof et  al. 2018).
Vitamins and trace elements are particularly
important in the maintenance of the body’s anti-
oxidant defences: vitamins C and E, selenium
(for glutathione peroxidase), copper, zinc and
manganese (all for superoxide dismutases), chro-
mium, iodine and molybdenum can all be added
to PN.  However, for many of these, the precise
requirements are not known. Although there is
A. Pierro and S. Eaton
evidence that selenium supplementation may be
beneficial, selenium status varies widely geo-
graphically, so global recommendations are dif-
ficult. It is suggested that if the duration of PN is
less than 4 weeks, of all the trace elements, only
zinc needs to be added. There is little specific evi-
dence for individual vitamin requirements, and
the current recommendations are to continue
with the available vitamin mixtures, which do not
appear to cause toxicity or deficiency in the
majority of children (Bronsky et al. 2018).
9.5.3 Complications of Parenteral
Nutrition (13)
9.5.3.1 Infectious Complications
In spite of significant improvement in the man-
agement of PN, including the introduction of
nutrition support teams, around 50% of surgical
infants on PN have at least one suspected episode
of sepsis, and around 30% of surgical neonates
have at least one positive blood culture. Careful
dressing and securement and dedicated teams for
central line insertion and maintenance, as well as
Chlorhexidine disinfectant, can reduce the risk of
catheter-born infections in children (Shi et  al.
2019; Ullman et  al. 2015; Mobley and Bizarro
2017). This may lead to impaired liver function,
critical illness and removal of central venous
catheters. It has always been assumed that the
central venous catheter is the major portal of
entry for micro-organisms causing septicaemia in
patients on parenteral nutrition. Infection with
enteric micro-organisms occurs significantly
later than presumed catheter-related infection,
supporting the hypothesis that a progressive
impairment of host defences and/or increased
intestinal permeability may allow translocation
of enteric organisms after an extended period of
PN. Important factors in reducing the incidence
of septic complications are placing intravenous
catheters under strict aseptic conditions, prepar-
ing the PN solutions in pharmacy in aseptic con-
ditions and using meticulous care when the
catheters are used.
9 Nutrition
9.5.3.2 Mechanical Complications
Mechanical complications related to the IV infu-
sion of nutrients are not uncommon. Histori-
cally, Wilmore and Dudrick started to use
central venous lines for long-term PN (Wilmore
and Dudrick 1969). Extravasation of PN solu-
tion is a common complication of peripheral
PN.  Unfortunately, even a low osmolarity solu-
tion is detrimental for peripheral veins leading to
inflammation and extravasation of the solution,
which can cause tissue necrosis and infection.
Extravasation injury is treated with occlusive
dressings or hyaluronidase irrigation, but there is
little evidence base for the best treatment in
­neonates. Intravenous lines may become clogged
from thrombus formation, calcium precipitates,
or lipid deposition (Noonan et al. 2018). There is
disagreement on the ideal position of central
venous lines (CVLs) for PN in infants. Some
authors advocate the atrium as the ideal position
because this would give less chance of catheter
dysfunction, whereas others believe that place-
ment in the superior vena cava would reduce the
risk of perforation. Current recommendations are
that the catheter tip should lie outside the atrium,
but because complications of either approach are
very rare (albeit potentially life-threatening),
there is a paucity of evidence from RCTs. Ultra-
sound controls are useful to detect catheter mis-
placement (Smit et al. 2020).
9.5.3.3 Hepatic Complications
The hepatobiliary complications related to PN
remain serious and often life-threatening. The
commonest hepatobiliary complication of PN in
neonates is cholestasis, sometimes termed
PN-associated cholestasis or PNAC; however,
recent evidence exists that it might be related to
intestinal failure including motility disorders and
is therefore termed as intestinal failure associated
cholestasis (IFALD). The clinical significance of
this cholestasis itself is unknown, but if untreated,
IFALD may occur, which can result in the need
for liver transplantation or can result in death.
Early IFALD (elevated liver enzymes) is thought
to be reversible, and established IFALD (total
bilirubin >50 μmol/L, with a conjugated fraction
of at least 50%) is potentially reversible if enteral
103
feeding is increased, PN is reduced and repeated
episodes of catheter-related sepsis are prevented.
Late IFALD is characterised by clinical signs of
end-stage liver disease and may require a liver
transplant. Patients with late IFALD frequently
also have problems with venous access due to
repeated episodes of thrombosis.
Various clinical factors are thought to contrib-
ute to the development of PN-related cholestasis.
These include prematurity, low birthweight,
duration of PN, immature entero-hepatic circula-
tion, intestinal microflora translocation to the
portal vein with recurrent septicaemia, failure to
implement enteral nutrition, short-bowel syn-
drome due to resection and intestinal motility
disorders, and number of laparotomies. In addi-
tion to the effects of extremely low birthweight
and length of time on PN, infants receiving PN
for either gastroschisis or jejunal atresia seem to
be at particular risk. PN-related cholestasis and
IFALD are diagnoses of exclusion without any
specific marker yet available. Therefore, infants
with cholestasis, who are receiving or have
received PN, must have an appropriate diagnostic
evaluation to exclude other causes of cholestasis,
such as bacterial and viral infections, metabolic
diseases and congenital anomalies. The aetiology
of PN-related cholestasis remains unclear.
Possible causes include the toxicity of compo-
nents of PN, lack of enteral feeding, continuous
non-pulsatile delivery of nutrients and host fac-
tors, infection and sepsis. In particular, the lipid
component of PN has been particularly impli-
cated, and many units now use alternative lipid
management strategies (see below). In addition
to lipid management, careful management of
these patients under a multi-disciplinary team is
beneficial so that centralisation of patients to
where appropriate specialist care is available
should be considered. Bowel lengthening proce-
dures, such as the STEP procedure or longitudi-
nal intestinal lengthening, may help with the
transition to enteral nutrition, whereas in those
with advanced liver disease and no prospect of
enteral autonomy, transplantation may be consid-
ered. PNAC and IFALD have been linked to the
use of soybean oil in the PN lipid emulsion,
although other factors are thought to be involved
104
(Madnawat et al. 2020). Soybean oil mostly con-
tains Ω-6 fatty acids, which are thought to be pro-­
inflammatory compared with Ω-3 fatty acids. In
addition, the amount of phytosterols delivered in
soybean-based lipid emulsions is relatively high,
and phytosterol accumulation has been postu-
lated to be involved in the pathogenesis of PNAC/
IFALD.  Hence, in recent years, there has been
great interest in adopting hepato-protective lipid
management strategies in PN of surgical infants
and children to either prevent or reverse cholesta-
sis. Several different approaches have been
adopted:
(i) Decreasing the amount of lipid adminis-
tered. This can also be achieved by limiting
the time on lipid by lipid-free hours or days.
This could potentially result in poor growth
due to inadequate calories.
(ii) Use of Omegaven®, a lipid emulsion of
10% fish oil. Fish oil is high in Ω-3 fatty
acids and low in phytosterols, so it has been
suggested to reverse cholestasis in surgical
infants on long term (Thavamani et al. 2019;
Fell et al. 2019). The use of Omegaven® as
the only lipid source could, however, poten-
tially result in essential fatty acid deficiency,
due to lack of Ω-6 fatty acids, and could also
result in poor growth, as the dose of
Omegaven® is limited to 1.0 g/kg/d (com-
pared with 3 g/kg/d for soybean-based lipid
emulsions).
(iii) Use of lipid emulsions containing a mixture
of long- (LCTs) and medium-chain (MCTs)
triglycerides. This has the advantage of
increasing fat utilisation, the ability to use at
up to 3 g/kg/d, and decreasing the phytoster-
ols and Ω-6 fatty acids administered whilst
ensuring adequate delivery of essential fatty
acids.
(iv) Use of mixed lipid emulsions, such as
SMOF®, which is a mixture of soybean,
MCT, olive and fish triglycerides. This has
the advantages of lowered amounts of phy-
tosterols and Ω-6 fatty acids, increased fat
utilisation, delivery of Ω-3 fatty acids and
the ability to use at up to 3 g/kg/d.
A. Pierro and S. Eaton
Despite these different approaches, the evi-
dence base supporting any lipid management
strategy is lacking due to the paucity of RCTs in
this area. It is difficult to design and implement
good quality RCTs, as units already use a variety
of alternate lipid management strategies. It is also
questionable whether maintaining an infant on a
high dose of soybean-based lipid emulsion after
the onset of PNAC/IFALD is ethical so that it is
difficult to decide on an appropriate comparison
group. Current paediatric guidelines for PN
(Lapillonne et al. 2018) recommend that soybean
oil-based lipid emulsions are not used as the sole
source of lipids for more than a few days and that
lipid management strategies, such as those above,
are instigated early before IFALD becomes
established.
9.6 Enteral Nutrition
The energy requirement of children fed enterally
is greater than the intravenous requirement
because of the energetic cost of absorption from
the gastrointestinal tract and energy lost in the
stools. Enteral feedings should be used in prefer-
ence to PN wherever possible. The transition
from parenteral to enteral feeding should be as
short as possible. Neonates undergoing abdomi-
nal surgery for congenital or acquired intestinal
dysfunction often require a period of PN; how-
ever, the authors recommend introducing enteral
feeding as soon as the gastric aspirate is less than
approximately 24 ml/day, even if the aspirate is
still bilious. There is evidence that during PN of
surgical infants, the introduction of minimal
enteral feeding preserves immune function.
Strictly speaking, the term ‘total parenteral nutri-
tion’ should be reserved for the period when
patients receive all nutrition parenterally.
Alternative feeding routes where children are
unable to feed orally include naso-gastric or oro-­
gastric tubes, naso-jejunal tubes, gastrostomy
tubes or jejunostomy tubes. Gastric feeding is
generally preferable to intestinal feeding because
it allows for a more natural digestive process, i.e.
allows the action of the salivary and gastric
9 Nutrition
enzymes and the antibacterial action of stomach
acid. In addition, gastric feeding is associated
with a larger osmotic and volume tolerance and a
lower frequency of diarrhoea and dumping syn-
drome. Thus, transpyloric feeds are usually
restricted to infants who (i) are unable to tolerate
naso- or oro-gastric feeds; (ii) are at increased
risk of aspiration; and (iii) have anatomical con-
traindications to gastric feeds, such as microgas-
tria. Neonates are obligatory nose breathers, and
therefore oro-gastric feeding may be preferable
over naso-gastric feeding in preterm infants to
avoid upper airway obstruction. However, naso-­
gastric tubes are easier to secure and may involve
a lower risk of displacement. In infants and
children requiring gastric tube feeding for
­ These infants may have had their small bowel
extended periods (e.g. more than 6–8 weeks), it is
advisable to insert a gastrostomy to decrease the
negative oral stimulation of repeated insertion of
nasal or oral tubes. The tube can be inserted using
an open, endoscopic or laparoscopic approach. In
children with significant gastro-oesophageal
reflux, fundoplication with gastrostomy tube or
enterostomy tube placement is indicated. In pre-
term infants with gastro-oesophageal reflux,
enteral feeding can be established via a naso-­
jejunal tube inserted under fluoroscopy. Naso-­
jejunal feeding usually minimises the episodes of
gastro-oesophageal reflux and their conse-
quences. However, it is common for these tubes
to dislocate back in the stomach. Regular analy-
sis of the pH in the aspirate is essential to monitor
the correct position of the tube. Feeding jejunos-
tomy tubes can be inserted through existing gas-
trostomy or directly into the jejunum via
laparotomy or laparoscopy.
9.6.1 Selection of Enteral Feeds
Breast milk is the ideal feed for infants because it
has specific anti-infectious activities and aids
gastrointestinal maturation and neurological
development (Lyons et  al. 2020). When breast
milk is not available, chemically defined formu-
lae can be used, which are designed either for
term infants or specifically for preterm infants. If
malabsorption is present and persists, an appro-
105
priate specific formula should be introduced. A
soy-based disaccharide-free feed is used when
there is disaccharide intolerance resulting in
loose stools containing disaccharides. For fat
malabsorption, a formula containing MCT should
be used. An elemental (free amino acids) or semi-­
elemental (protein hydrolysate containing di- and
tri-peptides) formula may be indicated when
there is severe malabsorption due to short bowel
syndrome or severe mucosal damage as in the
case of NEC. Semi-elemental preparations have
the advantage of a lower osmolality, are well
absorbed and have a more palatable taste. Infants
recovering from NEC pose a particular problem,
as malabsorption may be severe and prolonged.
resected, in addition to which the remaining
bowel may not have healed completely by the
time feeds are begun. Feeding may provoke a
relapse of the NEC, and feeding should therefore
be introduced cautiously. However, there is no
strong evidence regarding the time to re-­introduce
enteral feeds in infants who have had NEC. For
persistent severe malabsorption, a modular diet
may be necessary. Glucose, amino acid and MCT
preparations are provided separately, beginning
with the amino acid solution and adding the glu-
cose and then the fats, as tolerated. Minerals,
trace elements and vitamins are also added. These
solutions have a high osmolality and, if given too
quickly, may precipitate dumping syndrome,
with diarrhoea, abdominal cramps and hypogly-
caemia. It is important, therefore, to start with a
dilute solution and slowly increase the concentra-
tion and volume of each component. This may
take several weeks, and infants will need PN sup-
port during this period.
9.6.2 Administration of Enteral
Feeds
Enteral feeds can be administered as boluses,
continuous feeds or a combination of the two.
Bolus feeds are more physiological and are
known to stimulate intestinal motility, enterohe-
patic circulation of bile acids and gallbladder
contraction; continuous enteral feeding leads to
106 A. Pierro and S. Eaton

an enlarged, non-contractile gallbladder in lies, exclusive enteral feeding is commonly pre-

infants. Contraction is observed immediately cluded for some time after surgery due to large
after resuming bolus enteral feeds and gallblad- gastric aspirate and intestinal dysmotility.
der volume returns to baseline after four days. Therefore, appropriate calorie intake is established
Therefore, the mode of feeding has important initially by TPN. Supplementary enteral feeding is
bearings on the motility of the extrahepatic bili- introduced when intestinal motility and absorption
ary tree. Bolus feeds mimic or supplement meals improves. The percentage of calories given enter-
and are easier to administer than continuous ally is gradually increased at the expense of intra-
feeds since a feeding pump is not required. Bolus venous calorie intake. This transition time from
feeds are usually given over 15–20 minutes and total PN to total enteral feeding can be quite long.
usually every 3 hours; term infants can tolerate a The presence of significant gastric aspirate often
period of 4 hours without feeds before hypogly- induces clinicians and surgeons not to use the gut
caemia occurs. In preterm neonates or in neo- for nutrition. However, minimal enteral feeding
nates soon after surgery, 2-hourly feeds are can be implemented early in these patients even if
occasionally given. Where bolus feeds are not its nutritional value is questionable. Minimal
tolerated, e.g. in the presence of gastro-­ enteral feeding may be all that is required to
oesophageal reflux, continuous feeds should be enhance some immunological function.
administered via an infusion pump over 24 hours.
This modality of feeding is used in infants with
gastro-oesophageal reflux, delayed gastric emp- 9.7 Conclusions
tying or intestinal malabsorption. Infants with
jejunal tubes should receive continuous feeds and Despite high survival rates from surgical condi-
not bolus feeds as the stomach is no longer pro- tions, not enough attention is paid to nutrition in
viding a reservoir. the prevention of long-term adverse conse-
quences. This is particularly important in infants
where growth is in a critical period. Advances in
9.6.3 Complications of Enteral Tube artificial nutrition support, both parenteral and
Feeding enteral, have improved these outcomes.
Significant progress has been achieved by the
Enteral tube feeding is associated with fewer careful management of central venous catheters
complications than parenteral feeding. The com- and sophisticated PN solution compositions.
plications can be mechanical including tube Nevertheless, infectious complications, either
blockage, tube displacement or migration and through central venous catheters or by intestinal
intestinal perforation. Although infection is less bacterial translocation and chronic liver failure
of a risk than with PN, the risk of infected enteral after long-term PN, are still the challenges in the
feeds should not be ignored. Other complications management of neonates and infants with chronic
involve the gastrointestinal tract. These include intestinal diseases.
gastro-oesophageal reflux with aspiration pneu-
monia, dumping syndrome and diarrhoea.
Jejunostomy tubes inserted at laparotomy can
also be associated with intestinal obstruction. The References
use of hyperosmolar feeds has been associated
Bishay M, Lakshminarayanan B, Arnaud A, Garriboli M,
with the development of NEC, dehydration and, Cross KM, Curry JI et al (2013) The role of parenteral
rarely, intestinal obstruction due to milk curds. nutrition following surgery for duodenal atresia or ste-
In surgical infants, enteral feeding often nosis. Pediatr Surg Int 29(2):191–195
Bronsky J, Campoy C, Braegger C, nutrition EEECwgopp
results in vomiting, interruption of feeding,
(2018) ESPGHAN/ESPEN/ESPR/CSPEN guidelines
inadequate calorie intake and, rarely, NEC.  In on pediatric parenteral nutrition: vitamins. Clin Nutr
infants with congenital gastrointestinal anoma- 37(6 Pt B):2366–2378
9 Nutrition
Denne SC, Poindexter BB, Leitch CA, Ernst JA, Lemons
PK, Lemons JA (2006) Nutrition and metabolism in the
high-risk neonate. In: Martin RJ, Fanarof AA, Walsh
MC (eds) Fanaroff and Martin’s neonatal-­ perinatal
medicine, 8th edn. Mosby-Elsevier, Philadeplhia, PA,
pp 661–693
Domellof M, Szitanyi P, Simchowitz V, Franz A, Mimouni
F, nutrition EEECwgopp (2018) ESPGHAN/ESPEN/
ESPR/CSPEN guidelines on pediatric parenteral
nutrition: Iron and trace minerals. Clin Nutr 37(6 Pt
B):2354–2359
Dudrick SJ, Wilmore DW (1968) Long-term parenteral
feeding. Hosp Pract 3:65–78
Dudrick SJ, Wilmore DW, Vars HM (1967) Long-term
total parenteral nutrition with growth in puppies and
positive nitrogen balance in patients. Surg Forum
18:356–357
Fell GL, Cho BS, Dao DT et al (2019) Fish oil protects
the liver from parenteral nutrition-induced injury via
GPR120-mediated PPARy signaling. Prostaglandins
Leokot Essent Fatty Acids 143:8–14
Hall NJ, Drewett M, Wheeler RA, Griffiths DM,
Kitteringham LJ, Burge DM (2011) Trans-anastomotic
tubes reduce the need for central venous access and
parenteral nutrition in infants with congenital duode-
nal obstruction. Pediatr Surg Int 27(8):851–855
Hartmann G (1985) History of parenteral nutrition. Bibl
Nutr Dieta 35:1–8
Hill AG, Hill GL (1998) Metabolic response to severe
injury. Br J Surg 85:884–890
Iacone R, Scanzano C, Santarpia L et  al (2020)
Macronutrients in parenteral nutrition: amino acids.
Nutrients 12:772. https://doi.org/10.3390/nu12030772
Jochum F, Moltu SJ, Senterre T, Nomayo A, Goulet O,
Iacobelli S et  al (2018) ESPGHAN/ESPEN/ESPR/
CSPEN guidelines on pediatric parenteral nutri-
tion: fluid and electrolytes. Clin Nutr 37(6 Pt B):
2344–2353
Koletzko B, Goulet O, Hunt J, Krohn K, Shamir R (2005)
Guidelines on Paediatric Parenteral Nutrition of the
European Society of Paediatric Gastroenterology,
Hepatology and Nutrition (ESPGHAN) and the
European Society for Clinical Nutrition and
Metabolism (ESPEN), Supported by the European
Society of Paediatric Research (ESPR). J Pediatr
Gastroenterol Nutr 41(Suppl 2):S1–S87
Kotiya P, Zhao X, Cheng P et al (2016) Fish oil- andsoy
oil-based lipid emulsions in neonatal parenteral nuti-
tion: a systematic review and meta-analysis. Eur J Clin
Nutr 70:1106–1115
Lapillonne A, Fidler Mis N, Goulet O, van den Akker
CHP, Wu J, Koletzko B et  al (2018) ESPGHAN/
ESPEN/ESPR/CSPEN guidelines on pediatric
parenteral nutrition: lipids. Clin Nutr 37(6 Pt B):
2324–2336
107
Lyons K, Ryan CA, Dempsey EM et  al (2020) Breats
milk, a source of beneficial microbes and associated
benefits for infant health. Nutrients 12:1039. https://
doi.org/10.3390/nu12041039
Madnawat H, Welu AL, Gilbert EJ et al (2020) Mechanism
of parenteral nutrition-associated liver and gut injury.
Nutr Clin Pract 35:63–71
Mesotten D, Joosten K, van Kempen A, Verbruggen S,
nutrition EEECwgopp (2018) ESPGHAN/ESPEN/
ESPR/CSPEN guidelines on pediatric parenteral nutri-
tion: carbohydrates. Clin Nutr 37(6 Pt B):2337–2343
Mihatsch W, Fewtrell M, Goulet O, Molgaard C, Picaud
JC, Senterre T et al (2018) ESPGHAN/ESPEN/ESPR/
CSPEN guidelines on pediatric parenteral nutrition:
calcium, phosphorus and magnesium. Clin Nutr 37(6
Pt B):2360–2365
Mobley RE, Bizarro MJ (2017) Central line-associated
bloodstream infections in the NICU: successes and
controversies in the quest for zero. Semin Perinatol
41:166–174
Morgan A, Filler RM, Moore FD (1970) Surgical nutri-
tion. Med Clin North Am 54:1367–1381
Noonan PJ, Hanson SJ, Simpson PM et  al (2018)
Comparison of complication rates of central venous
catheters versus peripherally inserted central venous
catheters in pediatric patients. Pediatr Crit Care Med
19:1097–1105
Pierro A, Jones MO, Hammond P, Nunn A, Lloyd DA
(1993) Utilisation of intravenous fat in the surgical
newborn infant. Proc Nutr Soc 52:237A
Raman M, Almutairdi A, Mulesa L et al (2017) Parenteral
nutrition and lipids. Nutrients 9:388. https://doi.
org/10.3390/nu9040388
Shi Y, Yang N, Li Z et al (2019) Chlorhexidine disinfec-
tant can reduce the risk of central venous catheter
infection compaed with providone: a meta-analysis.
Am J Infect Control 47:1255–1262
Smit JM, Haaksma ME, Lim EHT et al (2020) Ultrasound
to detect central venous catheter placement associ-
ated complications: a multicenter diagnostic accuracy
study. Anesthesiology 132:781–794
Thavamani A, Mhanna MJ, Groh-Wargo S et  al (2019)
Enteral fish oil supplementation in the resolution of
parenteral nutrition associated cholestasis. J Neonatol
Perinatal Med 12:13–20
Ullman A, Cooke ML, Mitchell M et al (2015) Dressing
and securement devices for central venous catheters
(CVC). Cochrane Database Syst Rev 9:CD010367
van Goudoever JB, Carnielli V, Darmaun D, Sainz de
Pipaon M, nutrition EEECwgopp (2018) ESPGHAN/
ESPEN/ESPR/CSPEN guidelines on pediatric par-
enteral nutrition: amino acids. Clin Nutr 37(6 Pt
B):2315–2323
Wilmore DW, Dudrick SJ (1969) Safe long-term venous
catheterisation. Arch Surg 98:256–258
 Access for Enteral Nutrition
Julia Brendel and Michael W. L. Gauderer
10.1 Introduction
Enteral access is crucial for the management of
various surgical and nonsurgical pediatric condi-
tions, either for feeding or decompressing, or a
combination of both. Enteral nutrition support
can be used for short-, intermediate-, or long-­
term feeding. Direct enteral nutrition prevents or
treats a reduced nutrient state, which might other-
wise lead to inadequate development, growth
retardation, tissue breakdown, poor wound heal-
ing, or compromised immune function (Lord
2018). It has been shown that psychological well-­
being is linked to dehydration and malnutrition
(Abdelhadi et  al. 2016; Krasaelap et  al. 2020;
Lord 2018; Vermilyea and Goh 2016).
The advantages of enteral feeding over paren-
teral nutrition are well known, particularly the
preservation of gastrointestinal physiology and
the avoidance of intravenous access complica-
tions (Vermilyea and Goh 2016; Pang et  al.
2017).
However, abdominal pain, decreased appetite,
vomiting or nausea, and early satiety or postpran-
J. Brendel (*)
Department of Pediatric Surgery, Hannover Medical
School, Hannover, Germany
e-mail: brendel.julia@mh-hannover.de
M. W. L. Gauderer
Surgery and Pediatrics, University of South Carolina,
Greenville, SC, USA
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_10
10
dial fullness often limit caloric intake (Krasaelap
et al. 2020).
Recently, the American Society for Parenteral
and Enteral Nutrition set criteria for pediatric
patients who meet the criteria for enteral nutri-
tion as follows (Boullata et al. 2017):
–– Children with insufficient oral intake, particu-
larly children older than 1 year who are unable
to meet ≥60–80% of individual requirements
for ≥5 days and children younger than 1 year
who are unable to meet ≥60–80% of individ-
ual requirements for ≥3 days.
–– Children who meet the criteria for failure to
thrive, wasting, and stunting.
Previously, enterostomies were mainly per-
formed in patients with major congenital anoma-
lies of the abdominal wall and gastrointestinal
tract, but nowadays, indications are more likely
to be for nonsurgical conditions. Typically, these
are children with swallowing difficulties due to
central nervous lesions, patients requiring feed-
ing supplementation following chemotherapy or
trauma or chronic malnutrition secondary to
anorexia, chronic liver disease, cystic fibrosis,
extensive burns, and other conditions (Abdelhadi
et al. 2016; Boullata et al. 2017).
The choice of enteral access procedure
depends on the specific indication, the experience
of the medical team, and the available resources.
Furthermore, the patients’ specific needs, habi-
109
110
tus, comorbidities, and the caretakers’ experience
should be considered. Regular early and long-­
term follow-up appointments are important to
assure optimal functionality and minimal mor-
bidity (Zimmer and Gauderer 2020).
10.2 Naso-, Orogastric or Naso-,
Oroenteric Access
10.2.1 Indication
Naso- or orogastric tubes and naso- or oroen-
teric tubes are implemented for short-term and
intermediary use, lasting for a couple of weeks
to a few months (Abdelhadi et  al. 2016;
Ricciuto et al. 2015). However, in special situ-
ations, the long-term use of nasogastric access
in the outpatient setting is appropriate if the
pediatric patient or the caregivers are able to
self-place a nasogastric tube at home (Boullata
et al. 2017).
Besides nutritional support, these devices may
be used for the administration of medication or
fluids (Fig. 10.1) (Abdelhadi et al. 2016; Irving
et al. 2014). If properly placed, secured, and reg-
Fig. 10.1 Premature
infant with continuous
nasoenteric pump
feeding
J. Brendel and M. W. L. Gauderer
ularly flushed, naso- or orogastric tubes generally
decompress more effectively than gastrostomy
tubes (Vermilyea and Goh 2016).
In general, nasoenteric tubes should be pre-
ferred over oroenteric tubes due to patient com-
fort. However, in certain instances, such as
premature infants or neonates, orogastric tubes
are often used since infants are predominantly
obligate nasal breathers (Krasaelap et  al. 2020;
Gauderer et al. 2009).
Nasally placed tubes should be avoided in
patients with recent nasal surgery, nasal fractures,
or severe trauma to the midface (Lord 2018).
Patients with severe coagulation abnormalities,
esophageal varices or stricture, or alkaline inges-
tion are also considered high risk (Lord 2018;
Boullata et al. 2017).
The placement of a tube into the stomach
should be the primary option unless a small-­
bowel tube position is needed due to anatomical
abnormalities, high aspiration risk, to provide
nutrients beyond an obstruction, a bowel leak, or
a fistula (Lord 2018). Nasojejunal feeding tubes
should be placed past the ligament of Treitz to
minimize the amount of nutrition entering the
stomach (Simons and Bulat 2021).
10  Access for Enteral Nutrition
Dual-lumen tubes can provide gastric decom-
pression, along with small-bowel tube feeding
simultaneously (Lord 2018).
10.2.2 Insertion and Verification
of Placement
Naso- or orogastric tubes and naso- or oroenteric
tubes are available in various sizes, materials, and
with or without guidewires (Figs.  10.2, 10.3).
Size and material should be chosen carefully
according to the patient’s age, size, and the dura-
tion of placement.
The tubes are inserted through the nares or the
mouth, and the distal tips are positioned in the
stomach or small intestine. This is usually per-
formed bedside, either blindly or with the aid of a
manufactured device to enhance patient safety
Fig. 10.2  Different sizes (Charrière 5 and 8) and material
of naso/orogastric tubes
111
Fig. 10.3  Guidewired naso/orogastric tube
(Lord 2018; Gimenes et  al. 2020). If inserted
blindly, medical practitioners commonly follow
the nose-ear-xiphoid method to predict the length
of tubing required to reach the stomach. The
measurement starts from the tip of the nose to an
earlobe to the bottom of the xiphoid process
(Lord 2018).
Once the tube is placed, it should be secured
immediately with plaster stripes. It is also recom-
mended to mark and document the inserted
length for future references and to avoid unrecog-
nized dislocation.
To verify the proper tube position, simple air
insufflation and auscultation is a common method
but is no longer recommended as the sole method
due to the possibility of pitfalls (Vermilyea and
Goh 2016).
Although a plain abdominal X-ray is quick
and easy, there is an understandable hesitation in
its use due to radiation exposure (Vermilyea and
Goh 2016; Abdelhadi et  al. 2016; Irving et  al.
2014). Ultrasound verification is technically lim-
ited because intragastric contents, especially gas,
make interpretation difficult (Abdelhadi et  al.
2016; Adams et al. 2014; Tsujimoto et al. 2017).
Other accepted methods are visual observation of
gastric aspirate or gastric pH testing (Irving et al.
2014). The use of electromagnetic devices and
capnography is still controversial (Abdelhadi
et al. 2016; Gilbert and Burns 2012; Powers et al.
2011; Bear et al. 2016).
112
10.2.3 Complications
10.2.3.1 Misplacement
and Displacement
Misplacement is the placement of the tube tip in
an unintended anatomical position, and displace-
ment is the migration to an unintended anatomic
position for the proper position of the device
(Boullata et al. 2017).
An early complication of tube introduction is
accidental catheter placement into the trachea or
the bronchial tree, leading to pulmonary injury,
pneumonia, or even death (Lord 2018; Krasaelap
et  al. 2020). Perforation of the esophagus or
intracranial placement has also been reported
(Lord 2018; Isik et  al. 2014; Rahimi-Movaghar
et  al. 2005). Nasojejunal tubes are prone to
migrating into the stomach, especially after vom-
iting or endoscopic manipulation (Simons and
Bulat 2021).
The tube should be withdrawn immediately if
there are any signs of respiratory distress, such as
dyspnea, choking, or coughing (Lord 2018). In
alert patients, sips of water while the tube is
inserted might help guide it appropriately through
the esophagus (Lord 2018).
10.2.3.2 Long-Term Complications
Long-term complications include damage to the
nose or lip, nasooropharyngeal irritation and
infection, esophageal and gastric mucosal ero-
sion, and gastroesophageal reflux and aspiration
(Abdelhadi et al. 2016; Boullata et al. 2017; Lord
2018).
A recent multicenter study reported inadver-
tent feeding tube removal as the most frequent
mechanical device-related complication, which
was observed in up to 70% of patients (Gimenes
et al. 2020).
10.2.3.3 Tube Plugging
Obstruction is a common tube complication,
varying in the literature from 12.5 to 45.0%
(Gimenes et al. 2020).
Nasoenteric feeding tubes are more likely to
clog than tubes placed in the stomach due to their
longer lengths and narrow diameters (Lord 2018).
J. Brendel and M. W. L. Gauderer
Plunger movements back and forth with a
water-filled syringe usually helps to loosen up a
clog and encourage clog withdrawal. Sugared
sodas, pancreatic enzymes activated with baking
soda, cranberry juice, or meat tenderizer have
also been proven useful for declogging (Lord
2018).
10.3 Gastrostomy
This section gives only a short overview.
Gastrostomies are discussed extensively in
Chap. 61.
Gastrostomy is the preferred access for long-­
term enteral feeding as it is physiologic, is usu-
ally well-tolerated, and works for continuous
and bolus feedings as well as gastric venting and
drug application. There is no interference
between oral feedings and physiological activi-
ties. Nevertheless, gastrostomies have a long list
of potential early and late complications (Adams
et  al. 2014; Baker et  al. 2015; Landisch et  al.
2016).
10.3.1 Indications
Contemporary “nonsurgical” pathologies are the
most common indication for placement of a gas-
trostomy. It supports patients with failure to
thrive, malignancies, and trauma and/or inability
to swallow, as well as those needing feeding sup-
plementation, nonpalatable medications, and
chronic malabsorption syndromes (Fig.  10.4)
(Zimmer and Gauderer 2020). It is important to
mention that gastrostomy and gastrojejunostomy
procedures in neurologically impaired and chron-
ically ventilator-dependent children have a high
risk of postoperative complications as well as
morbidity and mortality (Liu et al. 2013; Chatwin
et al. 2013).
Regarding “surgical pathologies,” gastrosto-
mies are indicated for esophageal abnormalities,
such as esophageal stricture, after alkaline inges-
tion and esophageal atresia without fistula. They
may also be employed for gastric decompression
10  Access for Enteral Nutrition
Fig. 10.4  A 5-year-old with cystinosis demonstrating
feeding on a teaching doll
after repair of duodenal obstruction and major
wall defects with associated atresia (Gauderer
and Zimmer 2017).
In any condition with prolonged ileus or par-
tial luminal occlusion (e.g., complicated meco-
nium ileus, small bowel Hirschsprung’s disease)
or when a complex feeding regimen is likely
(e.g., those with intestinal lymphangiectasia), a
gastrostomy can facilitate management (Gauderer
and Zimmer 2017).
10.3.2 Techniques
A wide variety of gastrostomy techniques are
available. There are three basic types (Gauderer
and Stellato 1986; Gauderer 2006; Gauderer and
Zimmer 2017):
A. Formation of a serosa-lined channel from the
anterior gastric wall to the skin surface around
a catheter. The catheter is placed in the stom-
113
ach and exits either parallel or vertically to
the gastric serosa.
B. Formation of a tube or conduit from a full-­
thickness gastric wall flap to the skin
surface.
C. Percutaneous techniques, in which the intro-
duced catheter holds the gastric and abdomi-
nal wall in apposition, with or without the aid
of special fasteners.
With certain modifications, each of these
methods can be performed employing minimally
invasive approaches.
The open “Stamm” technique is the most
widely applied gastrostomy via laparotomy. It
can be used for children of all sizes, either as an
isolated intervention or along with other intra-­
abdominal procedures (Gauderer et  al. 2009).
The first gastrostomy without laparotomy was
the percutaneous endoscopic gastrostomy (PEG),
initially developed for high-risk pediatric patients
and eventually became known as the “pull” tech-
nique. (Gauderer et  al. 1980). The procedure is
employed in children of all ages, usually for the
purpose of long-term enteral feeding (Lalanne
et al. 2014; Srinivasan et al. 2009; Gauderer and
Zimmer 2017).
Various other methods of gastrostomy without
laparotomy have been introduced over the years
(Gauderer et  al. 2009). The percutaneous endo-
scopic “push” technique is performed with the
aid of needle-deployed gastric anchors or “T”
fasteners, followed by the Seldinger method of
guidewire introduction. Progressive tract dilata-
tions and insertion of a long tube or skin-level
gastrostomy device follow (Robertson et  al.
1996). A similar approach is used by interven-
tional radiologists and found to be suitable for
even very small stomachs (Cahill et  al. 2001;
Aziz et al. 2004).
Laparoscopic-aided techniques are essentially
expansions of the above methods and can also be
employed in infants (Rothenberg et al. 1999).
Several comparative studies have been pub-
lished regarding the outcome, efficiency, and
complications of different techniques, favoring
laparoscopic procedures (Liu et al. 2013; Baker
et  al. 2015; Merli et  al. 2016; Petrosyan et  al.
114
2016). However, Infants with an abnormal epi-
gastric anatomy, in whom the above-mentioned
techniques are difficult to perform, may benefit
from a hybrid procedure combining a mini-­
laparotomy and the PEG (Gauderer and Zimmer
2017).
10.3.3 Bolus and Continuous Feeds
Gastric feeds allow for either bolus or continuous
feeds. Bolus feedings are preferable as they mini-
mize disruption to lifestyle and the dependence
upon pump devices and allow the stomach to
completely empty and may therefore reduce bac-
terial colonization of the stomach (Krasaelap
et  al. 2020). In certain conditions and motility
disorders, such as gastroparesis, severe gastro-
esophageal reflux disease, and pediatric intestinal
pseudo-obstruction, continuous feeding is toler-
ated better (Krasaelap et al. 2020).
10.3.4 Complications
Although frequently considered a “simple” pro-
cedure, a gastrostomy has considerable potential
for early and late morbidity, particularly among
neonates (Gauderer and Zimmer 2017; Farrelly
and Stitelman 2016). Skin-level devices have
decreased the most common problems associated
with older, long tubes, independently from the
surgical technique (Gauderer and Stellato 1986;
Gauderer and Zimmer 2017).
Major bleeding is usually related to inade-
quate hemostasis at the time of catheter insertion
(Gauderer et al. 2009).
Separation of the stomach from the abdominal
wall occurs most frequently shortly after early
gastrostomy tube reinsertion before a firm adhe-
sion between gastric and abdominal walls has
occurred but can also occur at any time thereafter.
During the attempt to replace a dislodged cathe-
ter, the stomach is pushed away from the abdomi-
nal wall, leading to a partial or complete
separation of the stoma. If not recognized in time,
this can lead to severe peritonitis and death
J. Brendel and M. W. L. Gauderer
(Zimmer and Gauderer 2020; Gauderer and
Stellato 1986).
Infections may occur with any type of gastros-
tomy and are usually limited to the skin and sub-
cutaneous tissue. However, the whole abdominal
wall can also get infected (Gauderer and Stellato
1986; Goldin et al. 2016). Wound separation and
dehiscence are usually seen due to technical
problems after open procedures (Gauderer and
Stellato 1986; Gauderer et al. 2009).
The posterior gastric wall and other organs
can be damaged or perforated, not only during
the initial procedure, but also later during cath-
eter change (Gauderer and Stellato 1986).
Damage to the liver and spleen through the
improper use of retractors or other instruments
should meticulously be avoided. The distended
colon may be mistaken for the stomach
(Gauderer et al. 2009).
Gastro-colic fistulas can occur with any gas-
trostomy. This complication is more likely with
the percutaneous endoscopic techniques
(Gauderer and Stellato 1986; Gauderer and
Zimmer 2017).
Various problems can occur due to the mis-
handling of the gastrostomy or the gastrostomy
tube. Granulation tissue, skin irritation, and can-
didiasis are usually related to leakage and com-
pounded by occlusive dressings. Improper
feedings can cause metabolic abnormalities,
excessive reflux, and diarrhea. The tube catheter
becomes clogged if not regularly flushed after
each feeding or drug administration (Gauderer
and Zimmer 2017).
Accidental dislodgement of gastrostomy cath-
eters is quite common. The catheter must be
replaced before the tract closes, which can be in a
few hours unless it is already well matured and
epithelium-lined (Gauderer and Zimmer 2017).
Improper catheter reintroduction can damage
other organs, especially if long stylets or other
traumatic instruments are used (Gauderer and
Zimmer 2017).
Internal migration can occur with any gastric
tube but is particularly common with long,
balloon-­type catheters (Gauderer and Stellato
1986). Accidental inflation of a balloon-type
10  Access for Enteral Nutrition
catheter in the esophagus or small bowel might
lead to wall disruption (Gauderer and Zimmer
2017).
Over-zealous approximation of external
immobilizing devices (bumper) can lead to the
embedding of the inner cross-bar of the PEG
catheter, mushroom tip, or balloon in the gastric
and abdominal wall (external migration), result-
ing in the so-called “buried bumper syndrome”
(BBS) (Gauderer and Stellato 1986; Gauderer
et al. 2009; Abdelhadi et al. 2016).
10.4 Jejunal Access
This section gives only a short overview.
Jejunostomies are discussed extensively in
Chap. 61.
Jejunostomies are not used as frequently as
gastrostomies because they are less physiologic,
more difficult to place and maintain, and more
prone to complications (Abdelhadi et  al. 2016;
Gauderer et al. 2009).
Depending on their expected length of use,
jejunal access can be divided into short-term
(e.g., nasojejunal catheters), medium-term (e.g.,
trans-gastric jejunal tubes), and long-term access
(conventional jejunostomy) and, secondly, into
indirect and direct jejunal access, depending on
how the catheters are placed (Gauderer et  al.
2009).
10.4.1 Indications
Jejunostomies are implemented in children with
complex medical problems who are in need of
long-term enteral access and are unable to get a
gastrostomy. Postpyloric feedings are used to
overcome problems related to aspiration, gastric
emptying or paresis, microgastria, gastric outlet
obstruction, or gastroesophageal reflux disease
(Vermilyea and Goh 2016).
The usual purpose of jejunostomies is feeding,
whereas the administration of medication is a
less common use. An early “classic” combined
115
decompressing-feeding jejunostomy is the
Rehbein technique using a dual-purpose catheter
in infants with jejunal atresia. The larger lumen
provides decompression of the proximal dilated
bowel, while the thinner, trans-anastomotic tube
permits early feeding (Gauderer and Zimmer
2017).
10.4.2 Techniques and Feeding
Needle catheter jejunostomies are for short-term
use via direct jejunal access and can be employed
as an adjunct during other intraabdominal inter-
ventions (Gauderer et al. 2009).
A long-term approach is the direct catheter
placement into the jejunum with the formation of
a Witzel-type channel or the catheter placement
in a partially excluded loop (Roux-en-Y)
(Gauderer et al. 2009).
Postoperative care is similar to that after a
gastrostomy. However, feeding should be started
at a much slower rate to reduce the risk of ileus
and diarrhea and should be administered con-
tinuously by a pump (Zimmer and Gauderer
2020). Postpyloric bolus feedings bear the risk
of developing dumping syndrome (Krasaelap
et al. 2020).
10.4.3 Complications
Overall, jejunostomies have significantly more
complications than gastrostomies. Adhesive
bowel obstruction, dumping syndrome-­associated
diarrhea, intestinal volvulus around the stoma,
and internal hernia are specifically jejunostomy-­
related complications (Abdelhadi et  al. 2016;
Farrelly and Stitelman 2016; Gauderer et  al.
2009). Morbidity is related to tube dislodgement,
persistent reflux, and intestinal perforation
(Farrelly and Stitelman 2016; Campwala et  al.
2015; Demehri et  al. 2016). Gastrojejunal but-
tons that are too big in size may cause pyloric
obstruction in smaller children (Vermilyea and
Goh 2016).
116
10.5 Conclusion
Before choosing a specific enteral access proce-
dure, the medical team, parents, and caregivers
must carefully consider the advantages and dis-
advantages of each method, keeping the patient’s
comorbidities in mind as they mean major inter-
vention in the children’s life. Whenever possible,
every effort should be made to institute or resume
oral feedings.
References
Abdelhadi RA, Rahe K, Lyman B (2016) Pediatric enteral
access device management. Nutr Clin Pract 31(6):748–
761. https://doi.org/10.1177/0884533616670640
Adams SD, Baker D, Takhar A, Beattie RM, Stanton MP
(2014) Complication of percutaneous endoscopic
gastrostomy. Arch Dis Child 99(8):788. https://doi.
org/10.1136/archdischild-­2014-­306123
Aziz D, Chait P, Kreichman F, Langer JC (2004) Image-­
guided percutaneous gastrostomy in neonates with
esophageal atresia. J Pediatr Surg 39(11):1648–1650.
https://doi.org/10.1016/j.jpedsurg.2004.07.004
Baker L, Beres AL, Baird R (2015) A systematic review
and meta-analysis of gastrostomy insertion techniques
in children. J Pediatr Surg 50(5):718–725. https://doi.
org/10.1016/j.jpedsurg.2015.02.021
Bear DE, Champion A, Lei K, Smith J, Beale R,
Camporota L, Barrett NA (2016) Use of an electro-
magnetic device compared with chest X-ray to con-
firm nasogastric feeding tube position in critical care.
JPEN J Parenter Enteral Nutr 40(4):581–586. https://
doi.org/10.1177/0148607115575034
Boullata JI, Carrera AL, Harvey L, Escuro AA, Hudson
L, Mays A et  al (2017) ASPEN safe practices for
enteral nutrition therapy formula: see text. JPEN
J Parenter Enteral Nutr 41(1):15–103. https://doi.
org/10.1177/0148607116673053
Cahill AM, Kaye RD, Fitz CR, Towbin RB (2001) “Push-­
pull” gastrostomy: a new technique for percutaneous
gastrostomy tube insertion in the neonate and young
infant. Pediatr Radiol 31(8):550–554. https://doi.
org/10.1007/s002470100472
Campwala I, Perrone E, Yanni G, Shah M, Gollin G
(2015) Complications of gastrojejunal feeding tubes
in children. J Surg Res 199(1):67–71. https://doi.
org/10.1016/j.jss.2015.06.058
Chatwin M, Bush A, Macrae DJ, Clarke SA, Simonds AK
(2013) Risk management protocol for gastrostomy
and jejunostomy insertion in ventilator dependent
infants. Neuromuscul Disord 23(4):289–297. https://
doi.org/10.1016/j.nmd.2013.01.006
Demehri FR, Simha S, Herrman E, Jarboe MD, Geiger
JD, Teitelbaum DH, Gadepalli SK (2016) Analysis
J. Brendel and M. W. L. Gauderer
of risk factors contributing to morbidity from gas-
trojejunostomy feeding tubes in children. J Pediatr
Surg 51(6):1005–1009. https://doi.org/10.1016/j.
jpedsurg.2016.02.072
Farrelly JS, Stitelman DH (2016) Complications in
pediatric enteral and vascular access. Semin Pediatr
Surg 25(6):371–379. https://doi.org/10.1053/j.
sempedsurg.2016.10.006
Gauderer MWL (2006) Gastrostomy. In: Spitz L, Coran
AG (eds) Operative pediatric surgery, 6th edn. Hodder
Arnold, London, pp 330–355
Gauderer MWL, Zimmer J (2017) Acces for enteral nutri-
tion. In: Puri P (ed) Newborn surgery, 4th edn. CRC
Press, Boca Raton
Gauderer MWL, Stellato TA (1986) Gastrostomies:
evolution, techniques, indications, and complica-
tions. Curr Probl Surg 23(9):661–719. https://doi.
org/10.1016/0011-­3840(86)90020-­1
Gauderer MWL, Ponsky JL, Izant RJ (1980) Gastrostomy
without laparotomy: a percutaneous endoscopic
technique. J Pediatr Surg 15(6):872–875. https://doi.
org/10.1016/S0022-­3468(80)80296-­X
Gauderer MWL (2009) Gastrostomy and jejunostomy.
In: Puri P, Höllwarth ME (eds) Pediatric surgery:
diagnosis and management, 1st edn. Springer, Berlin,
London, pp 369–381
Gilbert RT, Burns SM (2012) Increasing the safety of
blind gastric tube placement in pediatric patients: the
design and testing of a procedure using a carbon diox-
ide detection device. J Pediatr Nurs 27(5):528–532.
https://doi.org/10.1016/j.pedn.2011.08.004
Gimenes FRE, Baracioli FFLR, Medeiros AP, Prado PR,
Koepp J, Pereira MCA et al (2020) Factors associated
with mechanical device-related complications in tube
fed patients: a multicenter prospective cohort study.
PLoS One 15(11):e0241849. https://doi.org/10.1371/
journal.pone.0241849
Goldin AB, Heiss KF, Hall M, Rothstein DH, Minneci PC,
Blakely ML et al (2016) Emergency department visits
and readmissions among children after gastrostomy
tube placement. J Pediatr 174:139–145.e2. https://doi.
org/10.1016/j.jpeds.2016.03.032
Irving SY, Lyman B, Northington L, Bartlett JA, Kemper
C (2014) Nasogastric tube placement and verifica-
tion in children: review of the current literature. Crit
Care Nurse 34(3):67–78. https://doi.org/10.4037/
ccn2014606
Isik A, Firat D, Peker K, Sayar I, Idiz O, Soytürk M
(2014) A case report of esophageal perforation:
complication of nasogastric tube placement. Am
J Case Rep 15:168–171. https://doi.org/10.12659/
AJCR.890260
Krasaelap A, Kovacic K, Goday PS (2020) Nutrition
management in pediatric gastrointestinal motility
disorders. Nutr Clin Pract 35(2):265–272. https://doi.
org/10.1002/ncp.10319
Lalanne A, Gottrand F, Salleron J, Puybasset-Jonquez
AL, Guimber D, Turck D, Michaud L (2014)
Long-term outcome of children receiving percu-
taneous endoscopic gastrostomy feeding. J Pediatr
10  Access for Enteral Nutrition
Gastroenterol Nutr 59(2):172–176. https://doi.org/
10.1097/MPG.0000000000000393
Landisch RM, Colwell RC, Densmore JC (2016)
Infant gastrostomy outcomes: the cost of complica-
tions. J Pediatr Surg 51(12):1976–1982. https://doi.
org/10.1016/j.jpedsurg.2016.09.025
Liu R, Jiwane A, Varjavandi A, Kennedy A, Henry G,
Dilley A et  al (2013) Comparison of percutaneous
endoscopic, laparoscopic and open gastrostomy inser-
tion in children. Pediatr Surg Int 29(6):613–621.
https://doi.org/10.1007/s00383-­013-­3313-­9
Lord LM (2018) Enteral access devices: types, function,
care, and challenges. Nutr Clin Pract 33(1):16–38.
https://doi.org/10.1002/ncp.10019
Merli L, de Marco EA, Fedele C, Mason EJ, Taddei A,
Paradiso FV et  al (2016) Gastrostomy placement in
children: percutaneous endoscopic gastrostomy or
laparoscopic gastrostomy? Surg Laparosc Endosc
Percutan Tech 26(5):381–384. https://doi.org/10.1097/
SLE.0000000000000310
Pang T, Sesia SB, Holland-Cunz S, Mayr J (2017)
Replaceable jejunal feeding tubes in severely ill chil-
dren. Gastroenterol Res Pract 2017:2090795. https://
doi.org/10.1155/2017/2090795
Petrosyan M, Khalafallah AM, Franklin AL, Doan T,
Kane TD (2016) Laparoscopic gastrostomy is supe-
rior to percutaneous endoscopic gastrostomy tube
placement in children less than 5 years of age. J
Laparoendosc Adv Surg Tech A 26(7):570–573.
https://doi.org/10.1089/lap.2016.0099
Powers J, Luebbehusen M, Spitzer T, Coddington A,
Beeson T, Brown J, Jones D (2011) Verification
of an electromagnetic placement device com-
pared with abdominal radiograph to predict
accuracy of feeding tube placement. JPEN J
Parenter Enteral Nutr 35(4):535–539. https://doi.
org/10.1177/0148607110387436
Rahimi-Movaghar V, Boroojeny SB, Moghtaderi A,
Keshmirian B (2005) Intracranial placement of
a nasogastric tube. A lesson to be re-learnt? Acta
117
Neurochir 147(5):573–574; discussion 574. https://
doi.org/10.1007/s00701-­004-­0463-­5
Ricciuto A, Baird R, Sant’Anna A (2015) A retrospec-
tive review of enteral nutrition support practices at
a tertiary pediatric hospital: a comparison of pro-
longed nasogastric and gastrostomy tube feeding.
Clin Nutr 34(4):652–658. https://doi.org/10.1016/j.
clnu.2014.07.007
Robertson FM, Crombleholme TM, Latchaw LA, Jacir
NN (1996) Modification of the “push” technique for
percutaneous endoscopic gastrostomy in infants and
children. J Am Coll Surg 182(3):215–218
Rothenberg SS, Bealer JF, Chang JH (1999) Primary
laparoscopic placement of gastrostomy buttons for
feeding tubes. A safer and simpler technique. Surg
Endosc 13(10):995–997. https://doi.org/10.1007/
s004649901154
Simons M, Bulat R (2021) Enteric tube placement
for gastroparesis: gastrostomy, gastrojejunostomy
and jejunostomy. Elsevier. https://doi.org/10.1016/
b978-­0-­12-­818586-­5.00029-­6
Srinivasan R, Irvine T, Dalzell M (2009) Indications
for percutaneous endoscopic gastrostomy and
procedure-­ related outcome. J Pediatr Gastroenterol
Nutr 49(5):584–588. https://doi.org/10.1097/
MPG.0b013e31819a4e8c
Tsujimoto H, Tsujimoto Y, Nakata Y, Akazawa M,
Kataoka Y (2017) Ultrasonography for confirmation
of gastric tube placement. Cochrane Database Syst
Rev 4:CD012083. https://doi.org/10.1002/14651858.
CD012083.pub2
Vermilyea S, Goh VL (2016) Enteral feedings in
children: sorting out tubes, buttons, and formu-
las. Nutr Clin Pract 31(1):59–67. https://doi.
org/10.1177/0884533615604806
Zimmer J, Gauderer MWL (2020) Access for enteral
nutrition. In: Puri P (ed) Encyclopedia of pediatric
surgery. Springer, Berlin, Heidelberg. https://doi.
org/10.1007/978-­3-­642-­38482-­0_19-­1
 Hematological Problems
in Pediatric Surgery
Peter McCarthy and Owen Patrick Smith
11.1 Introduction
Pediatric hematology is a broad specialty. It
encompasses disorders of white cells such as
severe congenital neutropenia, disorders of red
cells such as sickle cell disease, and disorders of
platelets such as immune thrombocytopenia pur-
pura. In addition, deficiencies of the coagulation
proteins required to ensure normal hemostasis,
including hemophilia A and von Willebrand dis-
ease, fall under the remit of the hematologist.
Transfusion medicine and advice on the appro-
priate use of blood products for patients is also an
important role (Arcesi et  al. 2006; Wynn et  al.
2017).
Disorders of the blood can be broadly grouped
into inherited or acquired, benign or malignant,
acute or chronic disorders with varied incidence,
etiology, pathogenesis, and outcomes.
Our understanding of hematological condi-
tions has advanced considerably due to signifi-
cant advances made in the field of molecular
biology, and pediatric hematologists have
worked hard to ensure that the management of
these conditions has kept up with these rapid
changes. Such advances aid in more accurate
P. McCarthy · O. P. Smith (*)
National Children’s Cancer Service, Children’s
Health Ireland at Crumlin, Dublin, Ireland
Systems Biology Ireland, School of Medicine,
University College Dublin, Dublin, Ireland
e-mail: peter.mccarthy@ucdconnect.ie;
owen.smith@olchc.ie
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_11
11
diagnosis, prognostication, and risk stratification
and ultimately lead to personalized therapeutic
approaches.
The majority of the pediatric hematologist’s
inpatient workload concerns the treatment of
malignancy, usually acute leukemia, but also
high-grade lymphomas (such as Burkitt lym-
phoma). In contrast, consultative hematology is
often centered around children with coagulation
abnormalities or the hematological manifesta-
tions of systemic diseases. The hematology day
ward is a busy place, where children with hemo-
globinopathies attend for transfusion, patients
with hemophilia attend for factor replacement
following a bleed, and children on chemotherapy
attend for review and supportive care.
In addition to clinical work, the pediatric
hematologist also heads the hematology labora-
tory, reporting on blood films and bone marrow
biopsies, overseeing the automated analyzer
results, pre-transfusion compatibility testing, and
coagulation testing. They are also tasked with
ensuring that a quality control system is in place
for delivering precise, accurate, and internation-
ally standardized blood test results for all users of
the laboratory.
Children with hematological problems may
require surgical input throughout their disease
course. At the time of diagnosis, they may require
surgical biopsy or central line insertion. Surgical
treatment may be required for complications of
their underlying condition, such as a neutropenic
119
120
patient on treatment for leukemia presenting with
an acute abdomen. Patients may also present with
surgical problems unrelated to their underlying
condition, the management of which may be
complicated by their hematological abnormality.
Surgical patients may manifest hematological
disturbances such as the neonate presenting with
a rapidly enlarging vascular mass who is noted to
be severely thrombocytopenic or the postopera-
tive child who develops wound sepsis, oozing
from cannulation sites, and disseminated intra-
vascular coagulation.
A knowledge of hematological basic science
and the pathophysiology and management of
common hematological disorders (McCann et al.
2005) is a prerequisite for a competent pediatric
surgeon (Oldham 2005).
11.2 Historical Overview
The history of hematology dates back to ancient
Egyptian times and the use of blood-letting tools.
It was not until the seventeenth century the first
microscope was produced, which allowed for the
identification of individual blood cells. It took
another 200  years before the first recorded suc-
cessful blood transfusion between humans
occurred in 1818. However, it required a further
100 years of research and discovery before blood
grouping was introduced.
The specialty of hematology has seen signifi-
cant advances in the last 60 plus years. There has
been explosive progress in the field of molecular
biology, which has revealed fundamental insights
into the causes of many hematological conditions
and sparked the development of numerous treat-
ments that continue to transform patient out-
comes today.
Examples of such advances include safer and
more effective blood transfusion products, the
development of recombinant growth factors that
stimulate blood cell production, hematopoietic
P. McCarthy and O. P. Smith
stem cell transplant procedures, and the discov-
ery of a plethora of targeted molecular therapies
for hematological malignancies and chimeric
antigen receptor T (CART) cell therapy.
11.3 Incidence
The incidence of hematological problems varies
widely depending on the underlying disorder.
While individually, many of these conditions are
rare in children, cumulatively they represent a
significant burden of disease.
Hemoglobinopathies (such as sickle cell dis-
ease) are inherited disorders of hemoglobin pro-
duction and function. Sickle cell disease is one of
the most frequent inherited genetic blood disor-
ders in the world. It mainly affects people of
African ancestry (approximately 80% of sickle
cell disease cases are believed to occur in sub-­
Saharan Africa) as well as people from the mid-
dle east, India, and Mediterranean regions.
Bleeding disorders are another important
group of inherited hematological disorders. The
most common of these is von Willebrand disease
(autosomal inheritance), estimated to affect 1 in
5000 young people. Hemophilia A (X-linked
inheritance) is the second most common inher-
ited bleeding disorder, impacting 1  in 10,000
patients per year.
There is great variation in the incidence of
childhood cancer internationally, and it is esti-
mated that over 140 cases per million person-­
years occur in children aged 0–14  years. These
vary between high-income countries and low-­
and-­middle-income countries. The most common
childhood cancer is acute leukemia (accounting
for almost one-third of cases), followed by cen-
tral nervous system tumors. The survival rates
from cancers such as acute lymphoblastic leuke-
mia (ALL) are much improved, with greater than
90% of children alive five years after diagnosis in
high-income countries.
11  Hematological Problems in Pediatric Surgery
11.4 Etiopathogenesis
11.4.1 Blood Formation
(Hematopoiesis)
The bone marrow is a mesenchymal derived tis-
sue divided into irregular interconnective spaces
by bone trabeculae. It consists of a complex
hematopoietic cellular component that is
extremely labile and continuously goes through
self-replication and differentiation processes.
These cells are supported by a micro-­environment
composed of stromal cells (endothelial cells,
fibroblast-like cells, adipocytes), extracellular
matrix, and vascular structures.
• Embryonic hematopoiesis (which predomi-
nantly is associated with red cell ­development)
begins in the yolk sac at the end of the third
week of gestation and declines to an insignifi-
cant level by the end of the first trimester.
• By the end of the first trimester, the liver is the
dominant source of hematopoiesis producing
all the hemopoietic elements. Hepatic hemato-
poietic activity reaches its maximum level at
around the third month and gradually declines
from the seventh month until birth. Bone mar-
row hematopoiesis begins to occur at around
the fifth month of gestation and continues to
increase thereafter.
• Every day, in normal adult bone marrow,
approximately 2.5 billion red cells, 1 billion
granulocytes, and 2.5 billion platelets are pro-
duced per kilogram of body weight.
• Hematopoiesis is regulated and sustained by a
complex cellular interaction of hemopoietic
and stromal elements and a network of cyto-
kine growth factors, including interleukins
and colony-stimulating factors.
• All the cells of the hematopoietic system orig-
inate from a pluripotent hemopoietic stem cell
(HSC). HSCs have the intrinsic capacity for
self-renewal and are low in number and divide
infrequently. The committed progenitors are
responsible for the massive amount of cell
121
proliferation required to maintain blood cell
production in numbers outlined above. The
common lymphoid progenitors develop into T
and B cells, whereas the common myeloid
progenitor gives rise to erythrocytes, mega-
karyocytes, monocytes, and granulocytes.
11.4.2 Mechanisms of Hemostasis
Normal blood coagulation or hemostasis is a
complex sequence of interrelated events by which
the body prevents blood loss from the vascular
tree. This is achieved by a multipathway interac-
tive system with multiple negative and positive
feedback loops, which ultimately ensure that
blood is at all times fluid within the vasculature,
but it also needs the ability to transform into a
clot when there is a breach in the integrity of the
vascular tree. The protein (pro- and anticoagu-
lants, outlined below) and cellular (endothelial
cells, monocytes, and platelets) components have
also shown to be intimately involved in the
inflammatory response, vasculogenesis, metasta-
sis, cellular proliferation, and tissue repair.
• Tissue factor, a cell surface glycoprotein is the
principal biological initiator of blood
coagulation.
• Exposure of circulating plasma factor VIIa to
tissue factor triggers the coagulation cascade
in vivo, which results in thrombin generation
(Fig. 11.1).
• Thrombin converts soluble fibrinogen to a
fibrin network, activates platelets, and stimu-
lates coagulation by positive feedback activa-
tors of cofactors, factors V and VIII, and the
zymogens II, VII, IX, X, XII, and XIII.
• Under physiological conditions, pro- and anti-
coagulant (see below) mechanisms are bal-
anced in favor of anticoagulation; however, at
sites of vascular damage resulting from
inflammation, trauma, etc., the anticoagulant
system is downregulated and thus procoagu-
lant forces prevail.
122
TF
+
VIIa
X
Va
PT
Xa
II IIa [ Thrombin ]
TT
I Ia [ Fibrin ]
I Insoluble
[ Fibrinogen ]
insoluble
Fig. 11.1  Blood coagulation is initiated (initiation phase)
when tissue factor (TF), expressed after injury to cell
(endothelial, monocytic cells, etc.) wall, is exposed to
FVIIa in the bloodstream. TF-FVIIa complex in turn acti-
vates FIX to FIXa and FX to FXa. FIXa with its cofactor
FVIIIa in turn also activates FX to FXa (amplification
phase). FXa with its cofactor FVa activates prothrombin
11.4.3 Natural Anticoagulation
Control Mechanisms
Several natural anticoagulant mechanisms have
been discovered that exert dampening effects
upon pro-coagulation and in turn halt the genera-
tion of thrombin. The major anticoagulant inhibi-
tors of blood coagulation include tissue pathway
inhibitor, antithrombin, and the protein C path-
way (Fig. 11.2). The protein C pathway regulates
the amount of thrombin in the micro-circulation,
whereas the tissue factor pathway inhibitor and
the antithrombin exert more of an effect in the
macro-circulation.
11.4.4 Platelets
Platelets derived from megakaryocytes play an
essential role in thrombosis and hemostasis.
P. McCarthy and O. P. Smith
IXa
VIIIa
APTT
Pro-thrombotic
Pro-inflammatory
Pro-angiogeneic
(II) to thrombin (IIa) (propagation phase). Thrombin con-
verts soluble fibrinogen to insoluble fibrin. Thrombin is
not only prothrombotic but activates platelets and is proin-
flammatory and promotes new vessel formation. Shown in
gray are three global coagulation screens prothrombin
time (PT), activated partial thromboplastin time (APTT),
and thrombin time (TT) (Puri and Höllwarth 2009)
Megakaryopoiesis in the bone marrow is
­governed by a complex interaction of cytokines
(e.g., thrombopoietin) and their receptors (e.g.,
c-mpl). Platelets play a major role in primary
hemostasis (adhesion, aggregation, and release).
Platelets may also have functional defects. These
disorders cause primary hemostatic bleeding
(mucocutaneous) ranging from mild to severe.
11.4.5 Blood Groups and Antibodies
There are 38 blood group systems corresponding
to red blood cell (RBC) surface antigens. The
ABO system is the most important. Antibodies
against blood group antigens A and B occur natu-
rally in people who lack these antigens; for
example, patients who are Group A will have
naturally occurring anti-B antibodies, while
patients who are Group O will have anti-A and
11  Hematological Problems in Pediatric Surgery 123

TF TFPI leukemia or lymphoma. The relevant pathophysi-

+
ology for each group of conditions will be dis-
cussed below.
VIIa
IXa

X
VIIIa 11.6 Hematological Disorders
Va
Encountered in Pediatric
AT+ Practice: A Surgical

GAG
Perspective
APC
Xa We will review a number of hematological disor-
PS ders encountered in pediatric practice from a sur-
EPCR gical perspective.
II IIa [ Thrombin ]
Tm

PC
11.6.1 Inherited Disorders
I Ia [ Fibrin ] of Coagulation
[ Fibrinogen ] Insoluble
insoluble
11.6.1.1 Hemophilia
Fig. 11.2  The initiation phase of coagulation is con- Factor VIII deficiency (hemophilia A) is the sec-
trolled by inhibiting the complex of TF, FVIIa, and FXa
ond most common inherited bleeding disorder
by tissue factor pathway inhibitor (TFPI). The amplifica-
tion phase of coagulation is blocked by the protein C path- with a frequency of approximately 1 in 5000 live
way. Protein C (PC) is activated by a complex of thrombin, male births. Factor IX deficiency (hemophilia B)
thrombomodulin (Tm), and endothelial protein C receptor is approximately one-sixth as common.
(EPCR) to APC, which in association with protein S (PS)
inactivates FVa and FVIIa. The thrombin formed in the
Diagnosis
propagation phase is controlled by antithrombin (AT)
(Puri and Höllwarth 2009) • The majority of severe and moderate (factor
VIII levels <1% and 1–5%, respectively) cases
anti-B and so on. These antibodies are predomi- present in the first few years of life.
nantly IgM and so can cause complement activa- • Hemophilia A and B are X-linked recessive
tion and acute intravascular hemolysis. Other disorders. Thus, males are affected and
clinically important antibodies do not occur natu- females can be carriers.
rally and require exposure to an antigen—either • One-third of cases of FVIII Deficiency have
by a blood product transfusion or via feto-­ no family history (spontaneous mutations).
maternal exposure. These antigens vary in their • When there is no family history, infants with
immunogenicity; e.g., the RhD and the Kell anti- moderate or severe disease usually present
gen are highly immunogenic and may lead to with:
hemolytic disease of the fetus and newborn –– Post-circumcision bleeding (Fig. 11.3).
(HDFN). –– Bad “toddler bruising”.
–– Soft tissue and muscle or joint bleeds at
6–18 months of age.
11.5 Pathophysiology –– Intracranial, iliopsoas, intra-abdominal
bleeding, or hematuria occasionally may
Hematological issues can arise due to inherited occur.
or acquired defects in normal hematopoiesis or • In children presenting with bruising and
hemostasis. They can also occur due to acquired severe bleeding, it is not uncommon for the
mutations leading to malignancy, such as acute first presumed diagnosis to be nonaccidental
124 P. McCarthy and O. P. Smith

a b

Fig. 11.3  Hemophilia A: (a) post-circumcisional hema- 22 mutation. (b) Full hematoma resolution 1  week later
toma in an infant with no family history of hemophilia. following replacement with recombinant FVIII (Puri and
Mutational analyses showed the presence of the inversion Höllwarth 2009)

injury (NAI). True coagulation defects need to

be excluded. A normal coagulation screen
does not exclude all significant coagulation
disorders, e.g., von Willebrand disease, FXIII
deficiency, and platelet function defects. The
presence of a bleeding disorder does not mean
that NAI is excluded.

Differential Diagnosis
Other inherited bleeding disorders should be con-
sidered—such as von Willebrand’s disease (see
below). If an acute presentation with no bleeding
history, acquired causes should also be consid-
ered—such as vitamin K deficiency or an
acquired inhibitor of a clotting factor.

Management
• The objectives of modern hemophilia man-
agement include the following:
–– Prevention of chronic joint damage
(Fig. 11.4).
–– Prevention of “life-threatening” bleeds.
–– Facilitation of social and physical well-­
being and helping children to achieve their Fig. 11.4  Hemophilia B: chronic severe hemophiliac
full potential. arthropathy of the right knee joint. The quadricep muscle
is severely wasted. This adolescent was only treated inter-
–– To provide a comprehensive service to the mittently with plasma throughout the first 5 years of life
family (including genetic counseling, train- (Puri and Höllwarth 2009)
11  Hematological Problems in Pediatric Surgery
ing in self-administration of factor, educa-
tion on how to prevent and manage bleeds).
• The multidisciplinary team is crucial to ensure
satisfactory achievement of all these
objectives.
• Successful treatment involves prompt and suf-
ficient intravenous replacement of FVIII/FIX
to hemostatic levels (Berntorp et  al. 2017;
Rayment et al. 2020).
• Prophylactic administration of factor concen-
trate converts a child who has severe or mod-
erately severe disease to a child with mild
disease. The child should no longer have
spontaneous bleeds, and prophylaxis protects
against the development of hemophilic
arthropathy. Because of the different plasma
half-lives of coagulation factors, dosing is
based on individualized pharmacokinetics.
Prophylaxis can be timed immediately prior to
sporting activities, thus allowing the child to
live a relatively normal life (Keeling et  al.
2008; Rayment et al. 2020).
• Prophylaxis usually requires a central venous
access device (see below) to be placed in the
child to facilitate regular intravenous
administration.
• Extended half-life recombinant factor concen-
trates are now becoming more commonly
used as standard treatment. These longer-­
acting FVIII and FIX products have proven
safe and efficacious at preventing bleeding in
both severe FVIII and FIX deficiency. They
reduce the need for frequent IV access and are
likely to improve compliance with therapy,
thus reducing hemophilia-related complica-
tions (Rayment et al. 2020).
• Minor surgical procedures such as endoscopic
biopsies will require factor levels to be brought
up to 100% of normal levels by giving a bolus
dose of factor concentrate immediately preop-
eratively. More major procedures, such as
cardiac surgery, may require a continuous
­
infusion of factor concentrate. Levels are
monitored periodically, and further doses are
titrated accordingly. All patients should be
discussed with their coagulation center prior
to any surgical interventions or procedures.
An individual management plan will be put in
place for each patient in advance.
125
• The formation of neutralizing antibodies to
FVIII/IX (known as inhibitors) is the single
biggest complication in the modern manage-
ment of hemophilia, and 15–20% of patients
will develop an inhibitor. These patients can-
not usually be treated with FVIII or FIX any-
more but require treatment with bypassing
agents such as FEIBA or NovoSeven®.
FEIBA is plasma derived and so theoretically
has a risk of transmission of blood-borne dis-
eases. NovoSeven® (recombinant FVIIa) is
very costly. Neither agent obtains hemostasis
as readily as factor replacement can in a
patient without inhibitors. No routine labora-
tory test is available to monitor treatment
response, and thus efficacy must be judged
clinically. Immune tolerance induction ther-
apy is used to eradicate inhibitors soon after
they develop (Astermark et al. 2007).
• The novel agent Emicizumab (a bispecific
antibody that binds activated FIX to FX)
replaces the normal function of FVIII and is
useful in the management of patients with
inhibitors to FVIII. It is now also being inves-
tigated as first-line prophylaxis for some pedi-
atric patients without inhibitors. It has the
advantage of subcutaneous administration
compared to recombinant factor and converts
the patient’s phenotype from severe to mild/
moderate.
• Gene therapy for FIX deficiency has been suc-
cessfully trialed in a number of patients. A
viral vector introduces the FIX gene into
hepatocytes. Sustained production of FIX has
ensued, which converts the patient’s pheno-
type from severe to mild or moderate, such
that they may no longer require regular pro-
phylaxis. Progress in FVIII gene therapy is not
quite so advanced as the FVIII gene is larger
and more difficult to introduce into the host
using a viral vector.
11.6.1.2 Central Venous Access
Devices
The intravenous administration of factor concen-
trates up to two or three times per week is fraught
with difficulty in the majority of young children
when only using peripheral veins. Similarly,
immune tolerance induction therapy using large
126
doses of factor concentrate, twice a day for
12–24 months for children with inhibitors to fac-
tor VIII/IX, is almost impossible without regular
venous access. These devices can be fully
implantable (PortaCath™, Deltac USA) or partly
externalized (single or double lumen Quintan™
Catheters). The use of a port is preferable to an
external device because it causes fewer limita-
tions to the child’s lifestyle, and it has been sug-
gested that there is a lower infective risk.
However, despite the obvious attractions of these
devices, they carry the risks of thrombosis and
infection, both of which may lead to morbidity/
mortality and permanent removal of the device.
The rate of infection is higher in children with
inhibitors.
There is now a growing consensus that long-­
term indwelling devices are necessary to facili-
tate the modern intensive treatment of congenital
coagulation disorders. With improved manage-
ment of the perioperative period and regular, fre-
quent re-education, particularly in those children
with inhibitors, many of the complications can be
avoided. CVADs are generally removed after the
age of 5  years following the development of
robust peripheral veins that can be readily
accessed for the administration of treatment.
More uncommonly, the use of arteriovenous
fistulae (AVF) as a reliable means of vascular
access in children with hemophilia has been
reported. Complication rates are reported to be
minimal. The vast majority of children (>95%)
achieved functional AVF that are still regularly
used for home treatment over a median period of
29 months, suggesting that this approach, i.e., the
creation of AVF as the first option for achieving
permanent venous access in children with severe
hemophilia, is warranted.
The need for venous access for prophylactic
treatment is reducing as the use of newer agents,
such as Emicizumab for FVIII deficiency, which
are delivered subcutaneously, becomes more
widespread.
11.6.1.3 von Willebrand Disease
(VWD)
von Willebrand factor (VWF) is a large plasma
protein involved in tethering platelets to the cap-
P. McCarthy and O. P. Smith
illary walls at sites of minor vessel injury (pri-
mary hemostasis). It also protects FVIII from
premature proteolytic cleavage within the
circulation.
Quantitative defects in VWF are termed type 1
VWD if partial and type 3 VWD if complete (i.e.,
no circulating VWF). Qualitative defects in VWF
are termed type 2 VWD (there are four subtypes).
Patients with VWD may or may not have an
abnormal platelet count or APTT.
Most patients with VWD are diagnosed in
adulthood. Children may come to clinical atten-
tion after positive screening results with a family
history of the disorder, or they may be investi-
gated after excessive surgical bleeding or muco-
cutaneous hemorrhages such as easy bruising,
gum bleeding, epistaxis, or heavy menstrual
bleeding.
Treatment will vary depending on the under-
lying defect and VWD type. Some patients
without a significant personal or family bleed-
ing history and mild functional deficiencies of
VWF may require no treatment prior to minor
procedures. For other patients, tranexamic acid
preoperatively and for 5–7 days postoperatively
may be sufficient. A DDAVP infusion can be
used for minor procedures in patients with mild
disease. Some patients will require plasma-
derived VWF concentrates such as Wilate® pre-
and postoperatively.
11.6.2 Platelet Disorders
The normal range of the platelet count in child-
hood is similar to that seen in adult life, being
about 150–400 × 109/l.
11.6.2.1 Differential Diagnosis
The causes of thrombocytopenia can be divided
into two broad categories: those arising from the
background of an established genetic defect
(inherited thrombocytopenia) and those that are
acquired. Neonatal thrombocytopenia may have
been acquired in the antenatal or perinatal period
or may also be due to inherited thrombocytope-
nia. Inherited thrombocytopenia is sometimes
accompanied by dysfunctional platelets (Bolton-­
11  Hematological Problems in Pediatric Surgery 127

Table 11.1  Inherited causes of thrombocytopenia (Puri

and Höllwarth 2009)
Disorders of platelet number
MYH9 disorders, e.g., May–Heggling anomaly
Congenital amegakaryocytic thrombocytopenia
Amegakaryocytic thrombocytopenia with radioulnar
synostosis
Thrombocytopenia—absent radius syndrome
X-linked thrombocytopenia with dyserythropoiesis
Severe disorders of platelet function
Bernard–Soulier syndrome
Glanzmann’s thrombasthenia
Wiskott–Aldrich syndrome
Disorders of receptors and signal transduction
Platelet cyclooxygenase deficiency
Thromboxane synthase deficiency
Thromboxane A2 receptor defect
ADP receptor defect (P2Y12)
Disorders of platelet granules
Idiopathic dense granule disorder
Hermansky–Pudlak syndrome
Chediak–Higashi syndrome
Gray platelet syndrome
Paris–Trousseau syndrome
Idiopathic a and d granule storage pool disease
Disorders of phospholipid exposure
E.g., Scott syndrome

Maggs et  al. 2006). Table  11.1 shows a list of
inherited causes of thrombocytopenia.
It is important to remember to confirm that the
low platelet count is genuine by careful inspec-
tion of the blood sample and smear to exclude
platelet clumps before initiating further
investigations.
Once established, the approach to the diag-
nosis of thrombocytopenia should be tailored to
the individual child or infant and mother if deal-
ing with neonatal thrombocytopenia. For exam-
ple, assessment of the child’s general well-being
is very important as healthy children usually
have an immune or an inherited etiology,
whereas the presence of lymphadenopathy, hep-
atosplenomegaly, mass lesions, hemangiomas
(Fig.  11.5), bruits, and congenital anomalies
points toward a totally different spectrum of
causes.
In neonates, it should also be emphasized
that obtaining a detailed maternal history,
including bleeding problems, pre-eclampsia,
and drug ingestion in the present and past preg-
Fig. 11.5  Capillary hemangioma in a 2-month-old boy
with Kasabach–Merritt syndrome (KMS). The lesion
involuted after 4 months of therapy involving vincristine,
prednisolone, and antiplatelet agents (aspirin and ticlopi-
dine) (Puri and Höllwarth 2009)
nancies and any history of viral infections (cyto-
megalovirus, rubella, herpes simplex, and HIV)
or connective tissue disease (systemic lupus
erythematosus [SLE]), will save time and
unnecessary investigation.
11.6.2.2 Immune Thrombocytopenia
Purpura (ITP)
ITP is defined as an isolated platelet count
<100  ×  109/l in the absence of other causes of
thrombocytopenia. It is the most common cause
of thrombocytopenia in childhood.
Diagnosis
• Predominantly seen in children aged between
2 and 5 years and is generally preceded by a
viral illness or prodrome. It usually presents in
the autumn and winter months.
128
• Bleeding is uncommon when the platelet
count is >50 × 109/l.
• Spontaneous bleeding remains uncommon but
more frequent when the platelet count is
<30 × 109/l.
• Diagnosis is one of exclusion in that the vast
majority of children will have had a preceding
viral infection or will have been vaccinated in
the previous month.
• Examination is usually normal with the excep-
tion of petechiae or mucocutaneous bleeding.
Laboratory workup is usually limited to FBC
(showing isolated thrombocytopenia), coagu-
lation screen, renal/liver/bone profile, and
LDH, all of which are otherwise normal. A
blood smear should be normal with the excep-
tion of thrombocytopenia.
• Bone marrow examination is not generally
required for the diagnosis to be made.
Pathophysiology
• Antiplatelet antibody formation leads to pre-
mature platelet destruction in the spleen and
also impaired platelet production from
megakaryocytes.
• T-cell mediated toxicity also contributes to
platelet destruction.
• There is no role for antibody screening or
measurement.
Management (Neunert et al. 2019)
• The majority of children will resolve sponta-
neously within 6–12  months without any
intervention.
• The rate of major bleeding is very low (<3%)
even when the child is severely thrombocyto-
penic. For these reasons, management of chil-
dren with newly diagnosed ITP presenting
with cutaneous bleeding alone, even with a
platelet count of <20  ×  109/l is usually
conservative.
• Platelet transfusions are ineffective due to
rapid platelet consumption and are generally
contraindicated.
• If treatment is required (e.g., more serious
bleeding at diagnosis, other risk factors for
bleeding, parental anxiety), then intravenous
P. McCarthy and O. P. Smith
immunoglobulin and prednisolone are front-­
line treatments. The majority of children will
respond but may have side effects.
• Some lifestyle restrictions are also required
until the platelet count rises above
50–75 × 109/l, e.g., avoidance of contact sports
along with rough play. Parents should be
warned to present urgently with any new
bleeding and to avoid nonsteroidal anti-­
inflammatory drugs (NSAIDs), such as ibu-
profen, which interfere with platelet function
and thus increase the risk of bleeding.
• For children who fail to respond, second-line
therapeutic interventions include thrombopoi-
etin receptor agonists (see below), anti-D
immunoglobulin, cyclosporin, azathioprine,
and other immune-modulating agents.
• Romiplostim and eltrombopag are novel drugs
that stimulate the thrombopoietin receptor and
increase platelet production by megakaryo-
cytes. Although they produce a sustained
platelet response in 60–70% of patients,
relapse tends to occur on treatment cessation.
They are used in children with persistent or
chronic ITP. There is a small risk of bone mar-
row fibrosis; thus, screening bone marrow
biopsies may be required for children on long-­
term treatment.
• Up to one-third of children may develop a
chronic form of ITP (thrombocytopenia last-
ing greater than 12 months) that can be symp-
tomatic. Treatment can be problematic and
hence splenectomy is worth considering. Also,
80% of children with chronic ITP will remain
in remission after 4  years. The downsides to
splenectomy are discussed below.
• Emergency or life-threatening hemorrhage
requires a different treatment approach, as
some agents require weeks or months before a
response is seen. Rapid-acting treatments
include intravenous immunoglobulin (effect
seen within 24–48  h), high dose IV methyl-
prednisolone, IV vincristine, emergency sple-
nectomy, or some combination of those above.
Platelet transfusions are occasionally used in
the emergency setting only and following dis-
cussion with the hematology team.
11  Hematological Problems in Pediatric Surgery
11.6.3 Disseminated Intravascular
Coagulation (DIC)
DIC is a clinicopathological entity resulting in
simultaneous and unregulated activation of the
coagulation and fibrinolytic pathways. It is a syn-
drome of serious clinical consequences, which is
encountered in all areas of pediatrics, particularly
in the intensive care unit.
11.6.3.1 Diagnosis
DIC is not a primary disease entity; it is second-
ary to an underlying usually severe and most
often systemic illness.
–– In DIC with hemorrhage, bleeding is typically
from multiple sites, indicating the systemic
nature of the process.
–– Purpura fulminans is seen in disseminated
meningococcemia. The skin lesions appear
hemorrhagic; however, microthromboses are
the underlying histological findings (Thachil
et al. 2012) (Fig. 11.6).
–– Diagnostic workup in DIC includes PT (pro-
longed), APTT (prolonged), fibrinogen
(reduced), D Dimer (raised), platelet count
(reduced), blood smear, and natural anticoag-
ulant factor levels.
11.6.3.2 Differential Diagnosis
DIC can be secondary to a number of different
conditions, with sepsis (of many potential etiolo-
gies) being the most common in all age groups.
Neonates can develop the condition following a
Fig. 11.6 Purpura fulminans secondary to severe
acquired protein C deficiency in association with menin-
gococcal septicemia (Puri and Höllwarth 2009)
129
perinatal insult such as birth asphyxia, meconium
aspiration syndrome, and respiratory distress
syndrome. Older infants and children can develop
DIC due to trauma (neurological, crush injury,
massive burns), malignancy (acute lymphoblastic
leukemia, acute promyelocytic leukemia), toxin
exposure (snake venom), immunological insults
(acute hemolytic transfusion reaction, transplant
rejection, autoimmune disease), and others.
11.6.3.3 Pathophysiology
DIC is a progressive, pathological process
resulting in profuse thrombin formation and
excessive activity of the fibrinolytic pathway,
and its most prominent clinical feature is a
bleeding tendency.
11.6.3.4 Management (Rajagopal
et al. 2017)
–– The patient and not the laboratory numbers
should be treated; i.e., blood product replace-
ment is only required in the setting of active
bleeding or prior to surgical intervention.
–– In the above settings, fresh frozen plasma
(FFP) is used if the PT and APTT are >1.5
times the upper limit of normal; fibrinogen
concentrate is used if fibrinogen is <1.0  g/dl
and platelets may be transfused to keep plate-
let count >50 × 109/l.
–– Other forms of intervention such as heparin
and natural anticoagulant concentrates have
been used in the past, but the current evidence
does not support their routine use or
recommendation.
–– Since DIC is not a primary disease entity,
treatment should be directed toward the under-
lying process causing the consumption.
11.6.4 Thrombotic Disorders
Thrombotic disease in children is much less com-
mon than in adults. Idiopathic thrombosis in chil-
dren is rare, and 95% of pediatric venous
thromboembolisms (VTEs) occur in the context
of serious illness. However, the incidence of
childhood VTE is increasing, largely because of
better imaging modalities and advances in neona-
130 P. McCarthy and O. P. Smith

tal care and tertiary pediatric care such as ECMO,
cardiopulmonary bypass, hemodialysis, and the
use of intra-arterial and intravenous indwelling
catheters.
11.6.4.1 Diagnosis
The peak incidence of thrombotic events is seen
in the neonatal period. When seen in childhood, it
is likely secondary to an acquired prothrombotic
state (Table 11.2) or much less likely due to an
inherited thrombotic tendency (Table  11.3).
When it does occur in childhood, it can be fatal or
associated with several sequelae such as amputa-
tion, organ dysfunction, and post-phlebitic
syndrome.
Children with idiopathic thrombosis should be
considered for screening for an inherited gene
defect predisposing to clot formation. The preva-
lence of thrombophilic defects in children with
VTE varies between studies from 10% to 78%. It
is unclear whether children presenting with sec-
ondary VTE, e.g., CVAD-related thromboses,
should have thrombophilia testing performed. It
is expensive, can cause significant concern for

Table 11.2  Acquired throm- Indwelling vascular catheters

botic tendency (Puri and Renal artery and vein thrombosis
Höllwarth 2009)
Acquired natural anticoagulant Nephrotic
deficiency syndrome → antithrombin
deficiency
Purpura fulminans → varicella
& protein S deficiency and
meningococcemia & protein C
deficiency
Necrotizing enterocolitis (NEC)
Respiratory distress syndrome
Heparin-induced thrombocytopenia/
thrombosis syndrome (HIT/HITTs)
Maternal anticardiolipin antibodies
(lupus anticoagulant)
Extracorporeal membrane
oxygenation (ECMO)
Hemolytic uremic syndrome/
thrombotic thrombocytopenic
purpura (HUS/TTP)
Birth asphyxia

Table 11.3  Inherited thrombotic High circulating levels of FVIII

tendency (Puri and Höllwarth Protein C deficiency
2009)
Protein S deficiency
Defects within the protein APCR and FVR506Q (factor V Leiden)
C pathway FIIG20210A (prothrombin gene variant)
Hyperhomocysteinemia Cystathionine B-synthase
Methionine synthase
Thermolabile
methylenetetrahydrofolate reductase
Antithrombin deficiency
Fibrinolytic pathway PAI-1 (4G/5G polymorphic status)
Plasminogenemia
Dysfibrinogenemia
Hemoglobinopathy
Platelet defects
11  Hematological Problems in Pediatric Surgery 131

parents and other family members, and does not c­ ardiopulmonary bypass and other such pro-
affect clinical management of the child and so is cedures. LMWH is given subcutaneously,
often felt to be unnecessary. usually once daily (Monagle and Newall
2018).
Differential Diagnosis • In more specific disease states such as inher-
The differential diagnoses for thrombotic condi- ited or acquired protein C or antithrombin
tions in children can be either acquired or inher- deficiency, factor concentrate replacement is
ited (see Tables 11.2 and 11.3). sometimes used.
• For children who develop heparin-induced
11.6.4.2 Management thrombocytopenia (HIT) (suspect if high 4 T
The indications for the use of anticoagulants in score), a rare prothrombotic complication of
infants and children have changed dramatically heparin therapy, direct thrombin inhibitors, or
over the past 20 years with the major advances in a heparinoid should be considered following
tertiary pediatric care described above. The fol- discussion with hematology.
lowing should be considered: • A wide array of direct oral anticoagulants
(DOACs) that do not require regular labora-
• Choice of anticoagulants is dependent on the tory monitoring are now available for the
duration of anticoagulation required. treatment of VTE in adults. Many of these
• Unfractionated heparin (UFH), low-­ agents are currently being trialed in children,
molecular-­weight heparin (LMWH), and vita- e.g., dabigatran, apixaban, and rivaroxaban.
min K antagonists (VKAs), e.g., warfarin, Unlike warfarin (prothrombin complex con-
have all been used safely in childhood. centrate) and dabigatran (idarucizumab), the
Heparin in children is monitored with Anti-Xa reveal agent for inhibitors of FXa such as riva-
levels. Warfarin is monitored using the inter- roxaban and apixaban (andexanet alfa) is not
national normalized ratio (INR). Home testing always readily available.
kits are available for parental use, with dosing
supervised by a warfarin clinic so that the
child does not have to attend the clinic for 11.6.5 Asplenia/Hyposplenism/
regular phlebotomy. Splenectomy
• VKAs take several days for the anticoagulant
effect to manifest; therefore, in the setting of The most common form of asplenia or hyposplen-
acute thrombosis, bridging anticoagulation is ism is surgical splenectomy. The usual hemato-
required with UFH or LMWH. logical indications for splenectomy include the
• UFH requires continuous intravenous infusion following:
plus frequent phlebotomy for monitoring pur-
poses. It is rapidly cleared in infants, meaning • Repeated splenic sequestration in children
that very large doses may be needed. In this with SCD.
population, less time is spent in the therapeu- • Thalassemia major with associated hyper-
tic range, and more bleeding complications splenism.
are seen than with use in adults. Antithrombin • Hereditary spherocytosis.
may need to be replaced concurrently with • Refractory immune cytopenias (e.g., chronic
UFH therapy to achieve a heparin effect in ITP).
certain patient groups. This should be guided
by local protocols and following discussion Splenectomy can be open or laparoscopic, the
with the hematology team. latter being preferred where appropriate facili-
• LMWH is the initial anticoagulant of choice ties and expertise exist. The downside of sple-
for most episodes of acute VTE, although nectomy is a lifelong risk of overwhelming
UFH is still used extensively during sepsis from encapsulated organisms such as
132
Streptococcus pneumoniae, Haemophilus influ-
enzae type B, and Neisseria meningitidis. There
is also a possible increased long-term risk of VTE
disease and pulmonary artery hypertension.
11.6.5.1 Diagnosis
Assessment of splenic filtration function is usu-
ally made by examination of the peripheral blood
for evidence of red cell inclusions, which are pit-
ted out during filtration by the normally function-
ing spleen. These inclusions include Howell Jolly
bodies and “pits” in red cells. The presence of
Howell Jolly bodies usually reflects significant
splenic hypofunction and the risk of overwhelm-
ing infection (Fig. 11.7).
Immune-mediated conditions associated with
functional hyposplenism include the following:
–– Chronic graft versus host disease (GvHD).
–– HIV/AIDS.
–– Coeliac disease/dermatitis herpetiformis.
–– Rheumatoid arthritis/SLE.
–– Thyroid disease.
–– Ulcerative colitis/Crohn’s disease.
In many of these diseases, there is not only
functional hyposplenism, but the spleen may also
become atrophied. This is seen especially fre-
Fig. 11.7 The arrowed red cell shows a dark dense inclu-
sion “Howell Jolly body”. Howell Jolly bodies are most
commonly seen in splenectomized (surgical or “auto”)
patients, severe forms of megaloblastic and hemolytic
anemias and hemoglobinopathies (Puri and Höllwarth
2009)
P. McCarthy and O. P. Smith
quently in patients with coeliac disease, the
majority of whom are adults.
11.6.5.2 Differential Diagnosis
Congenital absence of the spleen can be associ-
ated with multiple abnormalities, including car-
diovascular and visceral abnormalities, and some
of these have a genetic basis.
Loss of splenic substances as a result of infarc-
tion is seen in sickle cell disease and essential
thrombocytopenia, the latter being extremely rare
in children. These conditions are usually accom-
panied by functional hyposplenism.
11.6.5.3 Management
Vaccination against encapsulated organisms (to
include MenACWY, MenB, PCV13, Hib, and
PPV23, if not already received) should be per-
formed a minimum of 2 weeks prior to elective
splenectomy to minimize this risk or 2  weeks
after emergency splenectomy (Rubin and
Schaffner 2014). Patients should also receive the
annual influenza vaccine and PPV23 vaccination
every 5  years. Daily antibiotic prophylaxis
against pneumococcus is recommended for life
(Davies et  al. 2011). All patients should keep a
supply of therapeutic antibiotics at home for use
if they develop a febrile illness.
11.6.6 Anemia
Hemoglobin (Hb) values vary with age. Neonates
are relatively polycythemic with an Hb range of
150–210  g/l. This gradually drops to a nadir of
95–125 g/l by 2–3 months of age. Thereafter, the
Hb climbs such that the normal range is 110–
135 g/l from 1 year to puberty. The normal range
of Hb varies slightly between males and females.
Anemia can be caused by increased destruc-
tion or loss of red cells, e.g., hemolysis or acute
blood loss (see Table 11.4). This will be accompa-
nied within a few hours by a compensatory reticu-
locytosis; i.e., the reticulocyte count (the number
or proportion of immature red cells) will be high.
Conversely, a low or normal reticulocyte count in
the setting of anemia indicates inadequate bone
marrow activity or failure of red cell production.
11  Hematological Problems in Pediatric Surgery
Table 11.4  Causes of anemia (Puri 2016)
Increased red
cell
Decreased red cell production destruction
Nutritional: e.g., B12, folate, iron Acute blood
deficiency loss
Space related: bone marrow Hemolysis
infiltration, e.g., leukemia/other (see
malignancies Table 11.5)
Toxin related: e.g., chemotherapy
Hormone related: EPO deficiency with
chronic renal disease
Immune related: aplastic anemia, pure
red cell aplasia, transient
erythroblastopenia of childhood (TEC)
Acquired genetic disorders of
hematopoiesis: myelodysplastic
syndrome (MDS)
Hereditary genetic disorders of
hematopoiesis: Diamond–Blackfan
anemia, congenital dyserythropoietic
anemias
Cytokine related: anemia of chronic
disease, e.g., TB, connective tissue
disorders
11.6.6.1 Pathophysiology
Several factors are required for healthy erythro-
poiesis. Certain nutrients are prerequisites includ-
ing vitamin B12, folate, and iron. A certain
growth-promoting balance of cytokines is
required. This can be disrupted by inflammation.
The hormone erythropoietin, produced in the kid-
neys, is required. The bone marrow precursors
need space to expand, and an infiltrating malig-
nancy can cause anemia. Hereditary or acquired
genetic defects in the hematopoietic precursors
can cause abnormal erythropoiesis, e.g.,
Diamond–Blackfan anemia and myelodysplastic
syndrome. Aplastic anemia is thought to have an
immune-related pathogenesis. Infection can
directly interfere with red cell precursors, e.g.,
parvovirus B19, or through immune stimulation
and can inhibit red cell precursors, e.g., transient
erythroblastopenia of childhood. Toxins can
inhibit hematopoiesis, e.g., chemotherapy.
11.6.6.2 Diagnosis
Hemolytic anemias are usually accompanied by
raised reticulocyte counts, indirect bilirubin,
and lactate dehydrogenase (LDH). Haptoglobins
133
will be low. Raised urinary hemosiderin sug-
gests that hemolysis is intravascular (e.g.,
thrombotic thrombocytopenic purpura, TTP,
hemolytic uremic syndrome, HUS, mechanical
heart valves) as opposed to extravascular (tak-
ing place within the spleen and reticuloendothe-
lial system). A direct antiglobulin test will help
determine whether there is antibody-mediated
red cell destruction. Table 11.5 lists the causes
of hemolysis.
The mean corpuscular volume (MCV), a
marker of red cell size, can also provide useful
information. Iron deficiency and thalassemia
cause a low MCV. A high MCV accompanies a
B12 or folate deficiency, some cases of MDS,
and any cause of reticulocytosis.
Examining the blood smear can give further
pointers toward the cause of anemia, e.g., sphero-
cytes in hereditary spherocytosis and pencil cells
in iron deficiency anemia.
An FBC will help determine if the process is
restricted to the erythrocyte lineage only, e.g.,
hereditary spherocytosis, or involves the hemato-
poiesis more generally, e.g., acute leukemia. A
bone marrow aspirate and biopsy are helpful in
select cases only.
Many cases of anemia are multifactorial, e.g.,
combined nutritional deficiencies in inflamma-
tory bowel disease coupled with anemia of
chronic disease.
Table 11.5  Causes of hemolysis (Puri 2016)
Intrinsic to red cell Extrinsic to red cell
Hb variants, e.g., HbSS, Mechanical hemolysis,
HbSC, HbS b0thal e.g., mechanical heart
valve
Globin gene deficiency, e.g., Burns, toxins, others
thalassemia
Paroxysmal nocturnal Microangiopathic
hemoglobinuria (PNH) hemolytic anemia, e.g.,
TTP, HUS
Red cell enzyme defects, e.g., Autoimmune
G6PD deficiency, PK
deficiency
Red cell membrane defects, Alloimmune, e.g., Rh
e.g., hereditary spherocytosis, hemolytic disease of
hereditary eliptocytosis, the newborn
southeast Asian ovalocytosis
Parasites, e.g., malaria,
babesiosis
134
11.6.6.3 Management
Management of anemia depends on identifying
the underlying cause and treating this. It is impor-
tant to ascertain the cause of any nutritional defi-
ciencies and address this in addition to their
replacement (e.g., source of iron loss or malab-
sorption leading to iron deficiency anemia).
Hereditary Spherocytosis (HS)
This is an autosomal dominant hemolytic anemia
caused by a deficiency of a red cell membrane
protein (α or β spectrin, ankyrin, band 3, or pro-
tein 4.2). It is the most common hereditary ane-
mia in Caucasians. There is impaired interaction
between the red cell membrane and cytoskeleton.
Removal of redundant areas of membrane takes
place in the spleen so that the red cell loses its
biconcave shape and becomes spherical, with
resultant splenomegaly. The abnormally shaped
red cells are incapable of traversing the narrow
capillary network of the spleen and hemolysis
ensues.
The spectrum of clinical severity varies among
different kindreds (Bolton-Maggs et al. 2012):
• Some patients may be transfusion dependent.
• Other patients require transfusion during
infancy, puberty, intercurrent illness, or in
pregnancy only.
• Neonates may present with prolonged neona-
tal jaundice.
• Early pigment gallstones are a feature.
• There may be a family history of splenectomy
or early cholecystectomy.
11.6.6.4 Management of HS
Children with severe disease require splenec-
tomy, which usually “cures” the anemia. Those
with moderate disease may benefit from splenec-
tomy. In the setting of asymptomatic gallstones,
controversy exists over whether those patients
undergoing splenectomy may also benefit from
cholecystectomy. Children with chronic hemoly-
sis should also receive life-long supplementation
with folic acid.
P. McCarthy and O. P. Smith
Sickle Cell Disease (SCD)
Sickle cell disease is common in people of
African, Afro-Caribbean, and Middle Eastern
heritage. It is now the most common genetic dis-
ease in the UK. Sickle hemoglobin (HbS) results
from a point mutation in the β-globin gene. HbS
polymerizes when it becomes deoxygenated, and
then it polymerizes into molecular bundles that
interfere with RBC membrane structure and
deformability. The distorted RBCs cannot tra-
verse small blood vessels in the microcirculation,
and thus vasoocclusion leading to ischemia and
local tissue damage occurs. The inheritance of
one HbS gene (“Sickle Trait” or “HbAS”) is a
benign condition. Sickling disorders are seen
when two HbS genes are inherited (“HbSS”) or
when HbS is coinherited with certain other
β-chain variants: HbC, HbE, HbD, HbOArab,
and β -thalassemia.
11.6.6.5 Diagnosis of SCD
The diagnosis is suspected by blood film exami-
nation that shows characteristic sickle cells and
features of hyposplenism (Fig.  11.8). High-­
performance liquid chromatography (HPLC)
Sickle Cells
PC
TC
MC
SC
Fig. 11.8  A 9-year-old Nigerian boy with HbSS.  The
arrowed cells are markedly elongated with two pointed
ends. Also note the other classic findings of target cells
(TC), microcytes (MC), spherocytes (SC), and polychro-
matophilic cells (PC) (Puri and Höllwarth 2009)
11  Hematological Problems in Pediatric Surgery
confirms the diagnosis, and Hb electrophoresis
or isoelectric focusing may be required for con-
firmation or if there is doubt about the nature of
the abnormal Hb. Molecular testing is also play-
ing a more important role in the diagnosis of
inherited conditions such as hemoglobinopa-
thies. Many countries including the USA and the
UK now have universal neonatal screening
programs.
11.6.6.6 Differential Diagnosis
Sickle cell disease is usually suggested by the
typical picture of chronic hemolytic anemia and
recurrent vasoocclusive crisis. The diagnosis
should be distinguished from mild variants as
discussed above (such as HbSC disease). HbSC
disease has similar but less severe and less fre-
quent symptoms than HbSS. HbS-β0 thalassemia
also has a similar clinical phenotype to HbSS dis-
ease; however, it is slightly less severe.
Gaucher disease can also expand the marrow
cavity and cause bone marrow infarction.
However, unlike sickle cell disease, which causes
splenic auto infarction by the age of five years,
Gaucher disease causes splenomegaly. Depending
on the presentation, other potential differential
diagnoses include acute anemia due to bleeding
or another cause, hemolytic anemia, septic arthri-
tis, or a pulmonary embolism.
11.6.6.7 Clinical Features
and Management of SCD
(Murad et al. 2019)
–– Vasoocclusion is often precipitated by cold or
dehydration or infection. It causes severe pain
in the affected area. Infants may experience
their first vasoocclusive event, often dactylitis,
at around 6 months of age, when HbS replaces
HbF as the predominant form of Hb.
–– Vasoocclusion most frequently causes bone
pain, but it can also cause nonspecific abdomi-
nal pain (may mimic an acute abdomen) or
chest pain. Pain control is essential, and par-
enteral opioids are often required. Patients
should be well hydrated, and antibiotics are
given if there is any suspicion of infection.
135
–– Acute chest syndrome (ACS): This can be rap-
idly progressive and life-threatening. Children
present with dyspnea, fever, cough, and/or
chest pain. Physical examination may reveal
reduced air entry, crepitations, or wheeze.
CXR may show pulmonary infiltrates.
Infection often underlies the pathogenesis.
Bone marrow embolus or intrapulmonary
vasoocclusion may also contribute. Treatment
is with supplemental oxygen, hydration, anti-
microbials, top-up or exchange transfusion,
incentive spirometry, analgesia, and support-
ive care.
–– Cerebral disease: Up to a third of children will
have silent cerebral infarcts visible on MRI
brain that lead to progressive cognitive impair-
ment (Fig. 11.9). This may present as behav-
ioral difficulties or a decline in school
performance. Ten percent of children will
have overt stroke. Raised middle cerebral
artery velocity by transcranial color Doppler
(TCD) ultrasound scanning is a predictive fac-
tor for stroke.
–– Splenic or hepatic sequestration: Sequestration
is usually seen in those under 5  years. The
spleen or liver becomes engorged by sickled
RBCs and may cause rapid hemodynamic col-
lapse, and transfusion is usually required. The
more common splenic sequestration is a major
cause of mortality in young children. Parents
are taught to palpate the child’s spleen daily,
as sequestration can occur without warning.
–– Priapism mainly occurs in adolescents and
adults with SCD and can lead to impotence.
Up to 5% of prepubertal boys may be
affected.
–– Hyposplenism: Autoinfarction of the spleen
occurs in early childhood, resulting in func-
tional hyposplenism. This puts patients at risk
of overwhelming sepsis from encapsulated
organisms, most frequently Streptococcus
pneumoniae. From early infancy, patients
should take daily prophylactic penicillin, and
all patients should be vaccinated similar to
patients post-splenectomy and receive the
annual influenza vaccine.
136 P. McCarthy and O. P. Smith

Fig. 11.9  About 25%

of patients with SCD
develop cerebrovascular
complications, and about
80% of these are under
15 years of age. The
MRI brain shows an
area of infarction (I)
secondary to vessel
occlusion (stenosis (S)
and absence (A)) (Puri
and Höllwarth 2009)

–– Aplastic crisis: Parvovirus infection sup- 11.6.6.8 Chronic Complications

presses erythropoiesis and in SCD can cause
an acute drop in Hb with an accompanying
reticulocytopenia.
–– Other complications include obstructive
sleep apnea with nocturnal hypoxia, noctur-
nal enuresis, and increased incidence of
pregnancy-­related complications.
• Pulmonary hypertension.
• Chronic renal impairment.
• Chronic osteomyelitis or avascular necrosis
(AVN).
• Pigment gallstones.
• Proliferative retinopathy.
• Chronic leg ulceration in young adults.
11  Hematological Problems in Pediatric Surgery
11.6.6.9 Disease-Modifying Therapy
• Chronic transfusion therapy: Patients are
placed on a transfusion program after a stroke
or if they have abnormal TCDs. They are
transfused every 3–4 weeks, which suppresses
HbS production. Recent evidence indicates
that chronic transfusion therapy may preserve
cognitive function in patients with silent
infarctions. Transfusion therapy is also used
for other indications. However, the benefits of
transfusion must be weighed against the
impracticalities of such programs and the
complications that include iron overload and
red cell antibody formation.
• Hydroxyurea: Originally used as an oral che-
motherapy, hydroxyurea functions in SCD to
increase the quantity of HbF synthesized, thus
reducing the relative proportion of HbS in the
circulation. Hydroxyurea can reduce the num-
ber of painful episodes suffered. It can also
reduce the incidence of dactylitis and ACS and
reduce the frequency of hospitalization and
blood transfusion. For many children, it brings
about a marked improvement in quality of life,
although not all children respond, and some
lose their response over time. It is now recom-
mended for all children above the age of
9 months with SCD.
• Hematopoietic stem cell transplantation
(HSCT): This is potentially curative, although
the procedure is more complicated for those
with a more complex SCD history, who para-
doxically are those most in need of cure.
HSCT is not feasible in every patient and is
generally considered for those with previous
stroke, severe sickle-related pain, or recurrent
chest crises with a fully histocompatible unaf-
fected sibling donor.
11.6.6.10 Surgery in Sickle Cell
Disease
• Any child of the correct ethnic origin who is to
undergo surgery and whose sickle status is not
known should have a preoperative screening
test performed, e.g., sickle solubility test and
FBC. The “SICKLEDEX,” a solubility test, is
unreliable in children under 6 months old or in
those who have been recently transfused.
137
• Common surgeries performed in SCD include
splenectomy for recurrent splenic sequestra-
tion, cholecystectomy for pigment gallstones,
and adenotonsillectomy for obstructive sleep
apnea.
• Surgery and anesthesia can precipitate a sickle
cell crisis. Anemia and obstructive sleep apnea
also contribute to surgical risk. Mortality rates
vary in the literature from 1% to 10%.
Key Principles of Perioperative Management
• Keep the child well hydrated and maintain an
adequate ambient temperature.
• Maintain adequate oxygen saturations using
supplemental oxygen.
• Consult with hematology and anesthetic col-
leagues regarding the need for preoperative
top-up or exchange transfusion. A Cochrane
systematic review of the use of preoperative
blood transfusions in SCD reviewed three ran-
domized trials (Estcourt et al. 2020). One trial
showed that a conservative transfusion regi-
men (increasing Hb to 100 g/l) was as effec-
tive in preventing perioperative complications
as an aggressive regimen (decreasing HbS to
<30%). Two further trials showed no differ-
ence in outcome between patients receiving
preoperative transfusions and those receiving
none. However, it does appear that preopera-
tive blood transfusion may prevent the devel-
opment of sickle-related lung problems, but
this was only shown in one of the two trials.
• Incentive spirometry postoperatively reduces
the frequency of ACS.
Thalassemia
Thalassemia is a hemoglobinopathy caused by
deletion of either an α- or β-globin gene.
Thalassemia is found most frequently in
Southeast Asia, but it also occurs in Northern
Africa, the Middle East, Mediterranean Europe,
India, and Central Asia. Children with the mild-
est forms show only a microcytosis without ane-
mia. At the other end of the spectrum, deletion of
all four alpha-globin genes results in hydrops
fetalis and is incompatible with extrauterine life.
Children with β-thalassemia major present in
infancy with transfusion-dependent microcytic
138
anemia and failure to thrive. In the first decade of
life, they develop the complications of chronic
transfusion-related hemosiderosis, including dia-
betes, hypoparathyroidism, and osteoporosis.
They may have delayed puberty. In order to
­prevent death in the teenage years, an aggressive
iron chelation program must accompany chronic
transfusion therapy. In addition to oral iron chela-
tors, this usually involves nocturnal subcutane-
ous desferrioxamine therapy, which is
administered five to seven nights per week for
8–15  h (UK Thalassaemia Society 2016).
Adherence to therapy often becomes an issue.
Even with maximal chelation therapy, affected
persons can develop hepatic and cardiac failure
in later life. Bone marrow transplantation is
potentially curative.
11.6.7 Neutropenia
Neutrophils differentiate in the bone marrow for
approximately 7  days and then circulate in the
blood for approximately 6.5 hr. Newborns often
have neutrophilia for the first 2  weeks of life,
with a mean count of 11 × 109/l, whereas children
between 1 month and 8 years have mean levels of
3.6 × 109/l. Above this age, counts are similar to
adult levels.
11.6.7.1 Diagnosis
–– Neutropenia in children can have inherited or
acquired causes and can be transient or
chronic. Sometimes the inherited neutropenia
is part of a more complex syndrome (see
below).
–– The workup for a child with neutropenia
requires documentation of the neutropenia
over time, elimination of possible precipitat-
ing causes by history, and bone marrow exam-
ination in those children in whom a clear
cause is not found.
–– Children with neutrophil counts of less than
0.5  ×  109/l are at increased susceptibility to
bacterial infections. An untreated bacterial
infection in a severely neutropenic patient can
progress within hours to septic shock and
P. McCarthy and O. P. Smith
death. Treatment usually involves dual intra-
venous broad-spectrum antibiotics (exact
choice based on local neutropenic sepsis
guidelines and local antimicrobial resistance
patterns). Chronic severe neutropenia renders
patients susceptible to deep-seated fungal
infection.
11.6.7.2 Differential Diagnosis
–– Acquired transient causes include infections
(viral, bacterial), burns, drugs (e.g., chemo-
therapy, carbimazole), hemodialysis, hema-
tinic deficiencies (B12, folic acid).
–– Acquired chronic causes include bone marrow
infiltration (e.g., leukemia), myelodysplasia,
immune-mediated (alloimmune and autoim-
mune), hypersplenism, viral infections, bone
marrow suppression, and idiopathic.
–– Severe congenital neutropenia (Kostmann
syndrome) can be caused by mutations in sev-
eral genes encoding mitochondrial proteins
(HAX1, AK2), endoplasmic reticulum pro-
teins (ELANE/ELA2 and G6PC3), cytoskele-
tal regulator proteins (WAS), and
transcriptional regulator proteins (GF11).
Affected children have severe neutropenia and
recurrent bacterial infections. The untreated
mortality is high (70%), and >20% of children
treated with GCSF will go on to develop acute
leukemia within 10 years.
–– Other causes of isolated neutropenia are cycli-
cal neutropenia, WHIM syndrome (warts,
hypogammaglobulinemia, infection, and
myelokathexis), and benign chronic neutrope-
nia (BCN). BCN is autoimmune in etiology
(also known as autoimmune neutropenia) and
is associated with a much milder phenotype
than SCN. It is a transient cytopenia occurring
in the preschool years that lasts an average of
20 months and is associated with minor ENT
and skin infections. It is the most common
cause of chronic neutropenia in childhood
(Smith and Hann 2003).
–– Neutropenia may also be associated with
complex syndromes including cartilage hair
hypoplasia, Chediak–Higashi syndrome, dys-
keratosis congenita, primary immunodefi-
11  Hematological Problems in Pediatric Surgery
ciency (e.g., X-linked immunodeficiency with
hyper-IgM), Fanconi anemia, Shwachman–
Diamond syndrome, cyclical dysgenesis, and
metabolic disorders such as glycogen storage
type 1B disease.
11.6.7.3 Management
Treatment of patients with severe neutropenia
usually involves supportive measures such as
antibiotics, G-CSF (a recombinant cytokine that
stimulates granulopoiesis) therapy, and in those
cases associated with bone marrow failure, allo-
geneic bone marrow transplantation.
11.6.8 Leukemia
Acute leukemia is the most common pediatric
malignancy. Eighty percent of these children
have acute lymphoblastic leukemia (ALL), 15%
have acute myeloid leukemia (AML), and the
remainder present with rare pediatric leukemias
such as chronic myeloid leukemia and juvenile
myelomonocytic leukemia.
11.6.8.1 Diagnosis
• Acute leukemia presents with a prodromal ill-
ness lasting a few days to a few weeks.
Symptoms include lethargy, anorexia, general
malaise, dislike of being handled (bone pain),
fever and infection, and bruising or bleeding.
Untreated acute leukemia is fatal within days
to weeks.
• White cell count may be high or low at presen-
tation, and there are usually blasts (leukemic
cells) circulating in the blood. Pancytopenia
(anemia, neutropenia, and thrombocytopenia)
is common. Biochemistry may show an ele-
vated LDH and urate.
• Examination may reveal hepatosplenomegaly,
palpable adenopathy, gum hypertrophy, skin
infiltration, petechiae, or purpura. Papilledema
may be present with central nervous system
(CNS) involvement.
• A preliminary diagnosis can usually be made
from a blood smear. The lineage (myeloid or
lymphoid) of the leukemia is then confirmed
by flow cytometry. In cases of pancytopenia, a
139
bone marrow aspirate or trephine biopsy may
be needed to make a diagnosis.
11.6.8.2 Management of Acute
Leukemia
• Hyperhydration to prevent tumor lysis syn-
drome (TLS) on commencement of chemo-
therapy. TLS is manifest by electrolyte
disturbances and accumulation of uric acid
crystals in the kidneys with acute renal
failure.
• Blood product support.
• Broad-spectrum antibiotics intravenously if
there are infectious issues.
• Urgent placement of a tunneled CVAD and the
commencement of systemic anticancer
chemotherapy.
• Treatment of ALL is now response stratified;
i.e., those who respond promptly to chemo-
therapy without evidence of minimal residual
disease (MRD) by molecular techniques at the
end of induction are treated on a less drug
intense regimen. Those who have MRD at the
end of induction are escalated to a more
intense regimen. This allows individualized
treatment to maximize cure while minimizing
toxicity.
• Treatment for ALL lasts for approximately
3.5 years in boys and 2.5 years in girls. There
is an initial induction phase of therapy, fol-
lowed by consolidation, delayed intensifica-
tion, and maintenance therapy. Maintenance
therapy involves daily oral chemotherapy with
intermittent pulses of intravenous chemother-
apy and is delivered as an outpatient. All
phases of treatment are generally accompa-
nied by intrathecal chemotherapy, and drugs
are delivered usually by lumbar puncture
directly into the CNS to prevent CNS relapse.
• More than 85–90% of cases of ALL can be
cured by chemotherapy alone. Relapse treat-
ment involves salvage chemotherapy with or
without a bone marrow transplant.
11.6.8.3 Surgical Issues
in the Leukemic Patient
• CVAD: The child requires the urgent place-
ment of a tunneled CVAD for ease of delivery
140
of supportive care but also to allow safe deliv-
ery of vesicant chemotherapy. If a vesicant
(e.g., daunorubicin) is administered through a
peripheral vein, thrombophlebitis and associ-
ated extravasation can occur, with extensive
localized tissue destruction. This is a limb-­
threatening complication. Prior to CVAD
placement, the child may require blood
product replacement to achieve Hb >80 g/l,
platelets >50  ×  109/l, or plasma/fibrinogen
replacement to correct a coagulopathy.
• Typhlitis or neutropenic colitis (inflammation
of the caecum due to Gram-negative bacteria
of the gut flora) is a diagnosis unique to the
neutropenic patient. Its diagnosis is relatively
common in the haemato-oncology wards
where intensive chemotherapeutic protocols
are routinely used. Patients are febrile and
usually have right-sided or generalized
abdominal pain. It should be remembered that
no clinical findings differentiate typhlitis from
other abdominal diseases. CT and ultrasound
imaging show distention and thickening of the
caecum and bowel wall thickening with asso-
ciated marked pseudopolypoid formation of
the mucosa, respectively. Neutrophil recovery
is a good prognostic factor. Conservative man-
agement with broad-spectrum antibiotics and
antifungals with or without bowel rest is the
treatment of choice. Surgical intervention
should only be considered in the most severe
cases.
• Asparaginase-associated pancreatitis (AAP):
Asparaginase is a cornerstone drug in ALL
therapy; however, in 5–10% of cases, it can
cause pancreatitis (Raja et  al. 2012). The
pathogenesis of AAP is unknown. As with
other causes of pancreatitis, patients present
with abdominal pain and vomiting and may
have deranged blood biochemistry including
an elevated amylase or lipase, low calcium,
and a raised CRP. Ultrasound or CT confirms
the diagnosis, and serial imaging may be
required to detect the emergence of complica-
tions. Management involves drug cessation,
antibiotics until sepsis can be excluded, initial
bowel rest, total parenteral nutrition, and sup-
portive care. There may be a role for octreo-
P. McCarthy and O. P. Smith
tide and continuous regional arterial infusion
of protease inhibitors. Complications include
systemic inflammatory response syndrome
and multiorgan failure, pseudocyst forma-
tion, insulin-dependent diabetes mellitus, and
chronic pancreatitis.
• Mediastinal mass at presentation: Lympho-
blastic lymphoma is a variant of ALL, which
may present with a mediastinal mass without
derangement of the FBC.  The differential
diagnosis includes a variety of other malig-
nancies, and a histological diagnosis may be
required. Sometimes mediastinoscopy can be
avoided if flow cytometry of microscopically
normal bone marrow or of pleural fluid or
biopsy of an enlarged peripheral node reveals
the diagnosis. Patients may develop superior
vena cava obstruction syndrome and airway
encroachment. Steroids may shrink the mass;
however, they may also obscure the histologi-
cal diagnosis and are therefore only used pre-­
biopsy if the patient is in a critical condition.
Early senior anesthetic involvement is impera-
tive in order to protect the airway during sur-
gery.
11.6.9 Blood Products and Their Use
in Children
A list of blood products used in children in the
surgical setting is shown below:
–– Red cell concentrates (RBCs): Dose (ml)  = 
(Desired rise in Hb in g/dl)  ×  3  ×  Recipient
weight (kg).
The crossmatching of blood is designed to
ensure that an inadvertent exposure to a foreign
antigen does not occur and consists of the
following:
1. ABO and RhD grouping of the recipient.
2. Antibody screen of the recipient (or mother in
the case of neonatal transfusion)—serum is
tested against a “panel” of commercially
available RBCs that carry all clinically impor-
tant antigens between them.
11  Hematological Problems in Pediatric Surgery
3. A comparison of these results with any avail-
able historical record (a “group and screen”
finishes at this point).
4. Testing of patient serum against the RBCs to
be transfused.
–– Platelets: Dose  =  15  ml/kg. Usual maxi-
mum dose is one pool (“adult dose”)
unless bleeding or a specific target plate-
let count.
–– Fresh frozen plasma (FFP): Usual dose
15 ml/kg, usually used as a source of clot-
ting factors in DIC, hemorrhagic disease
of the newborn. All children receive
“pathogen-­ reduced plasma,” which has
undergone a viral inactivation process.
–– Fibrinogen concentrate: This has
replaced cryoprecipitate as the product
of choice for fibrinogen replacement as it
can be rapidly reconstituted in a small
volume, is virally inactivated, and con-
tains a standardized fibrinogen content.
In a massive hemorrhage situation,
fibrinogen can be the most significantly
depleted coagulation factor; therefore,
levels must always be checked. It is
mainly used in hypofibrinogenemia (usu-
ally in DIC or major hemorrhage); the
usual dose is 70 ml/kg.
–– RBCs and platelets are leucodepleted to
remove WBCs that can cause immune reac-
tions and harbor infections (e.g., CMV).
–– CMV-negative and irradiated products are
usually required for immunosuppressed
patients—refer to local guidelines.
11.6.9.1 Acute Complications
of Blood Transfusion
–– Hemolytic reaction: fever, dyspnea, back pain,
hemoglobinuria (with intravascular
hemolysis).
–– Urticarial and anaphylactic reaction.
–– Bacterial contamination—usually seen with
platelets (stored at 22 °C).
–– Transfusion-related acute lung injury (TRALI)
occurs due to anti-WBC or HLA antibodies in
donor or recipient. This causes an ARDS-like
picture.
141
–– Febrile nonhemolytic reaction: nonspecific
reaction to a foreign antigen. These must be
differentiated from more serious reactions.
–– Volume overload: deaths have been described
in the Serious Hazards of Transfusion (SHOT)
report. All children should be medically
assessed for risk factors prior to transfusion,
blood volumes should be carefully calculated,
and patients should be continually assessed
throughout the transfusion.
The various reactions have similar presenta-
tions (Narayan and Poles 2020). In practice, all
these possibilities need to be considered. When
faced with a suspected transfusion reaction,
1. Stop the transfusion.
2. Assess hemodynamic stability—resuscitate if
necessary.
3. Check the patient identification against the
blood product.
4. Examine the product for abnormal appear-
ance suggesting contamination.
5. Order full septic screen (include product if
bacterial contamination is a possibility) and
check FBC, renal profile, and coagulation
screen (look for indices of hemolysis, renal
failure, and DIC).
6. Order a CXR if dyspnea or hypoxia.
7. Repeat crossmatch and antibody screen.
8. Alert the transfusion laboratory urgently to
the possibility of a transfusion reaction as
another product recall may be required.
Other Adverse Reactions to Blood Product
Transfusion
–– Delayed hemolytic reactions occur after
5–10 days—evidence of hemolysis (decreased
Hb, raised LDH, raised bilirubin, reduced
haptoglobins) and possibly renal impairment
due to toxic effects of free Hb.
–– Infection can be bacterial, viral, protozoal
(Chagas’ disease), prion (vCJD).
–– Iron overload is seen with chronic RBC trans-
fusion, e.g., thalassemia major.
–– Post-transfusion purpura: there is a reaction to
an antigen on transfused platelets that the
142
recipient’s immune system recognizes as for-
eign. Severe thrombocytopenia ensues
7–10 days later. This is rare.
–– Transfusion-associated graft-versus-host dis-
ease is a rare but universally fatal complica-
tion of blood transfusion. It occurs in
immunocompromised hosts or where the
donor shares HLA types with the host. It can
be prevented by irradiating blood products for
certain immunocompromised recipients and
by avoiding interfamily donations.
11.7 Conclusions
Hematology is a rapidly evolving field of medi-
cine with many advances in recent decades in the
molecular understanding of hematological disor-
ders. Pharmaceutical companies invest large pro-
portions of their drug development budgets in
haemato-oncology drugs. This has a knock-on
effect on government healthcare budgeting. Due
to the need for extensive experience of a drug in
adults before it can be considered for use in chil-
dren, there is some delay before changes in adult
practice filter into pediatrics. The following is a
summary of treatments on the horizon and enter-
ing use for the most common pediatric hemato-
logical disorders.
ALL is now curable in >85–90% of cases with
chemotherapy alone. Current clinical trial focus
is on reducing long-term toxicities of treatment
regimens for patients with low-risk disease. For
those children with high-risk disease who cannot
afford a de-escalation of therapy, biologically tar-
geted therapies offer the potential to increase
cure rates without contributing much in the way
of toxicity (Ai and Advani 2015; Pehlivan et al.
2018). Blinatumomab, a bispecific T-cell engager,
directs the immune system to target B-ALL cells
that express surface CD19. A substantial propor-
tion of adults with relapsed or refractory ALL
achieved a complete response in early-phase clin-
ical trials. This has also been seen in a number of
trials of pediatric patients and has received FDA
and EMA approval for B-ALL. Inotuzumab is a
cytotoxic agent conjugated to a humanized
P. McCarthy and O. P. Smith
monoclonal antibody directed against CD22,
often expressed on the surface of ALL blasts. It
can induce molecular remissions in relapsed or
refractory ALL.  One of the most exciting new
developments for the treatment of B-ALL is chi-
meric antigen receptor T (CART) cells. These are
T-cells taken from the patient (autologous) or
from a third party (allogeneic) that are geneti-
cally engineered in  vitro to recognize a marker
expressed on the blasts of the patient’s leukemia
(e.g., CD19 or CD22). Clinical trials showed
very high response rates (60–100%) and promis-
ing cure rates, including patients who would have
previously been incurable. Commercial CART
products are now available for pediatric patients
with relapsed or refractory B-ALL.
Drug development for hemophilia aims to
address the following problems with recombinant
factor and bypassing agents: short half-life, poor
ease of delivery, suboptimal potency, and immu-
nogenicity. Longer-acting factor concentrates are
now widely available, and agents that require only
subcutaneous delivery are now available. Gene
therapy has proven successful in small groups of
trial patients. In addition to reducing spontaneous
bleeding, new therapies for hemophilia may make
surgery safer in this population.
In spite of improved survival in SCD through
neonatal screening programs, screening to pre-
vent complications, and hydroxyurea therapy, the
current life expectancy in the USA is less than
50 years. Fifty percent of patients will not benefit
in the long term from hydroxyurea therapy, either
through poor response, reluctant therapists, inad-
equate dosing, toxicities, or noncompliance.
More widespread use of hydroxyurea, coupled
with optimization of dosing, should further
impact mortality. Novel therapies are needed,
and treatments in the pipeline focus on modify-
ing effects downstream of sickling such as vas-
cular adhesion, inflammation, and hemolysis.
Increasing experience in transplanting patients
with SCD will likely make transplant safer. SCD
patients may benefit in the future from gene ther-
apy. Likewise, more patients with thalassemia
may benefit from transplant, and gene therapy
may one day provide a curative solution.
11  Hematological Problems in Pediatric Surgery 143

In the 1950s, children with leukemia died disease. Cochrane Database Syst Rev 7. https://doi.
within weeks of diagnosis, patients with hemo- org/10.1002/14651858.CD003149.pub4
Keeling D, Tait C, Makris M (2008) Guideline on the
philia were bed-bound throughout childhood and selection and use of therapeutic products to treat
died in early adulthood, patients with thalassemia haemophilia and other hereditary bleeding dis-
major died in early childhood, and patients with orders. A United Kingdom Haemophilia Center
SCD died in infancy or childhood. Advances in Doctors' Organisation (UKHCDO) guideline
approved by the British Committee for Standards in
the understanding of these disorders mean that in Haematology. Haemophilia 14(4):671–684. https://
leukemia, hemophilia, thalassemia, and SCD, doi.org/10.1111/j.1365-­2516.2008.01695.x
long-term survival is expected, and the focus of McCann SR, Foa R, Smith O, Conneally E (2005) Case-­
treatment has moved toward reducing long-term based haematology. vol Book, Whole. Blackwell,
Oxford
complications and improving quality of life. Monagle P, Newall F (2018) Management of thrombosis
Elective surgeries are now more frequently per- in children and neonates: practical use of anticoagu-
formed in these populations, and emphasis should lants in children. Hematology Am Soc Hematol Educ
be on reducing perioperative morbidity. Program 2018(1):399–404. https://doi.org/10.1182/
asheducation-­2018.1.399
Murad MH, Liem RI, Lang ES, Akl EA, Meerpohl JJ,
DeBaun MR et  al (2019) 2019 sickle cell disease
guidelines by the American Society of Hematology:
References methodology, challenges, and innovations. Blood
Adv 3(23):3945–3950. https://doi.org/10.1182/
Ai J, Advani A (2015) Current status of antibody therapy bloodadvances.2019000931
in ALL.  Br J Haematol 168(4):471–480. https://doi. Narayan S, Poles D, on behalf of the Serious Hazards
org/10.1111/bjh.13205 of Transfusion (SHOT) Steering Group: The 2019
Arcesi R, Hann I, Smith O (2006) Paediatric haematology, Annual SHOT Report (2020). Accessed www.shotuk.
3rd edn. Blackwell Publishing, Oxford org/shot-­reports/
Astermark J, Donfield SM, DiMichele DM, Gringeri A, Neunert C, Terrell DR, Arnold DM, Buchanan G, Cines
Gilbert SA, Waters J et al (2007) A randomized com- DB, Cooper N et  al (2019) American Society of
parison of bypassing agents in hemophilia complicated Hematology 2019 guidelines for immune thrombo-
by an inhibitor: the FEIBA NovoSeven Comparative cytopenia. Blood Adv 3(23):3829–3866. https://doi.
(FENOC) Study. Blood 109(2):546–551. https://doi. org/10.1182/bloodadvances.2019000966
org/10.1182/blood-­2006-­04-­017988 Oldham KT (2005) Principles and practice of pediatric
Berntorp E, Dolan G, Hay C, Linari S, Santagostino E, surgery. vol Book, Whole. Lippincott Williams &
Tosetto A et  al (2017) European retrospective study Wilkins, Philadelphia
of real-life haemophilia treatment. Haemophilia Pehlivan KC, Duncan BB, Lee DW (2018) CAR-T cell
23(1):105–114. https://doi.org/10.1111/hae.13111 therapy for acute lymphoblastic leukemia: transform-
Bolton-Maggs PH, Chalmers EA, Collins PW, Harrison ing the treatment of relapsed and refractory disease.
P, Kitchen S, Liesner RJ et  al (2006) A review of Curr Hematol Malig Rep 13(5):396–406. https://doi.
inherited platelet disorders with guidelines for org/10.1007/s11899-­018-­0470-­x
their management on behalf of the UKHCDO.  Br J Puri P (2016) Pediatric surgery. Berlin: Springer-Verlag
Haematol 135(5):603–633. https://doi.org/10.1111/ GmbH
j.1365-­2141.2006.06343.x Puri P, Höllwarth ME (2009) Pediatric surgery: diagnosis
Bolton-Maggs PH, Langer JC, Iolascon A, Tittensor and management. vol Book, Whole. Springer, Berlin,
P, King MJ (2012) Guidelines for the diagnosis and Heidelberg
management of hereditary spherocytosis—2011 Raja RA, Schmiegelow K, Frandsen TL (2012)
update. Br J Haematol 156(1):37–49. https://doi. Asparaginase-associated pancreatitis in children. Br
org/10.1111/j.1365-­2141.2011.08921.x J Haematol 159(1):18–27. https://doi.org/10.1111/
Davies JM, Lewis MP, Wimperis J, Rafi I, Ladhani S, bjh.12016
Bolton-Maggs PH (2011) Review of guidelines for Rajagopal R, Thachil J, Monagle P (2017) Disseminated
the prevention and treatment of infection in patients intravascular coagulation in paediatrics. Arch Dis
with an absent or dysfunctional spleen: prepared Child 102(2):187–193. https://doi.org/10.1136/
on behalf of the British Committee for Standards in archdischild-­2016-­311053
Haematology by a working party of the Haemato-­ Rayment R, Chalmers E, Forsyth K, Gooding R, Kelly
Oncology task force. Br J Haematol 155(3):308–317. AM, Shapiro S et  al (2020) Guidelines on the use
https://doi.org/10.1111/j.1365-­2141.2011.08843.x of prophylactic factor replacement for children and
Estcourt LJ, Kimber C, Trivella M, Doree C, Hopewell S adults with Haemophilia A and B.  Br J Haematol
(2020) Preoperative blood transfusions for sickle cell 190(5):684–695. https://doi.org/10.1111/bjh.16704
144
Rubin LG, Schaffner W (2014) Clinical practice. Care of
the asplenic patient. N Engl J Med 371(4):349–356.
https://doi.org/10.1056/NEJMcp1314291
Smith O, Hann I (2003) Essential paediatric haematology.
Martin Dunitz Publishers, London
Thachil J, Toh CH, Levi M, Watson HG (2012) The with-
drawal of Activated Protein C from the use in patients
with severe sepsis and DIC [Amendment to the BCSH
guideline on disseminated intravascular coagula-
P. McCarthy and O. P. Smith
tion]. Br J Haematol 157(4):493–494. https://doi.
org/10.1111/j.1365-­2141.2011.09019.x
UK Thalassaemia Society (2016) Standards for the clini-
cal care of children and adults with thalassaemia in
the UK
Wynn R, Bhat R, Monagle P (2017) Pediatric hematol-
ogy: a practical guide. Vol Book, Whole. Cambridge
University Press
 Genetics
12
James J. O’Byrne and Andrew J. Green

12.1 Introduction 12.2 A Clinical Genetic

Approach to Diagnosis
Genetic disorders are common in paediatric of Malformation Syndromes
practice and contribute to between 30% and 70%
of hospital admissions for children. Some
genetic conditions cause malformations and
many others cause a range of clinical problems,
including metabolic disorders, learning dis-
ability, and neurological disease (Green &
O’Byrne 2020). Almost every aspect of paediat-
rics will involve managing children with genetic
disorders. Clinical geneticists can offer expertise
both in diagnosing rare genetic disorders, advis-
ing families with known genetic disorders, and
now with the advent of gene replacement/manip-
ulation therapies, the possibility of treatment.
This chapter will focus on an aspect of paediatric
surgery, which is the diagnosis of congenital
anomalies, and different forms of genetic dis-
ease, which can cause congenital anomalies.
J. J. O’Byrne (*)
National Centre for Inherited Metabolic Disorders,
Mater Misericordiae Hospital, Dublin, Ireland and
School of Medicine, University College Dublin,
Dublin, Ireland
e-mail: jamesobyrne@mater.ie
A. J. Green
School of Medicine and Medical Science, University
College Dublin, National Centre for Medical
Genetics Our Lady’s Children’s Hospital, Crumlin,
Dublin, Ireland
e-mail: andrew.green@olchc.ie
12.2.1 Definitions
One child in 40 (2.5%) is born with a significant
major congenital anomaly (malformation, defor-
mation, disruption, or dysplasia). Congenital
anomalies account for 20–25% of perinatal and
childhood mortality. Most affected children have
a single isolated major congenital anomaly, with-
out any underlying syndrome. However, when a
child has more than one congenital anomaly, with
or without dysmorphic features, the possibility of
an underlying syndrome should be considered.
It is also important to note that about 10% of
the normal population will have a minor congeni-
tal anomaly, such as fifth finger clinodactyly or a
single palmar crease.
In the absence of any other problems, such
minor congenital anomalies are of no major
significance.
There are many different congenital anoma-
lies, some of which are extremely rare. Examples
of the more common congenital anomalies, listed
by the organ system affected and approximate
birth incidence, are shown in Table 12.1.
Awareness of the possibility of a genetic or
syndromal association for anomalies is very
important for the diagnosis and management of
the patient and family.

© Springer Nature Switzerland AG 2023 145

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_12
146 J. J. O’Byrne and A. J. Green

Table 12.1  Examples of major congenital anomalies dysplasia caused by a mutation in the FGFR3
Examples of major congenital Birth incidence gene. Most dysplasias are single gene disorders.
anomalies (per 1000 births) A sequence can be defined as a primary mal-
Cardiovascular 10 formation that results in secondary deformations.
Ventricular septal defect 2.5 An example would be Potter’s sequence. Potter’s
Atrial septal defect 1
sequence (see Fig. 12.1) is a group of anomalies
Patent ductus arteriosus 1
Fallot’s tetralogy 1 consisting of pulmonary hypoplasia, oligohy-
Central Nervous System 10 dramnios, talipes, cleft palate, and hypertelorism.
Anencephaly 1 All of these anomalies arise as a result of the fail-
Hydrocephalus 1 ure of urine production in the foetus. The cause
Microcephaly 1 of Potter’s syndrome and failure of urine produc-
Lumbosacral spina bifida 2 tion could be posterior urethral valves, dysplastic
Gastrointestinal 4
or cystic kidneys, or renal agenesis, all of which
Cleft lip/palate 1.5
Diaphragmatic hernia 0.5
can have genetic or non-genetic origins. Pierre-­
Oesophageal atresia 0.3 Robin sequence is the grouping of cleft palate,
Imperforate anus 0.2 micrognathia/retrognathia, and glossoptosis,
Limb 2 which can have at least 20 different causes.
Transverse amputation 0.2 An association can be defined as a clustering
Urogenital 4 of anomalies that is not a sequence, which occurs
Bilateral renal agenesis 2
more frequently than by chance, but has no prior
Polycystic kidneys (infantile) 0.02
Bladder exstrophy 0.03
assumption about causation. An example is the
association of Vertebral anomalies, Anal abnor-
malities, Tracheo-Esophageal fistula, Renal
A distinction has also to be made between sev- anomalies, and Limb anomalies (VACTERL
eral different forms of abnormality, with appro- association). There is no clear cause for VATERL,
priate definitions. although it can rarely occur in people with chro-
A disruption can be defined as an anomaly, mosome 22q11 microdeletions, and can also
which is caused by interference in the structure of rarely be mimicked by Fanconi’s anaemia.
a normally developing organ. A good example A syndrome is a group of symptoms and
would be the digital constrictions and amputa- signs, and a pattern of anomalies, where there is
tions caused by amniotic bands. Amniotic bands often a known cause or an assumption about cau-
are strands of tissue which cross from one wall of sation. The looser definition of ‘syndrome’ to
the amniotic sac to the other and can constrict describe any anomaly should be avoided. The
parts of the developing foetus. term can include chromosomal disorders, such as
A deformation can be defined as an anomaly Down’s syndrome, single gene disorders, such as
that is caused by external interference in the van der Woude syndrome, which can cause cleft
structure of a normally developing organ. An lip and palate with lower lip pits or non-genetic
example would be talipes equinovarus caused by causes, such as fetal alcohol syndrome. Many of
chronic oligohydramnios, perhaps from an amni- the syndromes, described before the genomic era,
otic leak. are eponymous or named after those who
A malformation can be defined as an anomaly described them (singly or as a group), for exam-
that is caused by an intrinsic failure in the normal ple Klippel-Trenaunay-Weber syndrome and a
development of an organ. Common examples vast array of named syndromes are now described
would include congenital heart disease, cleft lip in the literature. More recently, the trend has been
and palate, and neural tube defects. to name the syndrome to reflect the molecular
A dysplasia is an abnormal organisation of defect, such as PIK3CA-related overgrowth syn-
cells in a tissue, often specific to a particular tis- drome. Features of a syndrome can evolve over
sue. For example, achondroplasia is a skeletal time, and some conditions that may be difficult to
12 Genetics
Fig. 12.1 Potter’s
sequence is a group of
anomalies or secondary
deformations
(pulmonary hypoplasia,
oligohydramnios,
talipes, cleft palate, and
hypertelorism) arising
from a primary
malformation resulting
in the failure of urine
production in a foetus
recognise in the neonatal period or early infancy
may become more apparent as the child gets
older or vice versa.
12.2.2 An Approach to Diagnosis
of a Malformation Syndrome
History  When reviewing a child with a congeni-
tal anomaly, a comprehensive medical history
and examination is an essential starting point
towards a diagnosis. Several important aspects of
the history need to be explored including a
detailed three generation family history, with ref-
erence not only to a history of the same
anomaly,but also to other anomalies as well. The
history should include documentation of consan-
guinity, ethnicity, pregnancy losses, stillbirths
and neonatal deaths, and any history of potential
teratogens in the pregnancy, considering the
likely embryological timing of the anomaly.
Teratogens can include prescribed medications
that the mother has taken during pregnancy, rec-
reational drugs, maternal diabetes mellitus, and
prolonged maternal hyperthermia.
147
Examination and Investigations  In the pres-
ence of one congenital anomaly, a very careful
examination should be carried out to check for
any other subtle abnormalities or for dysmor-
phic facial features, e.g., to check for hydro-
cephalus in an infant with a spinal
meningomyelocoele. A diagnostic approach to
congenital anomalies is outlined in Fig.  12.2.
Deformations and disruptions need to be
excluded first. If the pattern of malformations
fits into a well-described malformation
sequence, then a cause for that sequence should
be sought. If the anomalies do not fit into a
sequence, then a syndrome or association diag-
nosis should be attempted. If a malformation
syndrome diagnosis is achieved, it is important
to remember that syndromes can be caused by
chromosomal, single gene (monogenic), multi-
ple genes (polygenic) disorders, or by environ-
mental agents. If cause is unknown and there is
more than one malformation or significant dys-
morphology, chromosomal analysis should be
requested. A clinical genetic opinion should
also be sought, as a clinical geneticist can often
help greatly in achieving a diagnosis, as well as
148 J. J. O’Byrne and A. J. Green

Fig. 12.2  A diagnostic

approach to congenital
anomalies. Deformations
and disruptions should
be excluded in the first
instance and if the
pattern of malformations
fits into a well-described
malformation sequence,
then a cause for that
sequence should be
sought. If the anomalies
do not fit into a
sequence, then a
syndrome or association
diagnosis should be
considered

in counselling parents about the likelihood of Table 12.2  Causes of congenital anomalies
recurrence of similar problems in other family Relative frequency
members. Causes of congenital anomalies (%)
Genetic
Chromosomal 6
Single gene 7.5
12.3 Genetic Aetiology Multifactorial/polygenic 20–30
of Congenital Anomalies Environmental
Drugs, infections, maternal 5–10
12.3.1 Introduction illness
Unknown 50
The genetic causes of congenital abnormalities, Total 100
outlined in Table  12.2, include chromosomal,
monogenic, or polygenic/multifactorial disor- tion can have specific implications for treatment,
ders, with the latter being the most frequent prognosis, and assessment of recurrence risk and
genetic cause. Other causes are classified under counselling of families.
environmental agents and include teratogens,
maternal illness (e.g., diabetes), and infections.
It is important to note that about 50% do not 12.3.2 Chromosome Disorders
have any clear cause and most are isolated or
non-­syndromal. Nonetheless, parents and fami- Disorders of either chromosome number or chro-
lies often want an explanation as to the origin of mosome structure affect about 6/10,000 births
their child’s anomaly, and it is therefore worth- and about 6% of all congenital anomalies have a
while to pursue a diagnosis, wherever possible. chromosomal cause (see Table 12.2). There is a
Furthermore, an accurate aetiological determina- finite number of disorders of chromosome num-
12 Genetics
ber, including Down’s syndrome, Edward’s syn-
drome, and Patau’s syndrome (discussed below).
There are potentially thousands of disorders of
chromosome structure, many of which are
extremely rare or unique to a particular family.
Many chromosome disorders, such as Down’s
syndrome, are usually new genetic events in the
affected child. However, there is a small subset of
chromosome disorders that can be inherited in
the form of a balanced chromosome rearrange-
ment called a translocation, present in one or
other healthy parent. Chromosome disorders are
detected by cytogenetic analysis and the three
most commonly used analyses are chromosomal
G-banding analysis (analysing the structure and
banding pattern of the 46 chromosomes within
each cell nucleus, usually in white blood cells),
which is often referred to as karyotyping, fluores-
cent in situ hybridisation (FISH) (a detailed anal-
ysis can use fluorescently labelled DNA probes,
targeted to a particular chromosome region,
hybridised to the patient’s chromosomes), and
high-resolution array comparative genomic
hybridisation (aCGH), more commonly referred
to as a microarray.
12.3.3 Specific Chromosome
Disorders
12.3.3.1 Down’s Syndrome
Down’s syndrome is the most common chromo-
some disorder, with an average incidence esti-
mated at 1 in 700 births, and is characterised by
the presence of an extra (third) chromosome 21.
The most frequent congenital anomaly in chil-
dren with Down’s syndrome is congenital heart
disease, usually an AV canal defect. However,
children with Down’s syndrome are also more
prone to duodenal and jejunal atresias, and
Hirschsprung’s disease. Children with Down’s
syndrome will also have a varying degree of
learning difficulty, and are also prone to deafness,
hypothyroidism, and have a higher-than-average
incidence of acute and chronic leukaemias. Over
90% of Down’s syndrome arises as a new genetic
event in the child, with a separate extra chromo-
149
some 21. Under these circumstances, there is no
need to check parents’ chromosomes. There is a
rare translocation form of Down’s syndrome,
which can be readily distinguished on analysing
the child’s chromosome. For the rare transloca-
tion form, the parents can be offered chromo-
some analysis, as the translocation forms can run
in families.
12.3.3.2 Patau’s Syndrome
Patau’s syndrome is caused by the presence of an
extra chromosome 13 and is a condition that is
usually lethal in the period just after birth. It
occurs in about 1  in 5000 births. Affected chil-
dren have congenital heart disease, polydactyly,
cleft lip and palate, microcephaly, and often a
single frontal lobe in their brain (holoprosen-
cephaly). Once recognised, despite the congeni-
tal anomalies, surgery is unlikely to alter the
outcome for the infant. Like Down’s syndrome,
95% are new genetic events. There are rare trans-
location forms, which can run in families, and
again the distinction between the common and
rare forms can be identified on analysing the
baby’s chromosome.
12.3.3.3 Edward’s Syndrome
Edward’s syndrome is caused by the presence of
an extra chromosome 18 and is a condition that is
usually lethal in the period just after birth. It
occurs in about 1  in 3000 births. Affected chil-
dren are usually extremely small and have con-
genital heart disease, exomphalos, renal
anomalies, and clenched hands and rocker bot-
tom feet. Once recognised, despite the congenital
anomalies, surgery is unlikely to alter the out-
come for the infant. Over 95% are new genetic
events in the affected infant.
12.3.3.4 Other Chromosome
Disorders
Klinefelter’s syndrome is a condition character-
ised by the presence of an extra X chromosome in
a male—i.e., 47,XXY.  Boys with the condition
rarely present surgically in childhood. However,
a significant number of boys with 47,XXY can
have delayed puberty, and can also present with
150
teenage gynaecomastia. Turner’s syndrome is the
presence of a single X chromosome in a
female—45,X.  Girls with Turner’s are more
prone to congenital heart disease, classically
coarctation of the aorta, and also have renal
anomalies. They have short stature and are very
unlikely to go through puberty spontaneously
due to ovarian dysgenesis. Children with a very
small deletion of one of their two chromosomes
22, detectable by FISH, have a condition called
Di George syndrome, or velocar-diofacial syn-
drome. They can have a wide range of congeni-
tal anomalies, most commonly, congenital heart
disease, but also cleft palate, laryngeal anoma-
lies, and renal anomalies. They can also have
hypocalcaemia, learning difficulties, and, some-
times, immunodeficiency.
12.4 Single Gene Disorders
Single gene or monogenic disorders are caused
by an alteration in one or both copies of one spe-
cific gene. Such gene alterations are sufficient to
cause the disorder. There are over 4000 single
gene disorders described, and about 7.5% of all
congenital anomalies are caused by a single gene
disorder (see Table 12.2). Single gene disorders
can be divided based on their inheritance pattern,
of which there are four principal modes of inheri-
tance: autosomal dominant, autosomal recessive,
and X-linked and mitochondrial. Other rare
forms of inheritance include disorders due to
abnormalities of genetic imprinting.
12.4.1 Autosomal Dominant
Inheritance
Autosomal dominant disorders are caused by a
pathogenic gene variant in one of the two copies
of a specific gene. Families will often have sev-
eral generations affected, and usually males and
females are equally affected. Each child of a per-
son with an autosomal dominant disorder has a
50:50 chance of inheriting the gene responsible
for the condition from its parent, and thereby
being affected. The hallmark of autosomal domi-
J. J. O’Byrne and A. J. Green
nant inheritance is father to son transmission.
However, not all people with an autosomal domi-
nant disorder have a family history of the condi-
tion. For some autosomal dominant conditions,
such as Marfan’s syndrome or neurofibromatosis
1, the de novo (new) mutation rate can be
20–30%.
There can often be variability in both expres-
sion and penetrance of autosomal dominant dis-
orders. For example, neurofibromatosis 1, an
autosomal dominant condition, will almost
always manifest in someone who has an altered
neurofibromatosis 1 gene. This means that the
condition has almost complete penetrance.
However, different people can manifest the con-
dition in different ways, with some people show-
ing mild skin lesions, and others with severe
intracerebral complications. This means that the
expression or expressivity of the condition is very
variable. In contrast, only 80% of those who have
a single altered gene for the rare hereditary form
of retinoblastoma will actually develop an eye
tumour. The penetrance in this situation is 80%,
but the expression of the altered gene is consis-
tent, as manifested by a retinoblastoma.
Autosomal dominant disorders are not com-
monly seen in neonatal surgical practice. In gen-
eral paediatric surgical practice, a subset of
children with Hirschsprung’s disease, Beckwith
Wiedemann syndrome, and pyloric stenosis can
have the condition as an autosomal dominant trait
with reduced penetrance. Some forms of cranio-
synostosis and oro-facial clefting can also be
caused by an autosomal dominant disorder. There
are also rare childhood cancer predisposition
syndromes such as polyposis coli, retinoblas-
toma, and Li-Fraumeni syndrome, which have an
autosomal dominant inheritance.
12.4.2 Autosomal Recessive
Inheritance
When a child is diagnosed with an autosomal
recessive disorder, both copies of a particular
gene responsible for the condition are altered in a
significant way. Both parents are usually there-
fore carriers for that condition, with one normal
12 Genetics
and one altered gene and it is likely that many of
the child’s relatives are, unknowingly, also carri-
ers. In most cases, being a carrier for an autoso-
mal recessive condition has no effect on that
person.
When both parents are carriers for an altera-
tion in the same gene, there is a 25% or 1  in 4
chance for each of their children of being affected
by the condition. The risk of a healthy carrier sib
of having a child with the same condition depends
on the chances of that sib’s partner also being a
carrier. A child of a person with an autosomal
recessive disorder will automatically be a carrier.
That child’s chances of being affected will
depend on whether its unaffected parent is a car-
rier for an alteration in the same gene.
Autosomal recessive disorders are commonly
encountered in paediatric practice, and the nature
of the disorder depends on the population being
studied. Each regional population has its own
recessive disorder, where the frequency of carri-
ers for that disorder is the highest. For instance,
cystic fibrosis is a very common autosomal reces-
sive disorder in Western Europe, whereas sickle
cell anaemia is the most common autosomal
recessive disorder in West Africa. Common
examples of autosomal recessive conditions
include cystic fibrosis, sickle cell anaemia, sev-
eral of the mucopolysaccharidoses, beta thal-
assaemia, spinal muscular atrophy, and congenital
adrenal hyperplasia. Direct genetic diagnosis is
available for many of these conditions, and in a
number of countries, newborn screening includes
testing for a number of autosomal recessive dis-
orders including galactosaemia, homocystinuria,
and cystic fibrosis. In some countries, carrier test-
ing for some of these diseases is offered to cou-
ples planning a pregnancy, or to women in the
early stages of pregnancy.
12.4.3 X-Linked Inheritance
X-linked conditions are caused by pathogenic
variants in a gene on the X chromosome. Males
only have one X chromosome, and therefore,
males with a mutated gene manifest the disease
as they have no normal gene. Females have two
151
X chromosomes, and therefore, even if a female
has an altered gene, the normal gene usually off-
sets the effect of the mutated gene and only some
females will manifest the disorder. The daughters
of a man with an X-linked recessive condition are
all obligate carriers, as they all inherit his X chro-
mosome. The sons of a man with an X-linked
condition are all normal, as they inherit his Y
chromosome, and not his X chromosome. When
a woman has a pathogenic variant on a X-linked
gene, each of her sons has a 50:50 chance of
being affected, and each of her daughters has a
50:50 chance of inheriting the affected gene copy
and possibly being affected. There can be a rela-
tively high mutation rate for some X-linked con-
ditions, and affected boys may not have any
family history of the condition. New mutations
are responsible for muscular dystrophy in about
one-third of boys with Duchenne. The classic
examples of such conditions are haemophilia A
and B, Duchenne and Becker muscular dystro-
phy, and Hunter’s syndrome.
12.4.4 Mitochondrial Inheritance
Mitochondrial inheritance is a very unusual pat-
tern of inheritance observed in inherited diseases
caused by single gene disorders of the mitochon-
drial genome. Most of the proteins necessary for
the mitochondrial function and structure are
encoded for by the nuclear genome and so follow
the standard Mendelian inheritance patterns.
Mitochondria, however, also contain their own
small genome of 18 kilobases, with many copies
per cell as each cell contains many mitochondria.
The mitochondrial genome does not follow the
Mendelian patterns of inheritance and replicates
independently and far more frequently than the
nuclear genome. Several important mitochon-
drial proteins are encoded by the mitochondrial
genome but mitochondria are only inherited via
oocytes, and not sperm. Therefore, where a gene
alteration is in the mitochondrial genome, it will
pass exclusively down the female line, but both
males and females can be affected. The children
of an affected male will not inherit his mitochon-
drial gene alteration. Children with mitochon-
152
drial disorders usually display multisystemic
involvement and can present with varied symp-
toms at any age, including myoclonic seizures,
acute acidosis, muscle weakness, deafness or dia-
betes. A number of point mutations and deletions
in the mitochondrial genome have been described
in patients with a wide variety of conditions,
including MELAS (Mitochondrial
Encephalopathy with Lactic Acidosis and Stroke-­
like episodes) or MERRF (Myoclonic Epilepsy
with Ragged Red Fibres on muscle biopsy).
12.5 Polygenic Disorders
Many paediatric disorders do not have a clear
mode of inheritance, and can be classed as poly-
genic or oligo-genic, where a disease may arise
as a result of the effects of several genes and may
also be influenced by several environmental
factors.
A good example is cleft lip and palate, which
usually occurs in the absence of a family history.
However, monozygotic twins have a high concor-
dance for cleft palate, suggesting a genetic influ-
ence. In addition, the likelihood of having a
further child with a cleft lip or palate is increased,
when a couple have already had one child with
the condition, suggesting a genetic influence.
However, the genetic influence does not follow
the Mendelian or single pattern.
Other examples of polygenic disorders include
neural tube defects, congenital heart disease, ves-
icoureteric reflux, and coeliac disease.
12.6 Next Generation Sequencing
With the advent of massive parallel sequencing,
or next generation sequencing, it is now possible
to sequence large sections of genomes in a short
timeframe. These rapid technological advances,
along with the continued identification of patho-
genic gene variants and the improvement in gen-
otype-phenotype correlations, has led to the
development of “gene panels” for hundreds of
conditions, whole exome sequencing and whole
genome sequencing. Gene panels are now avail-
able to investigate children with specific malfor-
J. J. O’Byrne and A. J. Green
mations. Examples of conditions that gene panel
testing is now available for include arthrogrypo-
sis, anophthalmia, craniosynostosis, non-­
syndromic Hirschsprung’s disease,
holoprosencephaly, lissencephaly, overgrowth,
skeletal dysplasias, vascular malformations, and
polycystic kidney disease.
With the rapidly falling costs of next genera-
tion sequencing, targeted gene panel testing, or
whole exome/whole genome sequencing, led by
an experienced clinical geneticist, can be a low
cost, time-effective test.
One such example of a very useful gene
panel is one that is applicable to disorders of the
Noonan syndrome spectrum, which contain
conditions such as Noonans, Lentigines, Elec-
trocardiographic conduction defects, Ocular
hypertelorism, Pulmonary stenosis, Abnormalities
of the genitals; Retarded growth and Deafness
(LEOPARD), cardiofaciocutaneous, and Costello
syndromes. These are genetically heterogeneous
group of autosomal dominant disorders that often
have overlapping clinical features which can be
difficult to differentiate. The gene panel now con-
tains greater than 10 genes (BRAF, KRAS, HRAS,
NRAS, MAP 2K1, MAP 2K2, PTPN11, RAF1,
CBL, SOS1, SHOC2), which can help distinguish
which condition is present. Another example of a
gene panel with a high clinical utility is that for
Meckel Gruber syndrome, which is the most com-
mon syndromic form of neural tube defect. It pres-
ents with a classic triad of clinical features of
occipital encephalocoele, cystic kidneys and
fibrotic change of the liver but the phenotype may
also include features such as postaxial polydac-
tyly, skeletal dysplasia, microphthalmia, genital
anomalies, cleft lip and palate and heart defects
and any one of a number of genes (CC2D2A,
CEP290, MKS1, RPGRIP1L, TCTN2,
TMEM216, TMEM67) may be causative.
Gene panel tests are constructed, analysed,
and reported with an intentional blindness to all
but a specifically selected list of genes, and clini-
cians should be cognizant of which genes were
not reported when interpreting a test result. Panel
tests are also more likely to identify a gene vari-
ant of unknown significance (VUS) than a delete-
rious gene variant. In addition, not all genes
included in panel tests are unequivocally linked
12 Genetics
to the disease/phenotype and for most genes, the
penetrance is highly variable, making it challeng-
ing to translate a specific pathogenic variant into
an absolute condition risk.
12.7 Conclusions and Future
Directions
As the capacity for diagnosis of genetic condi-
tions continues to develop at a rapid pace, medi-
cal subspecialties, traditionally rarely aligned
with clinical genetics/genomics, are encounter-
ing patients with genetic diagnoses on a far more
frequent basis. This requires the clinicians work-
ing in specialities like paediatric surgery to keep
abreast of developments in the area of clinical
genetics/genomics, a challenge not easily sur-
mountable in this era of such rapid technological
advance. Array CGH was added to mainstream
clinical practice in the last decade, while tests
such as gene panels, whole exome sequencing,
and whole genome sequencing, developed due to
the advent of next generation sequencing, are fast
becoming standard in clinic practice. Although
this exciting new technology will increase the
diagnostic rate of genetic disorders, it will bring
a new set of challenges such as an increased num-
ber of VUSs and the ethical dilemma of reporting
variations in genes not associated with the pheno-
type under investigation.
Non-invasive prenatal screening is another
new genetic technology that will influence neo-
natal surgical practice. It is now widely available
153
for the investigation of trisomies 21, 18, and 13
and is being expanded to screen for many other
genetic conditions. This will increase the number
of prenatal diagnoses, which will allow for early
preparation of required surgical management in
the newborn period or even in utero.
Paediatric, and in particular, neonatal surgical
practice will continue to become more involved
with the subspecialty of clinical genetics/genom-
ics as the surgeons encounter cases and condi-
tions with a known genetic cause on a more
frequent basis. The technology to investigate the
genetics/genomics of these conditions is becom-
ing available to an increasing breadth of special-
ties including paediatric surgery. Close liaison
between surgeons and clinical geneticists will be
essential in managing these investigations and
diagnoses correctly and overcoming the associ-
ated challenges.
References
Green AJ, O’Byrne JJ (2020) Pediatric clinical genetics.
Encyclopedia of Pediatric Surgery (Vol 1)
Contact a Family—a UK charity for families with dis-
abled children, which offers information on specific
conditions and rare disorders. www.cafamily.org.uk
OMIM, Online Mendelian Inheritance in Man—a data-
base of human genes and genetic disorders developed
by staff at Johns Hopkins www.ncbi.nlm.nih.gov/
Omim/
Orphanet—a database (in several languages) of genetic
disorders, clinical information, clinic listings and
research and diagnostic genetic testing for a wide
range of disorders www.orpha.net
 Ethical Considerations in Pediatric
Surgery
Rita D. Shelby, Donna A. Caniano,
and Benedict C. Nwomeh
13.1 Introduction
Physicians are frequently confronted with situa-
tions that require consideration of medical ethics,
also known as bioethics. In general, ethics is the
philosophic discipline concerned with questions
of right and wrong. Bioethics provide structure
for the standards and principles that govern the
practice of medicine, physician behavior, and
biomedical research. While the concept of ethical
practice continually evolves with fluid and evolv-
ing cultural norms in the background of an
increasingly diverse society and rapid technolog-
ical advancement, four major ethical principles
remain. Beauchamp and Childress defined four
principles that must be considered in any ethical
dilemma as: beneficence, nonmaleficence, auton-
omy, and justice (Beauchamp and Childress
1983). Beneficence encourages physicians “to do
or promote good over harm”; nonmaleficence,
defined as “first do no harm” or a duty to not
cause intentional harm; autonomy, which respects
the right of competent persons to have control
over one’s body and medical treatment received;
R. D. Shelby
Ohio State University Wexner Medical Center,
Columbus, OH, USA
D. A. Caniano · B. C. Nwomeh (*)
Department of Surgery and Pediatrics, Ohio State
University College of Medicine, Nationwide
Children’s Hospital, Columbus, OH, USA
e-mail: benedict.nwomeh@nationwidechildrens.org
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_13
13
and justice, which involves the fair and equitable
medical treatment to all persons. In addition to
these four core principles, virtue ethics also con-
tributes to the overall application of bioethics.
Virtue ethics emphasizes the physician and
patient relationship in regard to principle-based
actions and decisions. These particular virtues
include fidelity to trust, compassion, phronesis or
common sense, fortitude, integrity, honesty, and
self-effacement (Pellegrino and Thomasma
1987).
In particular, pediatric surgeons are often con-
fronted with clinical situations that involve deci-
sions about the continuation of life-sustaining
treatments for extremely premature infants, chil-
dren with life-threatening congenital anomalies,
and children with critical and surgical illness.
Additionally, there is a power differential between
the family and the surgeon responsible for the
child’s care (Statter 2013). Parents and physi-
cians often use the best interest standard, a cen-
tral ethical theme that is based on the idea that the
typical pediatric patient cannot often make ratio-
nal treatment decisions. However, when conflicts
arise, physicians sometimes invoke the best inter-
est standards to invite the intervention of state
authorities on behalf of the child.
The goal of this chapter is to review bioethical
principles and guidelines that are pertinent to
pediatric surgery. The chapter addresses surgical
and ethical concerns in operative management
including bariatric surgery, gender reassignment
155
156
procedures, and surgery involving transgender
patients. In addition, it will highlight other areas
of ethical consideration including surgical error,
professionalism, and physician burnout and how
they interact with the care of pediatric surgical
patients.
13.2 Guidelines for Ethical
Decision-Making
and Resolution of Ethical
Problems
Ethical dilemmas often arise when decision mak-
ers and healthcare professionals disagree on what
is in the best interest of the child in question. It is
imperative for the healthcare team to work toward
a resolution that promotes respect for all parties
and all views when an ethical problem arises.
Successful outcomes require several things
including: (1) development of common language
for the discussion of moral issues; (2) provide
training on how to intelligently articulate views
about issues; (3) have agreement on a decision-­
making approach for all to follow in the course of
their deliberations.
The following guidelines provide a frame-
work for the effective resolution of difficult
situations:
1. Identify the decision makers. For most cases
in pediatric surgery, this will be the parents,
unless the patient is a mature minor.
2. Ascertain “value data” from the parents, other
relevant family members, and the patient if
able. Examples of value data include spiritual-
ity and religious beliefs, cultural norms, com-
munity values, as well as views on quality of
life.
3. Present all relevant medical information,
including the prognosis. Provide clarification
to decision makers on any areas of uncertainty
and identify additional diagnostic testing that
could add value to the decision-making
process.
4. Define all benefits, risks, and chances of
achieving the desired outcomes for all avail-
able and reasonable treatment options.
R. D. Shelby et al.
5. Provide parents and patients with a profes-
sional recommendation for the best treatment
option.
6. Seek a consensus resolution that can be
accepted by all participants.
Most ethical dilemmas can be resolved with
the use of these guidelines. Successful outcome
can be achieved when all involved parties are
interested in the best decision for the patient,
even while holding divergent views that are
seemingly irreconcilable. If resolution is not pos-
sible, consideration of outside resources such as
ethics boards should be undertaken.
13.3 Informed Consent, Assent,
and Dissent
There is a consensus in the medical community
regarding the meaning of informed consent.
Informed consent is defined as the acceptance of
health interventions by the patient following
complete understanding of all benefits and risks
of that health intervention. Additionally, the
patient is made aware and clear regarding all
available alternatives and any collateral effects
(Conti 2017). The ideals of informed consent are
embedded in contemporary medical practice and
are based on the previously described ethical
principles of autonomy, beneficence, and justice.
While all the ethical principles are important,
autonomy stands out as the guiding principle for
informed consent. It is paramount in mentally
competent patients who can exercise the freedom
to choose for themselves. Furthermore, patient
autonomy and participation are associated with
improved healthcare outcomes.
In the pediatric population, autonomy is lim-
ited as medical decisions are often made by the
patient’s legal guardian. However, in some cir-
cumstances, adolescent and more mature patients
may have authority to make their own decisions
about medical treatments. It is ethically correct to
involve mature minors in decision-making oppor-
tunities. This inclusion is often based on an
assessment of the minor’s emotional maturing,
age, experience, and intelligence (Nwomeh and
13  Ethical Considerations in Pediatric Surgery
Caniano 2011). The concept of allowing mature
minors involvement in healthcare decision-­
making is termed assent (Parekh 2007).
Physicians, including pediatric surgeons have the
responsibility to determine the ability and com-
petence of a child to give consent or assent. The
assent of the pediatric patient should be appropri-
ate to their development and understanding.
Lastly, each state has defined specific instances
children can give their informed consent and this
is outlined in the American Academy of Pediatrics
policy statement (Informed consent, parental per-
mission, and assent in pediatric practice 1995).
Just as informed consent and assent are impor-
tant, making an informed decision to refuse cer-
tain medical therapies is also paramount. This
educated refusal of healthcare is known as dis-
sent. The ability to dissent is also based on appro-
priate development, education, and understanding
of the patient. It is important to note, that dissent
can be more problematic, especially in cases of
refusal of life prolonging procedures and
treatment.
Pediatric surgeons assume significant respon-
sibility for providing guidance to parents in pro-
tecting the best interests of the child, as the
exercise of autonomy is limited. As a result,
beneficence and nonmaleficence must assume
greater relevance in the consent process.
For surgical procedures, the essential compo-
nents of informed consent include: (1) adequate
information that facilitates decision-making pro-
vided by the surgeon; (2) a competent patient or
legal proxy who demonstrates full understanding
of the intervention, including the indications,
benefits, risks, and alternatives and (3) volun-
tarily consents to the proposed intervention.
13.3.1 Exceptions to Informed
Consent
There are several legitimate exceptions to the
right of informed consent, including exceptions
for medical emergency, public health emergency,
and the incompetent patient. The exceptions for
emergencies and for patients unable to give con-
sent are particularly relevant to situations requir-
157
ing emergency surgery. Several professional
societies have established guidelines for dealing
with patients in emergency settings. When imme-
diate action must be taken to prevent death or
other serious harm to the patient, the emergency
exception mandates that appropriate care not be
delayed.
13.4 Withholding and Withdrawal
of Life-Sustaining Treatment
With rapid advancement in technology, new
diagnostic and improved life-sustaining treat-
ments have also surfaced. These include an array
of measures such as dialysis, ventilators, and
organ transplantation, as well as advances in
medications, medication delivery, and ability to
provide nutrition. These advances improve the
ability to extend life beyond critical illness.
However, because of these advances, end of life
care and withdrawal decisions can be extremely
difficult and lead to ethical dilemmas. This is
because while these technological advancements
may allow for lifesaving therapies and prolonga-
tion of survival, they do not necessarily improve
quality of life.
What defines an optimal quality of life and at
what threshold life-sustaining measures should
be stopped is an especially difficult moral, ethi-
cal, and spiritual decision. Also, when to with-
hold life-sustaining treatment must be agreed
upon by the parents and physicians. These deci-
sions are typically made in non-emergent situa-
tions. Due to historical cases where parents and
physicians disagreed on withdrawal and with-
holding of life-sustaining measures, there are
now laws that help guide physicians and families
in making these decisions. Particularly, the Baby
Doe Law defines criteria under which withhold-
ing lifesaving treatments is acceptable (Pless
1983). This is applicable if the child is deemed
irreversibly comatose or treatment is determined
to be futile and would only prolong dying. While
many may believe it is “worse” to discontinue
life-sustaining treatment than to never have initi-
ated them, there is in actuality no ethical or legal
distinction between the two.
158
In the pediatric population, when making deci-
sions about withholding care or withdrawal of
life-sustaining treatment, again the best interest
standard is utilized to allow resolution of these
dilemmas. The use of the best interest standard
allows the focus to be patient centered and allows
for the assessment of benefits versus risk of con-
tinued treatment. All discussions regarding end of
life decisions should include consideration of
severity of the condition, availability of treatment,
ability to achieve medical goals, presence of other
serious medical impairments and conditions, life
expectancy, as well as any possible and associated
complications. Decisions to continue life-sustain-
ing treatment for critically ill children should be
based upon the pediatric surgeon’s assessment of
the benefits of such treatment, its likelihood of
success, and the anticipated burdens of therapy.
When there is prognostic uncertainty about out-
comes, the parents should be accorded latitude in
decision-making. Lastly, quality of life and ability
for families to be able to provide adequate
resources and care in the event of severe handicap
is also a critical component of these discussions.
As recommended by the American Academy of
Pediatrics, all discussions and medical decisions
should be family and patient centered in approach
(Informed consent, parental permission, and
assent in pediatric practice 1995).
Finally, there exists a special population of the
mature child or adolescent patient, who has the
capability for autonomy. Life-threatening ill-
nesses may impact the developmental under-
standing of pediatric patients and accelerate their
comprehension of serious illness, as well as their
own mortality. Furthermore, this experience may
promote their wishes to control decisions about
their healthcare. Mature or older children who
prove to be capable of giving assent should be
allowed to actively participate in healthcare dis-
cussions, including end of life issues.
13.5 Multiculturalism
Over the decades, there continues to be an expo-
nential growth in the cultural diversity of our
modern society. Culture has been defined as the
common and accepted way of thinking, feeling,
R. D. Shelby et al.
and acting for a group of people. Culture includes
all accepted beliefs, value patterns, attitudes, and
behaviors that are held in common by a group of
people. Culture can be viewed as the leading
mechanism by which people navigate and make
sense of their surrounding world through shared
meanings and patterns of behavior. However,
with the growth of cultural diversity and multi-
culturalism, there is an increasing body of evi-
dence that demonstrates unequitable delivery of
healthcare to cultural and ethnic minority groups
(Dossey 2015; Trubek and Das 2003).
Multiculturalism can lead to difficulties in deliv-
ering care to patients. This is due to a lack of
understanding of all the nuances and practices of
the diverse cultures. Cultural practices that
diverge from the “norms” of the majority can
prove to be frustrating, particularly when they
affect acceptance of and compliance with recom-
mended medical treatments. It is important for
pediatric surgeons to recognize the depth of
importance that culture plays in healthcare utili-
zation. Comprehension of this will elucidate both
how and why patients and families may behave in
certain ways.
Physicians tend to be socialized by traditional
Western medical training, which is often based
on a set of assumptions and values that focuses
on disease being the result of disruption or mal-
function of normal biologic processes. In con-
trast, some patients experience illness as an
extension of their cultural or spiritual wellness.
As a result, it is imperative that the pediatric sur-
geon be mindful of their own individual cultural
biases as well as the possibility of differences in
beliefs by patients. The treating pediatric surgeon
has responsibility to inquire about patient beliefs,
goals of treatment, and any other patient concerns.
Lastly, language barriers can often be associ-
ated with multiculturalism. In settings where
English is the dominant language, limited profi-
ciency in English may pose additional difficulties
in the multicultural medical setting. The pediatric
surgeon should utilize experienced medical trans-
lators to ensure accuracy of transmitted informa-
tion. The use of professional translations services
can significantly reduce healthcare disparities
associated solely with language barriers (Paredes
et al. 2018).
13  Ethical Considerations in Pediatric Surgery
13.6 Surgical Error
One of the tenets of ethical medical practice is
primum non nocere (“first do no harm,”). While
medical error cannot be completely eliminated,
as “to err is human,” (Kohn 2001) a worthy goal
is to minimize them. For several decades now, the
incidence of medical errors and their significant
impact on patient outcomes has been highlighted
in all aspects of society. Surgical errors are trau-
matic to the patient and family, and also to the
surgeons (Wu and Steckelberg 2012; Anderson
and Abrahamson 2017; Youngson 2014). While it
is extremely difficult, pediatric surgeons have an
ethical, legal, and moral responsibility to fully
disclose any possible error to families and
patients. Disclosure of errors is endorsed by all
professional organizations and increasingly
required by regulatory and governmental agen-
cies. Furthermore, open and honest disclosure
has been shown to reduce malpractice claims, as
well as improve the level of trust patients have in
physicians who are caring for them (Wei 2007).
Transparency in the disclosure of medical
errors is of utmost importance. The surgeon
should explain in clear, simple language what
happened, the anticipated effects, and how the
patient will be managed. Humility and accep-
tance of responsibility for all events is also key.
Some physicians may feel reluctant to admit
error, partly because risk managers routinely
advise that error disclosure not include any state-
ment that could constitute an admission of liabil-
ity. However, there is some evidence that a clear
apology and an explanation of what steps will be
instituted to prevent a similar error in the future
can help achieve resolution (Wei 2007).
Considerable skill and tact, as well as practice, is
required in the effective disclosure of errors.
13.7 Research and Innovation
in Pediatric Surgery
In order to provide the most up-to-date and best
evidence-based treatment for children, the par-
ticipation of the pediatric surgeon in research and
innovations is critical. While it is vital that
159
research endeavors be encouraged, understand-
ing research practices, as well as ethical chal-
lenges involved in conducting research and
protection of research subjects is paramount
(Rentea et al. 2020). Investigators are required to
protect not only the safety, but also, the privacy of
the research subject. Compliance with these
requirements is promoted and defined by the
Institutional Review Board (IRB). Pediatric
patients are defined as vulnerable subjects
because they do not have the ability to consent to
research themselves. Nevertheless, ethical tenets
of research require that research interventions
and practices not involve or cause greater than
minimal risk. When there is the possibility of
therapeutic benefit, then higher risk maybe ethi-
cally allowed. All the previously discussed bio-
ethical guidelines apply here as well. Additionally,
all the principles of informed consent, assent, and
dissent also apply here. Importantly, participation
of a child in research should be a shared decision-­
making process, there must be buy-in from the
family and the patient when maturity allows.
Surgical innovation, in contrast to research,
entails application of new techniques or thera-
pies, or modification of existing procedures,
which are thought to lead to the improvement in
care. Pediatric surgeons have been among the
most notable surgical innovators, with many of
the procedures beneficial to children today being
the product of their innovation. There is less reg-
ulation in surgical innovation, compared to inno-
vation of drugs or other medical therapies. This is
important to point out as it creates the potential
for abuse and can be harmful and dangerous. As
in other areas of research, innovation should be
directed toward the improvement of therapeutics
and the improvement in the delivery of care.
13.8 Ethical Issues in Bariatric
Surgery of the Pediatric
Patient
Over the last several decades, there has been
improved success in bariatric surgery and associ-
ated outcomes in the adult population. In previ-
ous times, bariatric surgery was not an option for
160
the management of obesity in children. However,
it has increasingly become apparent that obese
children are not safe from adverse comorbidities
associated with obesity such as type 2 diabetes,
hypertension, sleep apnea, and other serious
problems. Multiple research studies have demon-
strated that bariatric surgery is a safe and effec-
tive method of not only controlling weight, but
also eliminating some of these comorbidities.
With this evidence, the use of bariatric surgery in
children has become more in vogue (Pratt et al.
2018; Bolling et al. 2019; Armstrong et al. 2019).
However, the ethical issues of bariatric sur-
gery in children are complex. As with any sur-
gery, bariatric surgery has its own set of surgical
risk and complications that can lead to a lifetime
of problems. There is a concern that patients and
families will not be able to necessarily under-
stand the amount of risk they are undertaking.
Furthermore, undergoing bariatric surgery is a
significant commitment from the patient. It
requires not only a consent to undergo surgery,
but also a commitment to a lifestyle change.
These are all things that a younger child may not
have the ability to understand, and which their
families may not recognize. Informed consent
can be difficult in these situations as patients may
not be able to comprehend that irreversible nature
of some of the most effective procedures.
Furthermore, there may be a lack of understand-
ing that surgery is not a “quick fix” and there will
be further work required by the entire family unit.
These patients may have difficulty balancing
what they envision as immediate benefits of
weight loss, against unforeseen complications
such as reoperation and other unanticipated prob-
lems (Bolling et al. 2019; Caniano 2009).
Furthermore, there are known racial and eco-
nomic disparities in the manner through which
bariatric programs provide access and support to
a diverse population. Yet, obesity is dispropor-
tionally higher in African American, Hispanic,
and low-income children (Blacksher 2008).
Therefore, the ethical issue of justice demands
attention to fairness and equity in the distribution
of bariatric surgical resources, including presur-
gical education, counseling, and post-operative
follow up.
R. D. Shelby et al.
While the ethical issues are complex, the deci-
sion to proceed with bariatric surgery should only
be made after thorough counseling of the family
unit, and after it has been thoroughly determined
that less invasive methods would not be sufficient
to manage the obesity and comorbidities, and that
overall benefit of surgery outweighs the risk.
13.9 Ethical Issues
in the Treatment of Gender
Dysphoria in Children
and Adolescents
Pediatric surgeons may encounter children and
adolescent patients with gender dysphoria,
defined by the World Professional Association
for Transgender Health (WPATH) as discomfort
or distress caused by a discrepancy between a
person’s gender identity and that person’s sex
assigned at birth, including the associated gender
role and/or primary and secondary sex character-
istics (Kimberly et  al. 2018). Children may
exhibit signs of gender dysphoria as young as
2–4  years of age, with these characteristics:
strong desire to be of the other gender or insis-
tence that they are the other gender; preferences
for wearing clothes typical of the other gender,
for being the other gender during play, and for
toys, games, and activities usually engaged in by
the other gender, for playmates of the other gen-
der; rejection of toys, games, and activities typi-
cal of their assigned gender; strong dislike of
their sexual anatomy; and strong desire for the
physical characteristics that match their experi-
enced gender. While the pediatric surgeon does
not play an integral role in providing surgical ser-
vices to these patients, they do play a role in sup-
port and coordination of treatment for these
patients and their families.
The primary ethical obligation of pediatric
surgeons is to provide reassurance and emo-
tional support to the patient and parents about
gender dysphoria. Patients with gender dyspho-
ria are known to have high incidence of psycho-
logical distress (Kimberly et al. 2018; Murphy
2019). Support by the pediatric surgeon is to
approach care in a non-judgmental way, explain-
13  Ethical Considerations in Pediatric Surgery
ing to the patient and family that there is noth-
ing “wrong” with the child or abnormal, and
lastly coordination of care. These situations can
lead to significant emotional distress for both
the child and the family, and it is important that
the pediatric surgeon provides support by refer-
ral to appropriate counselors or psychological
health specialists.
Special importance is paid to patients choos-
ing to undergo gender or sex reassignment. It is
important for the pediatric surgeon caring for
these patients to understand what therapeutic
interventions are available and acceptable for
their patients. Children do not have as many
opportunities in obtaining hormonal and surgical
interventions to change their bodies. The World
Professional Association for Transgender Health
(WPATH), as well as the American Academy of
Pediatrics (AAP) recommends that surgical body
modification not be performed until patients are
of legal age and they have lived as their chosen
gender for at least 12 consecutive months
(Murphy 2019; Rafferty 2018). While surgical
interventions may not be available to those
younger than 18, except in special cases, hor-
monal therapy to suppress or induce certain fea-
tures is possible. Hormonal therapy can be
reversible or partially reversible with little to no
residual effect, hence it is thought to be a more
acceptable solution for younger patients. Once
the patient and family have reached a point and
agreed upon more permanent gender affirmation
interventions, the responsibility of the pediatric
surgeon then becomes a referral source. Surgical
management of these patients requires a multi-
disciplinary team, which includes disciplines like
Psychiatry, Endocrinology, Urology, and Plastic
surgery. Furthermore, pediatric surgeons’ most
important role maybe knowledge about and refer-
ral to a center for transgender and care coordina-
tion of that team.
Pediatric providers, including surgeons, have
an essential role in assessing gender concerns
and providing evidence-based information to
assist patients and families in their decision-­
making. Failure to do so can prolong gender dys-
phoria and can be detrimental to the patient.
Lastly, the pediatric surgeon must be aware of
161
their own bias that can interfere with the care of
these patients and infringe on their autonomy to
make personal choices.
13.10 Physician Wellness
The ethical duty of the physician to maintain
wellness and minimize burnout in order to pro-
vide the best care for the patient has risen to the
forefront of professional and public discussion.
Burnout is described as emotional exhaustion
resulting in cynicism and reduced work effective-
ness. It is also linked to poor physician health,
depression, and a rise in suicide rates amongst
physicians (Arnhart et al. 2019; Hu et al. 2019;
Bilimoria 2020). Other consequences are
impaired judgement, increase in medical errors,
decreased work performance, and decreased
emotional functioning (i.e., lack empathy toward
patients). Burnout appears to be higher among
surgeons. This is felt to be a result of more fre-
quent instances of workplace mistreatment, but
also long work hours, and constant interaction in
stressful situations. Self-care and management of
our own personal wellness is an ethical obliga-
tion. It addresses the ethical principle of nonma-
leficence, which implores us to do no harm. If we
are in a state of burnout, it is impossible to be the
effective and functional healthcare providers that
our patients deserve. Recognition of this has led
to the development of evidence-based practices
that help physicians not only identify burnout,
but also provide resources to start treating and
manage it. It is imperative that physicians remain
vigilant about the state of their own health, so
that they can remain the best providers they can
be to our patients.
13.11 Conclusion
There are many more moral dilemmas and ethical
concerns that we encounter in pediatric surgery
practice than what is covered in this chapter.
Regardless of the specific situation, the principles
of beneficence, nonmaleficence, autonomy, and
justice, and such virtues as trust, compassion,
162
prudence, justice, courage, phronesis, fortitude,
integrity, honesty, and self-effacement provide a
basis to navigate these situations. While there is
no universal solution for problems encountered
in practice, the topics covered here will serve as
an excellent guide to reach acceptable solutions.
References
Anderson JG, Abrahamson K (2017) Your health care may
kill you: medical errors. Stud Health Technol Inform
234:13–17
Armstrong SC, Bolling CF, Michalsky MP, Reichard KW,
section on obesity sectos (2019) Pediatric metabolic
and bariatric surgery: evidence, barriers, and best
practices. Pediatrics 144(6). https://doi.org/10.1542/
peds.2019-­3223
Arnhart K, Privitera MR, Fish E et  al (2019) Physician
burnout and barriers to care on professional applica-
tions. J Leg Med 39(3):235–246. https://doi.org/10.10
80/01947648.2019.1629364
Beauchamp T, Childress J (1983) Principles of biomedi-
cal ethics
Bilimoria KY (2020) Association for academic surgery
presidential address-fanning the burnout fire: how our
misconceptions and good intentions could fail tomor-
row’s surgeons. J Surg Res. https://doi.org/10.1016/j.
jss.2020.06.005
Blacksher E (2008) Children’s health inequalities: ethi-
cal and political challenges to seeking social justice.
Hast Cent Rep 38(4):28–35. https://doi.org/10.1353/
hcr.0.0035
Bolling CF, Armstrong SC, Reichard KW, Michalsky
MP, section on obesity sectos (2019) Metabolic and
bariatric surgery for pediatric patients with severe
obesity. Pediatrics 144(6). https://doi.org/10.1542/
peds.2019-­3224
Caniano DA (2009) Ethical issues in pediatric bariatric
surgery. Semin Pediatr Surg 18(3):186–192. https://
doi.org/10.1053/j.sempedsurg.2009.04.009
Conti AA (2017) From informed consent to informed dis-
sent in health care: historical evolution in the twenti-
eth century. Acta Biomed 88(2):201–203. https://doi.
org/10.23750/abm.v88i2.5441
Dossey L (2015) Medical racism. Explore (NY) 11(3):165–
174. https://doi.org/10.1016/j.explore.2015.02.009
Hu YY, Ellis RJ, Hewitt DB et al (2019) Discrimination,
abuse, harassment, and burnout in surgical residency
training. N Engl J Med 381(18):1741–1752. https://
doi.org/10.1056/NEJMsa1903759
Informed consent, parental permission, and assent in
pediatric practice (1995) Committee on Bioethics,
American Academy of Pediatrics. Pediatrics
95(2):314–317
R. D. Shelby et al.
Kimberly LL, Folkers KM, Friesen P et al (2018) Ethical
issues in gender-affirming care for youth. Pediatrics
142(6). https://doi.org/10.1542/peds.2018-­1537
Kohn LT (2001) The Institute of Medicine report on medi-
cal error: overview and implications for pharmacy.
Am J Health Syst Pharm 58(1):63–66. https://doi.
org/10.1093/ajhp/58.1.63
Murphy TF (2019) Adolescents and body modification for
gender identity expression. Med Law Rev 27(4):623–
639. https://doi.org/10.1093/medlaw/fwz006
Nwomeh BC, Caniano DA (2011) Emerging ethical issues
in pediatric surgery. Pediatr Surg Int 27(6):555–562.
https://doi.org/10.1007/s00383-­011-­2907-­3
Paredes AZ, Idrees JJ, Beal EW et  al (2018) Influence
of English proficiency on patient-provider com-
munication and shared decision-making. Surgery
163(6):1220–1225. https://doi.org/10.1016/j.surg.
2018.01.012
Parekh SA (2007) Child consent and the law: an insight
and discussion into the law relating to consent and
competence. Child Care Health Dev 33(1):78–82.
https://doi.org/10.1111/j.1365-­2214.2006.00641.x
Pellegrino ED, Thomasma DC (1987) The conflict
between autonomy and beneficence in medical eth-
ics: proposal for a resolution. J Contemp Health Law
Policy 3:23–46
Pless JE (1983) The story of Baby Doe. N Engl
J Med 309(11):664. https://doi.org/10.1056/
NEJM198309153091111
Pratt JSA, Browne A, Browne NT et  al (2018) ASMBS
pediatric metabolic and bariatric surgery guidelines,
2018. Surg Obes Relat Dis 14(7):882–901. https://doi.
org/10.1016/j.soard.2018.03.019
Rafferty J (2018) Ensuring comprehensive care and sup-
port for transgender and gender-diverse children and
adolescents. Pediatrics 142(4):e20182162. https://doi.
org/10.1542/peds.2018-­2162
Rentea RM, Oyetunji TA, Peter SDS (2020) Ethics
of randomized trials in pediatric surgery. Pediatr
Surg Int 36(8):865–867. https://doi.org/10.1007/
s00383-­020-­04665-­5
Statter MB (2013) The ethics consultation and the pedi-
atric surgeon. Semin Pediatr Surg 22(3):149–153.
https://doi.org/10.1053/j.sempedsurg.2013.05.003
Trubek LG, Das M (2003) Achieving equality: health-
care governance in transition. Am J Law Med
29(2–3):395–421
Wei M (2007) Doctors, apologies, and the law: an
analysis and critique of apology laws. J Health Law
40(1):107–159
Wu AW, Steckelberg RC (2012) Medical error, incident
investigation and the second victim: doing better but
feeling worse? BMJ Qual Saf 21(4):267–270. https://
doi.org/10.1136/bmjqs-­2011-­000605
Youngson GG (2014) Medical error and disclosure  – a
view from the U.K. Surgeon 12(2):68–72. https://doi.
org/10.1016/j.surge.2013.10.011
 Minimal Access Surgery in Infants
and Children
14
Amulya K. Saxena, Roberta V. Iacona,
and Keith Georgeson

“Surgery is always invasive; it is only the access Table 14.1  Minimal access surgery in pediatric surgery
that is minimal in the endoscopic approach” (organs and pathologies)
Upper GI tract:
Gastroesophageal reflux—Fundoplication (Nissen,
14.1 Introduction Toupet, Thal)
Heller’s myotomy procedure
Esophageal/gastric duplication
The application of Minimal Access Surgery Splenectomy
(MAS) in pediatric surgery has gained a progres- Liver and biliary tract pathologies, choledochal cyst
sively increased interest and application over the Pancreatic pathologies
Cholecystolithiasis
last 20  years (Saxena and Hollwarth 2009). Nutrition and feeding issues—Gastrostomy and
Thanks to the development of smaller size instru- Jejunostomy
ments as well as the improvement of specialized Lower GI tract:
pediatric anesthetic techniques, MAS has almost Hirschsprung’s disease
become the gold-standard approach for the vast Anorectal malformations
Appendicectomy
majority of pediatric surgical conditions includ- Meckel’s diverticulum
ing the neonatal ones (Table 14.1). Intussusception
Inflammatory bowel disease: Chron’s disease and
ulcerative colitis
Pilonidal sinus—Endoscopic treatment—EPSiT*
(* external)
Thoracic procedures:
Congenital pulmonary airway malformation
Recurrent pneumothorax
Hyperhidrosis
Tracheomalacia—Aortopexy
Bronchogenic cyst
Diaphragmatic eventration
A. K. Saxena (*) · R. V. Iacona Empyema
Department of Paediatric Surgery, Chelsea and Pulmonary sequestration
Westminster Hospital NHS Foundation Trust, Pediatric Urology and Gynecology:
Imperial College London, London, UK Pyeloureteric junction pathologies
e-mail: amulya.saxena@nhs.net Cystoureteric pathologies
Ovarian cysts
K. Georgeson
Renal pathologies
Department of Paediatric Surgery, University of
Undescended testes
Alabama Health Services, Birmingham, AL, USA
e-mail: Keith.georgeson@ccc.uab.edu (continued)

© Springer Nature Switzerland AG 2023 163

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_14
164
Table 14.1 (continued)
Pediatric Oncology:
Mediastinal masses
Lung lesions
Thymic pathologies
Abdominal masses
Retroperitoneal masses (adrenal, renal)
Ovarian masses
Neonatal Surgery:
Esophageal atresia
Chylothorax
Congenital diaphragmatic hernia (CDH)
Inguinal hernia repair
Pyloric Stenosis
Duodenal atresia
Malrotation—Ladd’s procedure
14.1.1 Laparoscopic
and Thoracoscopic Surgery:
Technical Considerations
The introduction of laparoscopic instruments for
either thoracic, abdominal, or pelvic cavity in
pediatric surgery should be performed under
direct vision (Hasson’s technique). Although
popular with adult surgeons, the use of the Veress
needle is discouraged among the pediatric popu-
lation, mainly due to a high risk of life-­threatening
complications, especially due to the shorter dis-
tance between the abdominal wall and the under-
lying structures and major vessels. Once the
abdominal/thoracic access is established safely
under direct vision, the ancillary working port(s)
are introduced according to the demands of the
specific surgical procedure.
Ports used for minimal access procedures can
be disposable or re-usable; however, most of the
3  mm options are available in re-usable forms.
Whereas, the disposable instruments are consid-
ered to be more reliable since they have not been
subjected to previous depreciation, their main
disadvantage is related to cost. Of note, some
instruments will only be available in disposable
options specifically based on the type of function
performed. Ports currently available are various
sizes that range from 2 mm (fragile and with lim-
ited application) to 3–5–11–12–15 mm. The lat-
ter larger sizes are often necessary even in
pediatrics because some instruments, such as sta-
plers and retrieval bag, are not commonly avail-
A. K. Saxena et al.
Table 14.2  Trocar tip options
Sharp pyramidal: traumatic. Easy leakage of gas
Sharp conical: less traumatic, dilate the tissue
Eccentric
Blunt conical
Devices with a small blade of a knife
Versa Step
Multiport trocar—SILS
able in smaller sizes. Besides the diameter, the
choice of the length of the port is also important,
the choice of which largely depends on the age
and weight of the child. The current available
lengths are 60–75–100 and 110 mm.
Ports are presently available with a variety of
trocars that aid in the insertion and secure place-
ment of the ports (Table 14.2). Fixation of ports
at the site of insertion is another issue that has
been addressed by the industry in designing vari-
ations in the sleeve of the ports to enable secure
placement. However, straight sleeve ports will
require fixation techniques to the abdominal wall,
mostly during neonatal procedures where the
thickness of the abdominal wall is less compared
to a school going child or teenager and chances
of displacement are much higher due to their
minimal lengths within the abdominal cavity.
MAS requires the creation of a working space.
This is obtained by the use of carbon dioxide
(CO2), which has the advantage of being reab-
sorbed by blood, not being toxic, cost-effective,
and is also non-combustible in the presence of
electrosurgical devices. Insufflation pressures
used in the pediatric age group ranges between
8 and 12  mmHg and these are even lower in
neonates and infants. When thoracoscopy is
performed, a pneumothorax can be created with
3–5  mmHg CO2. The flow rate is preferably
kept less than 1 L/min at least at the beginning
of the procedure after which it can be safely
increased if required up to 2  L/min. A higher
flow rate may lead to scapular pain and also
cause hypothermia.
A variety of scopes are available and can be
chosen depending on the requirements of the
type of procedure. Telescopes are available in
2–3–5–10–12 mm with angles ranging from 0° to
70°. The 30° scope is used most because it allows
14  Minimal Access Surgery in Infants and Children 165

to look behind structures, around corners, and 14.2.1 Diagnostic Thoracoscopy

below the surface of the abdominal wall. Newer
scopes are available which allow three-­ Thoracoscopy for diagnostic purposes is used
dimensional visualization of the internal infrequently due to the advances in diagnostic
structures. imaging. In carefully selected patients, how-
In terms of work port placement, this will fol- ever, the visualization of obscure structures or
low either the rule of sectorization (when the lesions in the thoracic cavity can be achieved
­target organ is on one side so that the optical port using thoracoscopy. Additionally, biopsies of
comes to lie on one side and the working ports on intrathoracic lesions are commonly performed
the other side of the target organ), or the triangu- thoracoscopically.
lation (when the optical port is placed about
10  cm from the target organ with two further
working ports on the same 10  cm arc on either 14.2.2 Biopsies for Pulmonary
side of the optical port, allowing a working space Pathologies
at a 60°–90° angle).
One of the most common indications for thora-
coscopy in children is to perform lung biopsies.
14.2 Thoracoscopic Procedures In fact, interstitial lung disease in immunocom-
promised patients, parenchymal lung disease,
For thoracoscopy to be applied in children, pulmonary masses associated with malignancies
appropriate anesthetic techniques play a central and refractory pleural lesion disease will benefit
role. General anesthesia and positive pressure from lung sampling. Patient positioning will
ventilation can impair visualization inside the depend on the position of the lung mass.
pleural space during thoracoscopy due to lung Generally, lateral positioning is the most com-
expansion. Single lung ventilation by intubation mon; however, lung biopsies might also require
of the contralateral mainstem bronchus is the prone positioning (posterior pleural lesions) or
most common technique used by pediatric anes- supine (anterior lesions). The camera port is usu-
thesiologists to partially collapse the ipsilateral ally positioned within the midaxillary line at the
lung for pediatric thoracoscopy. Double lumen level of the fourth or fifth intercostal space, below
tubes are not available in appropriate sizes for the tip of the scapula. Further ports are inserted
neonates and small children. Bronchial blockers, according to the baseball field positioning crite-
such as a Fogarty catheter passed through or ria. Thoracoscopic biopsies are performed by
beside an endotracheal tube, can be used to block using endoloops, endoscopic stapling or sealing
ventilation in the ipsilateral lung. Low flow and devices. There are a few relative contraindica-
low pressure (<6 cm H2O) infusion of CO2 into tions such as previous pleurodesis, advanced
the ipsilateral pleural space can help to partially empyema, pleural thickening of unknown etiol-
compress the lung and to increase the visualiza- ogy, cardiovascular instability, hypoxemia
tion of intrathoracic structures. This technique is despite oxygen therapy, bleeding diathesis, pul-
well tolerated in most neonates. Potential com- monary arterial hypertension, refractory cough,
plications of the technique, such as CO2 embo- drug hypersensitivity, and reduced general health
lism and hypotension due to impaired venous status with short suspected survival. It is impor-
return, have not been reported in significant num- tant to remember that in suspected malignancy,
bers. Stopping insufflation and relieving the the use of endobag is mandatory to retrieve the
pneumothorax can readily reverse hypercarbia specimen, in order to avoid seeding of tumor at
and hypoxia due to intrapleural infusion of CO2. the port site.
166 A. K. Saxena et al.

14.2.3 Pleural Empyema
The treatment of empyema using thoracoscopic
debridement has become a standard of care.
Children with a diagnosis of empyema should be
examined by either ultrasound or CT scanning.
Children with obvious fibrinous septi and locula-
tion of the empyema should be treated by thora-
coscopic debridement of the intrapleural space.
This technique is started using a chest tube to
access the pleural space of an anesthetized child.
If the chest tube does not sufficiently drain the
purulent material in the chest, the next step
should be thoracoscopic access of the pleural
pace followed by thoracoscopic debridement of
the fibrinous septi within the pleural space. Once
the pleural space has been accessed with a port, a
suction device is used as a probe to separate the
lung from the parietal pleura. CO2 is infused to
further develop this space. A scope is then placed
through the original port and other ports are
sequentially introduced under vision. The fibrin-
ous material within the pleural space is removed
using grasping forceps. Usually, most of the
fibrinous debris can be removed within 1  hour.
The pleural space is irrigated with saline and
aspirated using a suction device. One or two
chest tubes are left in place to drain the pleural
space. Using this technique, recovery from the
empyema is usually much more rapid than fol-
lowing chest tube drainage alone or chest tube
drainage with fibrinolytic agents to break up the
fibrinous septi. The chest tubes are usually left in
place for 3 or 4  days and then removed when
drainage becomes minimal.
tive recovery time. Generally, 3–5 mm ports are
used and the application of 3–5  mm sealing
devices is gaining popularity for these proce-
dures. In older children, where it is possible to
insert 5–12 mm ports, endostapler application is
possible. Usually, the lungs are freed using seal-
ing energy sources followed by stapling or liga-
tion of the bronchus and major vessels. Depending
on the size, the specimens are generally removed
through in an endobag or in a piecemeal fashion
through a slightly enlarged 5 mm port site. The
chest drain is usually left following lobectomies.
In pulmonary sequestrations, the first step of
the procedure is to identify the systemic artery
feeding the sequestration, which usually origi-
nated below the level of the diaphragm. Depend-
ing on the size of the vessels, sutures, clips, or
sealing device are used for ligation.
14.2.5 Primary Spontaneous
Pneumothorax
Most primary spontaneous pneumothorax (PSP)
patients show apical emphysematous-like
changes such as bullae or blebs which can be
demonstrated by high-resolution chest com-
puted tomography (CT) scans. Indications for
the thoracoscopic approach to primary sponta-
neous pneumothorax are simultaneous bilateral
PSP, recurrent PSP, failed conservative manage-
ment with persistent air leak >3–5 days or sig-
nificant hemopneumothorax. The procedure
involves atypical lung resections using endo-
scopic staplers.

14.2.4 Congenital Pulmonary Airway 14.2.6 Mediastinal Masses

Malformations
and Pulmonary Sequestration Mediastinal masses detected in pediatric age will
generally be related to either congenital anoma-
Congenital Pulmonary Airway Malformations lies or neoplasms. The most frequent indication
(CPAM) including bronchogenic cyst, broncho-­ for thoracoscopy is the presence of a mediastinal
pulmonary sequestrations, and congenital lobar lesion such as bronchogenic cyst, esophageal
emphysema (CLE) are conditions that benefit duplication, thymic lesions, or tumors (neuroblas-
from the thoracoscopic approach since this toma). Thoracoscopic resection of foregut dupli-
approach significantly improves the postopera- cation and bronchogenic cyst in present times is
14  Minimal Access Surgery in Infants and Children
considered safe and the preferred approach.
Bronchogenic cysts usually have no discrete con-
nections to other important structures in the chest.
Esophageal duplications, on the other hand, have
a common wall with the esophagus and some-
times have a luminal communication. All medias-
tinal dissections are performed with a bougie in
the esophagus. If there is a luminal communica-
tion between the esophagus and the esophageal
duplication, the mucosa and muscle of the remain-
ing esophagus is closed using sutures placed tho-
racoscopically. Great care must be taken not to
compromise the esophageal lumen during the clo-
sure of the esophagus. Thoracoscopic biopsy of
undefined mediastinal masses is a simple tech-
nique that avoids the large thoracotomy wound
and commonly employed for such biopsies.
14.2.7 Patent Ductus Arteriosus
Thoracoscopic clipping of a patent ductus arteri-
osus (PDA) is performed using three or four
ports. The patient is placed in a semi-prone posi-
tion to allow the lung to fall away from the poste-
rior thorax. The ductus is carefully dissected,
preserving the recurrent laryngeal nerve, which
loops around the PDA.  A clip applier is passed
directly through a small incision in the chest wall.
The clip is then applied under endoscopic vision.
The recurrent laryngeal nerve is observed while
the clip is being applied to avoid injury to the
nerve. Placement of a chest tube is optional,
depending upon the surgeon’s preference.
14.2.8 Esophageal Atresia
Esophageal atresia with or without tracheoesopha-
geal fistula is now being commonly approached by
the thoracoscopic approach (Iacona and Saxena
2020). The patient is positioned in the Cuschieri
modified lateral decubitus position (side depend-
ing on the aortic arch), and the initial camera port
is placed below the tip of the scapula. Two further
ports are placed in the midaxillary and the poste-
rior axillary line. The Azygos arch is recognized
167
and may or may not be divided. If divided, clips,
diathermy, or suture are the options that can been
employed. The distal trachea-esophageal fistula is
then visualized, ligated, and divided. Both upper
and lower esophageal pouch are then approxi-
mated and the anastomosis is performed hand-
sewn with 10–15 sutures tied intracorporeally. If
an anastomosis is not possible, an approach with
traction sutures is preferred; this is followed by a
delayed anastomosis in a few days. Suturing in this
setting is tedious because of the very small work-
ing space in the posterior chest. The procedure
should be performed by experienced pediatric
endoscopic surgeons comfortable in performing
thoracoscopic surgery in a small thoracic space.
14.2.9 Congenital Diaphragmatic
Hernias and Diaphragmatic
Eventration
Both Bochdalek and Morgagni diaphragmatic
hernias are repaired using endoscopic tech-
niques. Thoracoscopic repair of Bochdalek dia-
phragmatic hernias is performed in stable
patients but is not usually attempted in infants
with pulmonary hypertension and severe pulmo-
nary hypoplasia. Bochdalek diaphragmatic her-
nias are repaired using a primary closure
technique as well as by the application of a patch
graft. Repair of Bochdalek diaphragmatic hernia
is performed both thoracoscopically and laparo-
scopically; however, the author favors the thora-
coscopic approach. The reconstruction of the
diaphragm usually starts at the medial portion of
the defect. In larger left-sided defects, care must
be taken not to include the esophagus into the
first stitch. The tension should be assessed step-
by-step, since excessive tension will be the main
cause of recurrence. In case of doubt, either a
patch or conversion into open repair should be
considered.
Morgagni diaphragmatic hernias, on the other
hand, are better approached by laparoscopy
(Alqadi and Saxena 2019). Laparoscopy enables
the visualization and the confirmation of a bilat-
eral defect. The defect is repaired by an extra-
168
abdominal suturing technique, in which the
posterior rim of the hernia tis sutured to the full
thickness of the anterior abdominal wall using
the port closure needle without the need for a
mesh. In case of a large defect or tension in the
sutures, a mesh is necessary to avoid recurrence.
Eventration of the diaphragm and a high-­
riding paralyzed diaphragm are also plicated with
a thoracoscopic technique. If CO2 infusion is
used for the enhancement of intrathoracic visual-
ization, there is the added benefit of the dia-
phragm being pushed down by the pressure of the
infused pneumothorax. This pressure enlarges
the involved pleural space and aids the plication
of the hemidiaphragm. The plicating sutures are
placed and tied thoracoscopically.
14.3 Laparoscopic Procedures
Successful laparoscopy depends on adequate
expansion of the peritoneal cavity for clear visual-
ization of intra-abdominal structures achieved
through a pneumoperitoneum. Although many
gases have been utilized for maintaining a pneumo-
peritoneum, carbon dioxide (CO2) is the standard
gas that is used for procedures. The physiological
effects of CO2 pneumoperitoneum may be more
pronounced in infants and children than in adults.
Neonates have a higher level of end-tidal CO2 dur-
ing carbon dioxide pneumoperitoneum. This
enhanced absorption of CO2 pneumoperitoneum
can result in hypercapnia and respiratory acidosis.
Hypercapnia is usually overcome by increas-
ing the minute ventilation and by evacuating the
CO2. Postoperatively, there is a risk that the child
may be unable to maintain the increased ventila-
tory effort needed to ventilate off the CO2.
Increased intra-abdominal pressure also has
cardiovascular effects in the child. The younger
the child, the greater the potential cardiovascular
compromise. Pressures within the peritoneal cav-
ity above 10 mmHg can lead to the compression
of the inferior vena cava with a reduced venous
return to the heart and to secondary hypotension.
The infants’ sensitivity to intraperitoneal pres-
sures greater than 10  mmHg must be carefully
considered for safe laparoscopy.
A. K. Saxena et al.
14.3.1 Inguinal Hernia
Laparoscopic inguinal hernia repair and the eval-
uation of a patent contralateral processus vagina-
lis is commonly performed around the world.
Once a pneumoperitoneum is established, the
patency of the patent processus vaginalis can be
identified and the repair is performed. Placement
of a purse-string or figure-of-eight suture laparo-
scopically or transabdominally under laparo-
scopic vision is used to close the internal ring.
Variants in pediatric closures include simple clo-
sures of the inguinal ring or dissection of the
peritoneum at the level of the internal ring and
closure of the ring. Non-absorbable monofila-
ment or braided sutures are used in laparoscopic
repairs. A potential advantage of the laparoscopic-­
assisted inguinal hernia repair includes the avoid-
ance of injury to the cord structures and testicles,
and also closure of a patent contralateral side
without any further incisions in the same session.
The procedure is performed as a standard in neo-
nates using 3 mm instruments (Walsh et al. 2020).
In the classic open hernia repair, the hernia sac is
stripped away from the vas deferens and testicu-
lar vessels; many studies have suggested that this
stripping of the cord structures may produce
unintended permanent changes in the vas defer-
ens and testicles.
14.3.2 Pyloric Stenosis
The principles of the extramucosal pyloromyot-
omy described by Fredet and Ramstedt still
remains the standard management for the treat-
ment of pyloric stenosis also in the laparoscopic
approach. In the laparoscopic approach, a 5 mm
port is inserted under direct vision to accommo-
date a 30° scope. Two further abdominal access
points are established with or without work ports.
The author’s preference is to grasp and secure the
stomach using the left access work point. Using
the right access point, a sleeved retractable lapa-
roscopic knife is inserted to perform the pyloro-
myotomy, after which the muscle is spread using
a pyloromyotomy spreader also inserted from the
right side (Saxena 2013). Completion of the
14  Minimal Access Surgery in Infants and Children
procedure involves administering of air in the
stomach and confirming the integrity of the
mucosa; after which the air is evacuated though
the nasogastric tube. The procedure has a learn-
ing curve to become proficient in order to avoid
complications such as incomplete myotomies or
mucosal perforations.
14.3.3 Duodenal Atresia
Laparoscopic management of duodenal atresia
is becoming popular amongst pediatric surgeons
since the first reports of successful application
of this technique 2 decades ago (Mentessidou
and Saxena 2017). The neonate is positioned
supine at the end of the table and a 5  mm 30°
scope placed in the umbilicus is preferred for
this procedure. Abdominal inspection is per-
formed to exclude associated anomalies, such as
malrotation. Two further 3 mm ports are inserted
under direct vision within the right and left iliac
fossa. Once the duodenum is identified, a stay
suture is used to suspend the proximal pouch
to the abdominal wall. The proximal and distal
segments of the duodenum are mobilized and
approximated. The proximal pouch is incised
transversely and distal segment longitudinally
using laparoscopic scissors; avoiding electro-
cautery. A nasogastric tube is passed through
the distal segment to rule out a second atre-
sia or a possible web. The anastomosis is per-
formed intracorporeally with interrupted sutures
or with a running suture; both these suturing
techniques have shown comparable outcomes.
Alternatively, a duodeno-jejunostomy is an
option which is easier to perform laparoscopi-
cally than duodeno-­duodenostomy and has sim-
ilar outcomes.
14.3.4 Malrotation
Malrotation, with or without mid-gut volvulus,
can be managed laparoscopically. The malrota-
tion is detorsed, and a Ladd’s procedure per-
formed using three or four access ports (Chacon
and Saxena 2019). Appendectomy in neonates
169
and infants is performed by exteriorizing the
appendix through a port site and performing the
procedure conventionally. Performing a laparo-
scopic Ladd’s procedure in the presence of a vol-
vulus may be challenging as it may be difficult to
achieve sufficient visualization in the presence of
dilated bowel. There is some controversy in
reports emerging as to whether all the steps of
Ladd’s procedure are being followed or being
achieved in the laparoscopic approach besides
the detorsion of the volvulus and the division of
peritoneal bands attached to the cecum; these
include widening of the mesenteric root, appen-
dectomy and positioning of the small bowel to
the right and the colon to the left. Conversion rate
is high for this procedure and mainly related to
the degrees of volvulus, age of the patient, and
learning curve.
14.3.5 Gastroesophageal Reflux
and Gastric Feeding
Fundoplication, with or without gastrostomy, is a
frequent procedure performed in children.
Common indications include: primary aspiration,
gastroesophageal reflux, profound neurologic
impairment, and severe pulmonary and cardiac
disease with failure to thrive or recurrent aspira-
tion. Most infants and children needing fundopli-
cation are excellent candidates for a laparoscopic
approach. The fundoplication is performed using
3 mm instruments. The distal esophagus is mobi-
lized and secured in the abdomen, and the crus
are approximated behind the esophagus. The fun-
dal wrap is formed loosely around the newly cre-
ated intra-abdominal esophagus. Laparoscopic
fundoplication can be achieved using the Nissen,
Toupet, or Thal technique with comparable out-
comes depending on the volume of cases and
experience of the surgeon performing these
procedures.
Laparoscopic gastrostomy is usually per-
formed by placing a gastrostomy button. If the
gastrostomy is performed without fundoplica-
tion, an effort is made to site the gastrostomy at
an appropriate distance from both the gastro-
esophageal junction and the pylorus. Adequate
170
distance from the fundus is important, in case a
fundoplication is needed at a later date. The tech-
nique involves securing the stomach by a laparo-
scopic grasper passed through an abdominal
incision (predetermined site of the gastrostomy)
and passage of the needle alongside into the
stomach under laparoscopic vision to progress
the Seldinger wire. The gastrostomy button is
then positioned in the same steps as the percuta-
neous endoscopic gastrostomy. Many pediatric
surgeons now prefer the laparoscopic-assisted
gastrostomy to percutaneous endoscopic
­gastrostomy tube placement, as with the laparo-
scopic approach, the visualization of the entire
left upper quadrant allows for the selection of the
site of gastric entry, helps in avoiding nearby
structures such as the colon, and can easily be
combined with the performance of other laparo-
scopic procedures.
If a percutaneous gastro-jejunostomy tube
(PEG-J) is required, the stomach is insufflated by
the endoscope and under laparoscopic and endo-
scopic vision, T-fastners or U-sutures passed
through the abdominal wall into the stomach and
back out through the abdominal wall to secure
the stomach. A Seldinger technique is used to
obtain access to the gastric lumen, followed by
the dilation of the tract over a guide wire; after
which the PEG-J tube is inserted, with the jejunal
limb correctly positioned beyond the pylorus
under endoscopic vision.
14.3.6 Achalasia
The treatment of choice still remains the extramu-
cosal longitudinal modified Heller cardiomyot-
omy with the anterior Dor fundoplication to
avoid the onset of gastroesophageal reflux, main-
taining the myotomy open and protecting the
mucosa. Endoscopic myotomy limited to esopha-
geal rosette or the perusal endoscopic myotomy
technique (POEM) utilizing a submucosal tunnel
to reach the inner circular muscle bundle of the
lower esophageal sphincter are recent options to
achieve the myotomy. POEM is a safe, effective,
and feasible technique for the pediatric
population.
A. K. Saxena et al.
14.3.7 Gall Bladder Pathologies
The main indications for laparoscopic cholecys-
tectomy are the presence of symptomatic gall-
stones followed by gall bladder polyps and biliary
dyskinesia. The minimal access approach with
the use of four ports involves the retraction of the
gall bladder by grasper positioned in one port.
After the identification of Calot’s triangle, liga-
tion of the cystic artery and duct is performed.
Cholecystectomy is performed using a monopo-
lar hook employed in the liver–gallbladder inter-
face, while the gallbladder is retained under
traction. It is mandatory to stay close to the gall-
bladder wall to avoid the injury of the duct of
Luschka, which, if present will connect the gall-
bladder fossa entering the right/left hepatic duct
or either the cystic duct and if not seen can cause
postoperative bile leak. The gallbladder is
retrieved within an endobag to avoid spillage of
gallstones in the abdomen. Reported complica-
tions of cholecystectomy include cystic duct
leaks, duct of Luschka leaks, bleeding, infection,
herniation and deep vein thrombosis and injury to
the biliary tree.
14.3.8 Splenic Pathologies
Splenectomy in pediatric age found is mainly
performed for hematologic disorders.
Laparoscopic splenectomy involves a 4-port
technique. The spleno-colic ligaments are dis-
sected to release the lower pole attachments. This
is followed by the dissection of the spleno-­
phrenic ligament. Once the spleen is released
from its attachments the splenic vessels are
ligated using suture ties, endoscopic titanium
clips, vessel sealing devices, or vascular endo-
scopic staplers. The spleen after resection is
placed in an endobag and piecemealed to extract
it from the umbilical port site. Morcellators
should be avoided as they risk serious and fatal
injuries to the underlying structures. Also, the
introduction of sharp objects into the endobag
risks breaching of its integrity and spillage of the
splenic pulp into the abdominal cavity risking
splenosis.
14  Minimal Access Surgery in Infants and Children
14.3.9 Pancreatic Pathologies
Pancreatic tumors are rare in children. The most
common pancreatic tumors in pediatric popula-
tion are: pancreatoblastoma, solid pseudo papil-
lary tumors, and islet cell tumors. Among these,
insulinomas are insulin-secreting tumors arising
from pancreatic beta cells with 10% associated to
the multiple endocrine neoplasia (MEN) type 1.
Congenital hyperinsulinism of infancy (CHI) is a
condition characterized by the inappropriate
secretion of insulin causing persistent hypoglyce-
mia and leading toward serious neurologic
sequelae. However, recent advancements in mini-
mally access surgery have enabled surgeons to
perform laparoscopic pancreatectomy safely.
Most of these reports involve the adult population
with distal pancreatectomies. Few reports have
described the role of laparoscopic approach for
the treatment of these pathologies in children.
14.3.10 Meckel’s Diverticulum
The surgical technique consists in Meckel’s
diverticulum utilizes the capability of diagnostic
laparoscopy and management at the same time.
Though over the past 3 decades multiple laparo-
scopic approaches have be advocated, the
authors’ preference is for a laparoscopic-assisted
resection of Meckel’s diverticulum. Once the
diverticulum is identified, the diverticulum is
exteriorized through the umbilical port site and
the resection and anastomosis are performed. The
umbilical port site is enlarged especially with the
extension of the linea alba so as to accommodate
safe reduction of the anastomosis and prevent its
forced reposition. The enlarged skin incision
within the umbilical port site incision can be con-
cealed at the time of port site closure.
14.3.11 Intussusception
Laparoscopic approach is indicated when
attempts at enema reduction have failed or where
enema reduction is contraindicated; it can be also
useful as a diagnostic tool when other modalities
171
of investigations like ultrasound and contrast
enema were inconclusive like intussusception of
small bowel. This approach is contraindicated in
hemodynamic unstable children and when there
is evidence of small bowel obstruction with
marked abdominal distention and dilated bowel
loops. Laparoscopic reduction involves grasping
the intussusception and the intussuscepiens, and
pulling back the intussuscepiens while holding
the intussusception; this generally helps avoid
tears during the reduction. In case of multiple
presentations of intussusception and the absence
of lead points, a laparoscopic ileo-cecopexy can
be performed using non-resorbable sutures to
secure part of the terminal ileum approximated to
the cecum.
14.3.12 Appendicectomy
Laparoscopic appendectomy has taken its place
as the primary technique used to treat appendici-
tis. In pediatric surgery centers, converting to an
open operation after attempting laparoscopic
appendectomy is a very uncommon event. It is
performed with three ports using the sectoriza-
tion or the triangulation port placement. The
authors prefer the latter approach, as suprapubic
port placements in sectorization have been
reported to be associated with injuries to the
bladder during port placement. The appendix is
identified and separated from any attachments
using blunt and sharp dissection. The mesoap-
pendix is divided with bipolar forceps cautery
and scissors sequentially till the base of the
appendix is reached. The appendiceal stump is
ligated with 3 endoloops ligatures (and resected
between the 2 proximal and the distal endoloops)
or with a stapling device. The appendix is deliv-
ered through the umbilical port site.
Usually a urethral catheter is inserted at the
beginning of the procedure in order to empty
the bladder and decrease the risk of bladder
injury. Single-port appendicectomy is also per-
formed in some centers as well as the transum-
bilical laparoscopic-­ assisted appendicectomy
(TULAA). The laparoscopic approach has shown
a decrease in morbidity with reduced length of
172
hospital stay and rate of wound infections. The
incidence of a postoperative intraperitoneal
abscess is about the same as after open appendec-
tomy. Morbidity seems to be diminished with a
more rapid discharge from the hospital being
reported in several very large series. Laparoscopic
appendectomy has become the standard of care
for children with appendicitis.
14.3.13 Anorectal Malformations
High imperforate anus is usually repaired by a
posterior sagittal anorectoplasty (Georgeson et
al. 2000). The alternative laparoscopic repair of a
high imperforate anus starts by the dissection of
the rectourethral or rectovesicular fistula transab-
dominally. The muscle complex is identified by
muscle stimulation on the perineum. A tract is
developed through the muscle complex by identi-
fying the appropriate landmarks from both the
perineal and abdominal aspects. This tract is
sequentially dilated using a radially expanding
trocar. The fistula is pulled down to the perineum
through the tract and secured to the perineal skin
with sutures. The laparoscopic repair of a high
imperforate anus mimics the repair of a low
imperforate anus where the fistula to the perineum
is mobilized and pulled through the external
sphincter. Continence after laparoscopic pull-
through has not been fully assessed although the
early results appear to be promising.
The laparoscopic approach is mainly deemed
to treat high prostatic or bladder neck fistula in
males and congenital high rectovaginal fistula or
cloaca with high rectum in girls. The laparo-
scopic approach offers performing of the abdom-
inal part of the surgical procedure using the
minimal access route, thereby avoiding the mor-
bidities associated with laparotomy. The goals of
laparoscopic-assisted pull-through for high ano-
rectal malformations include precise placement
of the rectum inside the sphincter complex with-
out dividing and weakening the muscles and the
diminished soft tissue scarring around the rectum
leading to improved rectal compliance.
A. K. Saxena et al.
14.3.14 Hirschsprung’s Disease
The management of Hirschsprung’s disease has
been radically changed by minimally invasive
techniques (Georgeson KE 1995, 2002b). A pull-
through is utilized after laparoscopic identifica-
tion of ganglion cells proximal to the transition
zone. Although the pull-through can be per-
formed using a Duhamel or Swenson technique,
the most common procedure performed endo-
scopically is an endorectal pull-­ through.
Hirschsprung’s disease is now managed by a
single primary pull-through as opposed to the
two- or three-stage procedures formerly used to
correct the disorder.
Laparoscopic-assisted trans-anal endorectal
pull-through is a versatile and effective tech-
nique for all left and transverse colon agangli-
onic segments. A laparoscopic-assisted Duhamel
procedure is however preferred in case of
ascending colon and total colonic ganglionosis.
Contraindications to primary laparoscopic-­
assisted endorectal pull-through are associated
life-threatening anomalies, deteriorating general
health, severe enterocolitis, and severe dilatation
of the proximal bowel.
14.3.15 Inflammatory Bowel
Disease
Surgical procedures available for the manage-
ment of these conditions include total colectomy,
subtotal colectomy, ileocecal resection, ileorec-
tal anastomosis with or without the creation of
pouch (Georgeson 2002a). Four ports are used:
12 mm port is placed in the umbilicus, one 5 mm
port in the right upper quadrant, one in the left
flank region, and one 12  mm in the right iliac
region where the ileostomy is to be fashioned. A
fifth port may be used in the epigastric region, if
needed. Generally, the colon dissection starts at
the sigmoid colon by creating a mesenteric win-
dow, after which an endostapler is employed to
create a rectal stump. Mobilization of the colon
continues up to the hepatic flexure, always close
14  Minimal Access Surgery in Infants and Children
to the bowel wall in order to avoid injuries to
surrounding structures. Once the colon is entirely
mobilized, it is extracted from the right iliac
region by extending the incision. Ileostomy is
created also at this port site.
A three-port technique is preferred for the
laparoscopic right colectomy. Two 12  mm and
one 5 mm ports are used (one port in the suprapu-
bic area and other in the right iliac fossa).
Following the mobilization of the right lateral
peritoneal attachments by sharp dissection, the
mesocolon is dissected by using sealing devices.
The colon is dissected intracorporeally using a
linear endostapler. The anastomosis is finally per-
formed either intracorporeal or extracorporeal
suturing method.
14.4 Genitourinary System
14.4.1 Vesicoureteral Reflux
Different techniques are available for the sur-
gical treatment of vesicoureteral reflux (VUR).
Minimal access techniques were recently
developed in order to reduce postoperative
pain, avoid postoperative hematuria, and
shorten hospitalization. They include endo-
scopic treatment (not covered in this chapter),
vesicoscopic reimplantation, and laparoscopic
reimplantation. When laparoscopic Lich-
Gregoir reimplantation is performed, a cystos-
copy precedes the laparoscopic phase in order
to gain access to the bladder. Usually, a trans-
peritoneal approach is preferred using 3 ports,
a 5 mm umbilical port and two 3 mm ports on
the right and left flanks, respectively. The ure-
ter is identified at the level of the iliac vessels,
the peritoneum is breached at the vesicoure-
teral junction and mobilized to achieve suffi-
cient freedom for a tension free reimplantation.
The bladder dome is suspended to the abdomi-
nal wall with a stay suture. After bladder fill-
ing, the detrusor muscle is split to expose the
bladder mucosa. After placement of the ureter
in the muscle cleft, the detrusorrhaphy is
performed.
173
The vesicoscopic approach involves the inser-
tion of a 5-mm step-port inside the bladder under
cystoscopic vision. Following this the urethral
catheter is inserted to deflate the bladder and
allow the insufflation of CO2. Under direct vision,
further 2 ports (3-mm) are inserted. A 3–4 Fr
catheter is then inserted to cannulate the ureter
and secure it during its mobilization. Once ade-
quate length is gained, a submucosal tunnel is
created and the reimplantation is performed.
Limitations however with this technique are
small bladder capacity, surgeon’s skills in intra-
corporeal suturing and knot-tying and previous
failed endoscopic treatment or thick bladder wall.
14.4.2 Ureteropelvic Junction
Obstruction
Ureteropelvic junction obstruction (UPJO) is the
most common cause of hydronephrosis in infants
and children. The patient is positioned in lateral
decubitus or supine position with the affected
side slightly elevated. The three-port technique is
employed using a 5  mm 30° scope introduced
through the umbilicus and two ports triangulated
to reach the ureteropelvic junction. A transab-
dominal stay suture is placed to suspend the pel-
vis in order to facilitate suturing. The anastomosis
is performed with either continuous or inter-
rupted suturing. A double J stent is passed
through one of the ports and advanced with the
pusher. If the UPJO is related to an external
obstruction such as a crossing vessel, a vascular
hitch can be performed.
14.4.3 Benign Renal Pathologies
Minimal access surgery is one of the best appli-
cations for nephrectomy in benign conditions.
The approach can be either transperitoneal or ret-
roperitoneal according to surgeon preference and
skills. The patient is positioned in a prone lateral
decubitus will be required for the retroperitoneal
approach. A 10 mm 30° scope is preferred for the
transperitoneal approach while a 0° for the retro-
174
peritoneal approach. Three ports are needed in
both approaches. Due to the smaller space, retro-
peritoneoscopic approach is considered techni-
cally more demanding. For the retroperitoneal
approach, a single 5 mm incision is sufficient to
access the retroperitoneal renal space. This first
incision is placed midway between the 12th rib
and the iliac crest at the lateral border of the
sacro-spinal muscle, and the initial working
space is created with the use of a balloon (80–
120  ml air instillation) or swab to create the
working space. The camera port is then inserted,
and a 5  mm work port is inserted under direct
vision laterally to the camera port between the
11th rib and the iliac crest. The second work port
is placed in a medial position to achieve triangu-
lation. Ligation of the renal vessels for nephrec-
tomy can be achieved by sutures or vessel sealing
devices.
14.4.4 Ovarian Pathologies
Ovarian pathologies that can be approached by
laparoscopic surgery include ovarian cysts,
benign ovarian neoplasms, ovarian torsions,
streak gonads, and giant para-ovarian cysts.
Open surgery is the preferred option in malig-
nant neoplasms to avoid the risk of spillage and
tumor upstaging. Simple ovarian cysts are
punctured and evacuated under laparoscopic
vision. In case of large hemorrhagic cysts, a
cyst fenestration is performed and the contents
evacuated by aspiration. Ovarian torsions are a
good indication for laparoscopic surgery, with
the present recommendation to salvage necrotic
ovaries in children. Oopheropexy is performed
on the ipsilateral affected ovary and prophylac-
tically on the contralateral normal ovary based
on the surgeon’s preference. The authors’ pref-
erence is to perform oopheropexy using a non-
resorbable braided suture with a 2-point fixation
of the ovaries. Streak gonads can be resected
using vessel sealing devices, with a preference
for salpingo-­oophorectomy in this pathology.
Giant para-ovarian cysts are evacuated through
the initial port site incision to offer space in the
A. K. Saxena et al.
abdomen to perform the laparoscopic proce-
dure. The cyst wall resection can be performed
using vessel sealing devices. Benign teratomas
can be safely managed laparoscopically as an
ovary-sparing procedure without rupturing the
tumor cysts and mass (Raicevic and Saxena
2019). Careful dissection after breaching the
capsule enables step-by-step enucleation of the
entire mass leaving the residual ovarian tissue
in situ.
14.4.5 Impalpable Testis
In case of impalpable testis, laparoscopy offers
both a diagnostic and a management option.
Once the lie of the testis is determined, four
options are possible: (1) The presence if the testis
inside the inguinal ring warrants the management
with an open orchidopexy in the same setting.
Abdominal testis that cannot be mobilized for
scrotal fixation can be managed by one of the
following first stage options: (2) Performing a
laparoscopic first-stage Fowler-Stephens proce-
dure in which the testicular vessels are cauter-
ized by bipolar forceps and dissected. During the
second-­stage laparoscopic Fowler-Stephens pro-
cedure, a large peritoneal flap is created around
the testicle before it is mobilized into the scro-
tum via the Prentiss maneuver, in which a port is
introduced through the scrotal incision and
passed directly into the abdomen through the
external inguinal ring to retrieve the testis for
orchidopexy. (3) Vessel-sparing technique-I:
Laparoscopic application of a traction suture on
the testis and securing it to the abdominal wall
on the contralateral port site. In the second stage
that is also performed with, the testis is released
from its abdominal wall attachments and due to
sufficient length gained, an orchiopexy is per-
formed either using the Prentiss maneuver or by
routing the testis through the inguinal canal. (4)
Vessel-sparing technique-II: Laparoscopic appli-
cation of a traction suture on the testis and pass-
ing the suture through the ipsilateral inguinal
canal for traction fixation in the scrotum. Since
the testis is placed in the inguinal canal with this
14  Minimal Access Surgery in Infants and Children
technique, the second stage does not involve
laparoscopy, but only an incision in the inguinal
canal and orchidopexy.
14.5 Pediatric Oncology
The use of laparoscopy, thoracoscopy, and
robotic techniques in pediatric oncology is being
increasingly reported; however, controversies
regarding its application in specific tumors
remain. Although minimal access undoubtedly
improves postoperative morbidity (pain, length
of hospital stays, and cosmesis), uncertainty
toward its compliance with the oncologic princi-
ples remain. Minimal access approach is
employed for the management of abdominal and
thoracic tumors such as adrenal tumors (adreno-
cortical carcinoma, pheochromocytomas, adrenal
adenomas, sarcomas, ganglioneuromas, ganglio-
neuroblastomas, and neuroblastoma), pancreatic
tumors (pancreatoblastomas and pseudo papil-
lary tumors), liver tumors, ovarian tumors, sacro-
coccygeal teratoma (Altman type 4 or to ligate
the feeding vessel in Altman type 1–3).
Irrespective of the minimal access approach,
the following criteria should be respected in min-
imal access tumor surgery: (1) optimal exposure
of the operative field both in the abdomen and
thorax (with preference for single lung ventila-
tion for thoracic masses), (2) dissection should
proceed from the periphery to the central located
vital structure with instruments familiar to the
surgeon, (3) safe and secure removal of the tumor
in endobags after resection and extending inci-
sions if necessary, and (4) a clear plan to manage
intraoperative bleeding.
14.6 Conclusion
Minimally invasive surgical techniques are play-
ing an expanding role in pediatric surgery.
Despite the growing use of endoscopic surgery in
175
children, the current literature supporting the
safety and efficacy of thoracoscopy and laparos-
copy in children is based on relatively small num-
bers of patients. It seems clear, even with these
small numbers, that children tolerate minimal
access techniques well but have specific issues
that must be recognized and respected to achieve
safe results.
References
Alqadi G, Saxena AK (2019) Laparoscopic Morgagni her-
nia repair in children: systematic review. J Ped Endosc
Surg 1:85–90
Chacon CS, Saxena AK (2019) Approach to intestinal
malrotation in children in the laparoscopic era. J Ped
Endosc Surg 1:137–142
Georgeson KE (2002a) Laparoscopic-assisted total colec-
tomy with pouch reconstruction. Semin Pediatr Surg
11:233–236
Georgeson KE (2002b) Laparoscopic-assisted pull-­
through for Hirschsprung’s disease. Semin Pediatr
Surg 11:205–210
Georgeson KE, Fuenfer MM, Hardin WD (1995) Primary
laparoscopic pull-through for Hirschsprung’s disease
in infants and children. J Pediatr Surg 30:1–7
Georgeson KE, Inge TH, Albanese CT (2000)
Laparoscopically assisted anorectal pull-through for
high imperforate anus: a new technique. J Pediatr Surg
35:927–931
Iacona RV, Saxena AK (2020) Thoracoscopic repair of
esophageal atresia with distal tracheoesophageal
fistula (type c): systematic review. Surg Laparosc
Endosc Percutan Tech. https://doi.org/10.1097/
SLE.0000000000000832
Mentessidou A, Saxena AK (2017) Laparoscopic repair
of duodenal atresia: systematic review and meta-­
analysis. World J Surg 41(8):2178–2184
Raicevic M, Saxena AK (2019) Review of laparoscopic
management of mature cystic teratoma of ovaries in
children. J Indian Assoc Pediatr Surg 24(2):92–96
Saxena AK (2013) Ergonomic laparoscopic spreader for
infantile hypertrophic pyloric stenosis. Technol Health
Care 21(6):625–630
Saxena AK, Höllwarth ME (2009) Essentials of pediatric
endoscopic surgery. Springer, Heidelberg
Walsh CM, Ng J, Saxena AK (2020) Comparative analy-
sis of laparoscopic inguinal hernia repair in neonates
and infants. Surg Laparosc Endosc Percutan Tech.
https://doi.org/10.1097/SLE.0000000000000815
 Surgical Safety in Children
Iain Yardley and Holbrook Charlotte
15.1 Introduction
Despite medicine’s long and noble history, it is
only relatively recently, perhaps in the last thirty
years or so, that there has been a widespread
acknowledgement that healthcare is intrinsically
risky. Despite the best of intentions, it carries a
risk of harming the patients, rather than making
them well. During the 1990s, several studies
(Brennan et  al. 1991; Leape et  al. 1991) and
reports (Dept. of Health 2000; Institute of
Medicine 1999) were published that laid out the
scale of the problem in shocking detail. As many
as one in ten patients admitted to hospital was
exposed to some form of harm during their stay.
One dramatic headline compared the number of
deaths from complications occurring during
healthcare to a fully laden passenger aircraft
crashing every week. These studies documented
the many different ways in which patients could
come to harm during their care. These included
clearly defined, discrete events such as operating
on the wrong body part (Kwaan et al. 2006), or
accidentally giving a massive drug overdose
(Kaushal et al. 2001), to occurrences that could
be dismissed as inevitable components of hospi-
tal care such as hospital-acquired infections
I. Yardley (*) · H. Charlotte
Evelina London Children’s Hospital, London, UK
e-mail: iain.yardley@gstt.nhs.uk;
charlotte.holbrook@gstt.nhs.uk
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_15
15
(Dept. of Health 2000; Institute of Medicine
1999). Whilst it is possible to identify the effects
of unsafe care, defining what constitutes safe care
is rather more elusive and can only really be
described by the absence of dangerous care.
World Health Organisation (WHO) defines
“Patient Safety” as “the prevention of errors and
adverse effects to patients associated with health
care” (WHO n.d.). Of course, many adverse out-
comes are not related to errors, either of omission
or commission, and are unavoidable conse-
quences of the patient’s condition. Patient safety
does not aim to eliminate adverse outcomes,
rather it aims to minimise them by eliminating
those that are potentially preventable. For exam-
ple, blood loss may be inevitable during surgery
and so haemorrhage is not in itself necessarily
due to an error. However, a failure to prepare for
blood loss, such as not having adequate venous
access for resuscitation or blood products for
transfusion is a potentially preventable error.
Over the intervening thirty years, patient
safety has matured into its own specialty with its
own terminology, methodology and whole scien-
tific journals dedicated to reporting studies in the
field. As a young specialty, patient safety bor-
rowed heavily from the experiences of other
high-risk industries, most notably the aviation
industry, which had gone on its own safety jour-
ney in the latter half of the twentieth century. An
early realisation for those working in the field of
patient safety was that the attitude generally
177
178
taken to errors in healthcare, the “person
approach”, was fundamentally flawed. In this
model, if something untoward happened, it was
assumed that it was because someone (or a num-
ber of people) had done something wrong and the
remedy was to reprimand, remind or educate the
individual so that they would not make the same
mistake again. Of course, in the vast majority of
cases, the individual caregivers involved in a
safety incident are nothing but caring and consci-
entious and the real reason that incidents had
occurred was that the system in which these dedi-
cated individuals worked and in which patients
were cared for was flawed and vulnerable
(Reason 2000). Rather than attempting to change
individuals’ behaviour, changing and strengthen-
ing the systems that staff work in is much more
likely to lead to improvements in safety—the
“system approach”. This approach acknowledges
and accepts that even the most diligent and highly
trained individuals will make errors from time to
time and that trying to identify the errors and
mitigate them is more likely to be successful than
attempting to prevent them from occurring in the
first place. Perhaps the most famous illustration
of the system approach is James Reason’s “Swiss
Cheese Model” (Reason 1990). This model
describes systems as consisting of several layers
of protection (the slices of cheese) between the
caregiver and the patient; as long as at least one
of the layers catches any error an individual may
make, it is prevented from affecting the patient.
However, no layer of protection is failsafe and
may miss an error, allowing it to pass to the next
layer (the holes in the cheese). It is only when all
the layers of protection collectively fail (the holes
in the cheese lining up) that the error can reach
the patient and cause harm. A system approach to
improving safety will aim to do two things: (1)
Increase the number of layers of protection. (2)
Make the holes in the layers as small as possible.
To take the example of errors in drug dosing and
administration, an example of increasing the lay-
ers of protection is to have a pharmacist check
each prescription written by the doctors. An
example of making the holes as small as possible
is to have pre-prepared giving sets with the cor-
I. Yardley and H. Charlotte
rect dose and dilution of the drug, so that ward-­
based staff do not have to prepare the dose
themselves.
15.2 Measures to Improve Safety
As its own discipline, patient safety has amassed
a significant body of evidence of successful inter-
ventions to improve safety. Unfortunately, com-
paratively little work on patient safety has been
undertaken in paediatric surgery, with the major-
ity of studies that have concentrated on paediatric
surgery published only in the past decade
(MacDonald and Sevdalis 2017). This leaves
paediatric surgeons reliant on lessons learnt in
other surgical disciplines. There have been many
studies published that investigate the nature of
safety incidents in surgery but rather fewer that
describe successful means to reduce their occur-
rence. Key safety interventions that have been
developed for surgery include checklists, inci-
dent reporting and training in non-technical
skills.
15.2.1 Checklists
The publication of the results of a worldwide trial
of new, WHO-backed, surgical checklist in 2008
is considered a landmark in patient safety
(Haynes et al. 2009). The study claimed the use
of the three-part, nineteen-item, checklist
(Table 15.1) nearly halved death rates associated
with surgical care and significantly reduced com-
plication rates. Other studies, including a ran-
domised controlled trial (Haugen et  al. 2014),
also found significant improvements with the use
of the checklist and backed up these findings.
This led to widespread promotion of the use of
the checklist and its introduction to surgical prac-
tice across the world. Subsequent studies exam-
ining the impact of the checklist in clinical
practice out with a trial have been more equivo-
cal, however, with some finding that the use of
the checklist was associated with a lower crude
hospital mortality rate (Jammer et al. 2015) but
15  Surgical Safety in Children 179

Table 15.1  Surgical safety checklist (Based on WHO others finding no change after the introduction of
Surgical Safety Checklist 2009, who.int.) the checklist (Urbach et al. 2014). It would seem
Sign out (before that checklists themselves are not sufficient, but
the patient leaves that their use may well form a useful part of a
Sign in (before Time out (before the operating
anaesthetic) incision) room) broader effort to improve safety by simplifying
Has the patient Confirm all team Verbally and standardising workflows, providing feedback
confirmed his/ members have Confirms: and guidance to staff and addressing the wider
her identity, site, introduced safety culture (Bosk et al. 2009).
procedure and themselves by
consent? name and role.
Is the anaesthesia Confirm the The name of the
machine and patient’s name, procedure 15.2.2 Reporting and Learning
medication procedure and
check complete? where the Incident reporting was an important part of safety
incision will be improvement in aviation, with all staff being
made
Is the pulse Has antibiotic Completion of
encouraged to report safety issues. This was
oximeter on the prophylaxis been instrument, taken on enthusiastically by healthcare safety
patient and given within the sponge and workers and significant investment in reporting
functioning? last 60 minutes? needle counts and learning systems was made, with some
Does the patient Anticipated Specimen amassing millions of reports over the years
have a: Critical Events labelling
Known allergy? To Surgeon Whether there (National Reporting and Learning Service n.d.).
are any While these have enabled potentially preventable
equipment safety issues (such as misplacement of nasogas-
problems to be tric tubes or intrathecal administration of metho-
addressed
trexate) to be identified as recurrent and prompted
Difficult airway What are the To Surgeon,
or aspiration critical or Anaesthetist and attempts to reduce the risk of these events occur-
risk? non-routine Nurse ring again, the ability of reporting and learning
steps? systems to eliminate these issues remains
Risk of >7 ml/kg How long will What are the key elusive.
blood loss? the case take? concerns for
recovery and In surgical practice, a more local form of
management of reporting and learning maybe more effective at
this patient? improving safety. Widely used in the airline
What is the industry and in the military, a process of debrief-
anticipated blood
loss?
ing gives an opportunity for all involved in an
To Anaesthetist event to reflect and to feed back their experi-
Are there any ences. In surgical practice, debrief should ideally
patient-specific occur in the immediate post-operative period and
concerns? enable the day’s events, both positive and nega-
To Nursing
Team:
tive, to be reviewed and discussed. This process
Has sterility can improve team cohesion and safety culture
(including but, crucially, needs the engagement and leader-
indicator results) ship of senior surgeons in a way that enables all
been confirmed? team members to contribute (Lark et  al. 2018).
Are there
equipment issues
Even with a team orientated approach like
or any concerns? debriefing, there is still a significant individual
Is essential burden on the surgeon to prevent and mitigate
imaging complications. This will best come about through
displayed?
a detailed understanding of each condition
180
treated, the commonest complications and the
best response to each complication (Linnaus and
Ostlie 2016).
15.2.3 Non-technical Skills
Non-technical skills are yet another aspect of
safe practice that have been borrowed by health-
care from aviation. The concept refers to a set of
skills beyond the ability to operate, including
situational awareness, decision-making, commu-
nication and leadership (Flin et al. 2007). These
skills are vital for safe and effective surgical
practice but have been neglected in traditional
surgical training where the focus has been on
clinical knowledge and technical expertise. They
can, however, be measured, taught and seen to
improve over time (Yule et al. 2009).
15.3 Safety Concerns Specific
to Surgery
Adverse events in surgery often appear starker,
and consequently receive more attention than
other types of safety incident. This is due to a sur-
gical procedure being a well-defined and discrete
event, whereas other stages in a patient’s care,
such as establishing a correct diagnosis, occur at
less distinct points in time. There are three
surgery-­specific incident types that have been
termed “never events”. Never events are patient
safety incidents that are considered preventable
when national guidance or safety recommenda-
tions that provide strong systemic protective bar-
riers are implemented by healthcare providers
(NHS Improvement 2015). If they do happen,
then it suggests there may be serious failings in
the approach taken by the organisation concerned
to patient safety (Lembitz and Clarke 2009). The
never events that relate specifically to surgical
care are (1) operating on the wrong patient or on
the wrong body part, (2) inserting the wrong
implant/prosthesis and (3) leaving items inside
the patient unintentionally following surgery.
Despite the term “never event”, these incidents
I. Yardley and H. Charlotte
still occur across healthcare settings and have
shown little sign of decreasing in frequency
(Moppett and Moppett 2016). Although it may
seem hard to comprehend how such errors could
occur with competent and conscientious staff on
duty, the systems in operation during surgical
care are incredibly complex and there are many
opportunities for errors to occur, no matter how
assiduous individual staff members are. For
example, when performing a left inguinal herni-
otomy the surgeon needs to make no fewer than
eight mental steps after encountering the supine
patient to fully transfer their understanding and
concept of “left” to that of the patient’s left groin
and operate on the correct side (Seiden and
Barach 2006). An error in any one of these steps
will potentially lead to operating on the incorrect
side.
In addition to these clearly defined errors,
there are many more subtle ways in which surgi-
cal patients can suffer harm during their care.
These sources of harm can be preoperative, such
as delays in diagnosis; intraoperative, such as
allowing the patient’s body temperature to drop
below an acceptable limit; and during the postop-
erative period, such as developing surgical site
infection. It is often impossible to identify one
single “cause” of the adverse outcomes due to the
complexity of the care provided and the multi-
plicity of factors affecting outcomes. No single
distinct event, error or omission will be respon-
sible for some unexpected outcomes, such as a
surgical site infection. Rather, it is a combination
of factors including timely surgery, appropriate
antibiotic prophylaxis, good temperature control
intra-operatively and so on that is required to
reduce the risk of infections. This has led to the
concept of “safety bundles”, where several mea-
sures known to improve outcomes are packaged
together and deployed in a patient’s care. A suc-
cessful example in paediatric surgery is a safety
bundle to reduce wound infection following
stoma closure in children. This includes bowel
preparation, careful skin preparation, rationalised
prophylactic antibiotics, intra-operative warming
and adherence to a careful closing protocol
(Ahmad et al. 2020).
15  Surgical Safety in Children
15.4 Safety Concerns Specific
to Paediatric Surgery
The paediatric surgical patient is susceptible to
all of the same hazards as any other surgical
patient but in addition these patients have unique
attributes that increase their vulnerability to harm
during their care.
15.4.1 Size
Paediatric surgical patients are simply physically
smaller than other surgical patients, sometimes
extremely small (RCPCH 2009). This leads to
increased risk from several sources. Firstly,
decreasing the size of a body rapidly increases
its surface area to volume ratio, which means
that they will lose heat far more rapidly than
older patients and so are at risk of hypothermia
and its associated complications. Particularly in
the case of premature infants, their small size
makes them more physically fragile than older
patients, with risk of iatrogenic injury including
fractures, particularly during transfer within or
between healthcare facilities (Madar and
Milligan 1994). Physical fragility is also a source
of risk when inserting tubes, such as the danger
of oesophageal perforation with a nasogastric
tube (Gander et al. 2009) or bladder perforation
with a urinary catheter (Birs et al. 2016). Vascular
access to deliver vital medications and fluids and
to monitor the patient’s condition is also more
challenging, in smaller patients, often requiring
multiple attempts (Cuper et al. 2012). Tiny, frag-
ile veins make extravasation injuries a signifi-
cant risk (Wilkins and Emmerson 2004). A
reduced circulating blood volume means that a
small volume blood loss can lead to significant
compromise of the paediatric circulation.
Finally, operating on smaller anatomical struc-
tures makes surgery more technically challeng-
ing and demanding, increasing the risk of
complications.
181
15.4.2 Anatomy
Although, of course, an infant or child’s basic
anatomical set up is the same as that of an adult,
there are significant differences that can make the
child vulnerable to harm during surgical care,
particularly if the caregiver is not familiar with
these differences. For example, the liver is pro-
portionately much larger in new-borns and con-
sequently not so well protected by the lower ribs.
This, in combination with the fragility of the
liver capsule, can make intraoperative liver lac-
eration more likely. Similarly, the underdevelop-
ment of the pelvis makes the bladder an
intra-abdominal organ and so more vulnerable to
damage during surgery, particularly laparoscopic
surgery (Hotonu and Gopal 2019).
15.4.3 Growth
Although growth is desirable in a child and a vital
indicator of health, it can create a source of risk
for the paediatric surgical patient, particularly the
long-term patient. As children grow, they may
outgrow implanted devices such as vascular cath-
eters, creating the risk of extravasation injury.
They will also need medication doses to be
reviewed and adjusted regularly, particularly dur-
ing periods of rapid growth such as puberty to
prevent underdosing.
15.4.4 Immature Organ Systems
In the neonatal surgical patient, many organ sys-
tems are not yet fully mature and do not function
as the “adult” version would. Although this is
clearly the case in infants born prematurely, it is
also true of the term newborn. This includes ana-
tomical variations such as the cardiovascular sys-
tem’s change from the fetal to the newborn
circulation leaving the potential to shunt from the
left circulation to the right and vice versa during
182
surgery. It also includes more intrinsic forms of
immaturity such as the myocardium itself, the
liver’s ability to metabolise drugs and toxins and
the kidney’s urine concentrating capacity. In
addition to this, there are immaturities in the non-­
solid organ systems including the coagulation
cascade, the endocrine and the autonomic ner-
vous systems (Caliskan 2017). All these immatu-
rities make care of the neonate in the perioperative
period challenging, especially for inexperienced
or non-specialist practitioners.
15.4.5 Acuity of Problems
Many paediatric and neonatal surgical conditions
present acutely and unpredictably. Even where
conditions can be diagnosed antenatally, the tim-
ing of birth is not usually fully controllable. In
many paediatric surgical conditions (for example
gastroschisis, intussusception and necrotising
enterocolitis), surgical intervention is required
urgently and so cannot be planned for in advance.
It may be necessary to undertake procedures out
of normal working hours when risks are known
to increase (Hull et al. 2012).
15.4.6 Co-morbidities
It is not just the surgical condition that affects
outcome in paediatric surgical patients. Although
much of paediatric surgery is carried out in other-
wise healthy children, a significant proportion is
performed in children with serious co-existing
medical problems. Over two thirds of babies
undergoing surgery in the neonatal period are
classified as ASA 3 or higher (Bucher et  al.
2016). The surgical procedure may be related
only indirectly to the co-existing condition, for
example gastrostomy insertion in children with
metabolic conditions, or as a direct consequence
of the co-morbidity, for example tumour resec-
tion in an oncology patient. Regardless of the rea-
son for the procedure, co-morbidities add
significantly to the risk of surgery and must be
actively managed perioperatively.
I. Yardley and H. Charlotte
15.4.7 Healthcare System Limitations
The system in which healthcare is provided can
itself present risks to the paediatric surgical
patient. If a patient presents to a facility that is
unable to provide the care required, then a refer-
ral and transfer to another facility may be
required. This has the potential to introduce
delays and worsened outcomes in a number of
conditions (Thakkar et al. 2018). Also, paediatric
patients will also eventually reach an age where
they cannot be cared for in a paediatric facility,
necessitating transition to adult services. This
transition is not always a smooth one, leaving
patients vulnerable to falling between two pro-
viders, missing out on care needed and suffering
psychological, if not physical harm (Yardley
et al. 2009).
15.5 Summary and Conclusions
Healthcare provision is an inherently hazardous
undertaking. Paediatric Surgery is no exception
to this, carrying not just generic healthcare risks;
but also some additional, specific risks due to the
intrinsic characteristics of the patients cared for,
the nature of the conditions treated, and the sys-
tem in which care is delivered. Over the past
thirty years, much has been learned about how to
improve patient safety. Although little of this is
specific to paediatric surgery, plenty can be
applied to the field. There remains, however, a
need for more research into paediatric, and par-
ticularly neonatal, surgical safety.
References
Ahmad H, Nordin AB, Halleran DR, Kenney B, Jaggi
P, Gasior A et  al (2020) Decreasing surgical site
infections in pediatric stoma closures. J Ped Surg
55(1):90–95
Birs AS, Perez JA, Rich MA, Swana HS (2016) Delayed
diagnosis of iatrogenic bladder perforation in a neonate.
Case Rep Urol. https://doi.org/10.1155/2016/1425373
Bosk CL, Dixon-Woods M, Goeschel C, Pronovost
P (2009) Reality check for checklists. Lancet
374:444–445
15  Surgical Safety in Children
Brennan AT, Leape LL, Laird NM, Hebert L, Localio AR,
Lawthers AG et al (1991) Incidence of adverse events
and negligence in hospitalized patients—results of
the Harvard Medical Practice Study I. N Engl J Med
324:370–376
Bucher BT, Duggan EM, Grubb PH, France DJ, Lally
KP, Blakely ML (2016) Does the American College
of Surgeons National Surgical Quality Improvement
Program pediatric provide actionable quality
improvement data for surgical neonates? J Ped Surg
51:1440–1444
Caliskan E (2017) Neonatal care for anesthesiologists [Cited
2020 Nov 23]. In: Barria RM, Editor. Selected Topics
in Neonatal Care. Intechopen. https://doi.org/10.5772/
intechopen.71952. Available from: https://www.inte-
chopen.com/books/selected-­topics-­in-­neonatal-­care/
neonatal-­care-­for-­anesthesiologists
Cuper NJ, de Graaff JC, van Dijk ATH, Verdaasdonk
RM, van der Werff DBM, Kalkman CJ (2012)
Predictive factors for difficult intravenous cannula-
tion in pediatric patients at a tertiary pediatric hos-
pital. Paediatr Anaesth 22(3):223–229. https://doi.
org/10.1111/j.1460-­9592.2011.03685.x
Department of Health (2000) An organisation with a
memory: report of an expert group on learning from
adverse events in the NHS Chaired by the Chief
Medical Officer, London, The Stationary Office
Flin R, Yule S, Paterson-Brown S, Maran N, Rowley DR,
Youngson GG (2007) Teaching surgeons about non-­
technical skills. Surgeon 5:86–89
Gander JW, Berdon WE, Cowles RA (2009) Iatrogenic
esophageal perforation in children. Pediatr Surg Int
25:395–401
Haugen AS, Søfteland E, Almeland SK, Sevdalis N,
Vonen B, Eide G et  al (2014) Effect of the World
Health Organization checklist on patient outcomes:
a stepped wedge cluster randomized controlled trial.
Ann Surg 261:821–828
Haynes AB, Weiser TG, Berry WR, Lipsitz SR, Breizat
A-HS, Dellinger EP et  al (2009) A surgical safety
checklist to reduce morbidity and mortality in a global
population. N Engl J Med 360:491–499
Hotonu SA, Gopal M (2019) Bladder injury in a child dur-
ing laparoscopic surgery. J Surg Case Rep 2019, Issue
2, February 2019. https://doi.org/10.1093/jscr/rjz043
Hull L, Arora S, Aggarwal R, Darzi A, Vincent C, Sevdalis
N (2012) The impact of nontechnical skills on techni-
cal performance in surgery: a systematic review. J Am
Coll Surg 214:214–230
Institute of Medicine (1999) To err is human: building a
safer health system. National Academy of Sciences,
New York
Jammer I, Ahmad T, Aldecoa C, Koulenti D, Goranović
T, Grigoras I et  al (2015) Point prevalence of surgi-
cal checklist use in Europe: relationship with hospital
mortality. Br J Anaesth 114(5):801–807. https://doi.
org/10.1093/bja/aeu460
183
Kaushal R, Bates DW, Landrigan C, McKenna KJ, Clapp
MD, Federico F, Goldmann DA (2001) Medication
errors and adverse drug events in pediatric inpatients.
JAMA 285:2114–2120
Kwaan MR, Studdert DM, Zinner MJ, Gawande AA
(2006) Incidence, patterns and prevention of wrong
site surgery. Arch Surg 141:353–358
Lark ME, Kirkpatrick K, Chung KC (2018) Patient safety
movement: history and future directions. J Hand
Surg Am 43(2):174–178. https://doi.org/10.1016/j.
jhsa.2017.11.006
Leape LL, Brennan AT, Laird N, Lawthers AG, Localio
AR, Barnes BA et  al (1991) The nature of adverse
events in hospitalized patients—results of the
Harvard Medical Practice Study II.  N Engl J Med
324:377–384
Lembitz A, Clarke TJ (2009) Clarifying “never events”
and introducing “always events”. Patient Saf Surg 3:26
Linnaus ME, Ostlie DJ (2016) Complications in common
general pediatric surgery procedures. Sem Ped Surg
25(6):404–411
Macdonald AL, Sevdalis N (2017) Patient safety improve-
ment interventions in children’s surgery: a systematic
review. J Ped Surg 52:504–511
Madar RJ, Milligan DWA (1994) Neonatal transport:
safety and security. Arch Dis Child Fetal Neonatal Ed
71:F147–F148
Moppett IK, Moppett SH (2016 Jan) Surgical caseload
and the risk of surgical Never Events in England.
Anaesthesia 71(1):17–30. https://doi.org/10.1111/
anae.13290
National Reporting and Learning Service (n.d.). https://
report.nrls.nhs.uk/nrlsreporting/. Accessed 23 Nov
2020
NHS Improvement (2015) Revised Never Events policy
and framework. Available at: https://improvement.nhs.
uk/resources/never-­events-­p olicy-­a nd-­f ramework/.
Accessed 23 Nov 2020
Reason J (1990) Human Error. Cambridge University
Press
Reason J (2000) Human error: models and management.
BMJ 320:768–770
Royal College of Paediatrics and Child Health (2009)
UK-WHO growth charts  – neonatal and infant close
monitoring (NICM). Available at: https://www.rcpch.
ac.uk/resources/uk-­w ho-­g rowth-­c harts-­n eonatal-­
infant-­close-­monitoring-­nicm. Accessed Nov 23 2020
Seiden SC, Barach P (2006) Wrong-side/wrong-site,
wrong-procedure, and wrong-patient adverse events:
are they preventable? Arch Surg 141:931–939
Thakkar HS, Yardley I, Kufeji D (2018) Management of
Paediatric Testicular Torsion  – Are we adhering to
Royal College of Surgeons (RCS) recommendations.
Ann R Coll Surg Engl 100(5):397–400. https://doi.
org/10.1308/rcsann.2018.0041
Urbach DR, Govindarajan A, Saskin R, Wilton AS, Baxter
NN (2014) Introduction of surgical safety checklists in
Ontario, Canada. N Engl J Med 370:1029–1038
184
Wilkins CE, Emmerson AJB (2004) Extravasation inju-
ries on regional neonatal units. Arch Dis Child Fetal
Neonatal Ed 89:F2745
World Health Organization (n.d.) Patient Safety. Available
at http://www.euro.who.int/en/health-­topics/Health-­
systems/patient-­safety. Accessed 23 Nov 2020
Yardley IE, Pauniaho S, Baillie CT, Turnock RR, Coldicutt
P, Lamont GL, et  al (2009) After the honeymoon
I. Yardley and H. Charlotte
comes divorce: long-term use of the antegrade conti-
nence enema procedure. J Pediatr Surg 44(6):1274–
1276; discussion 1276–7. https://doi.org/10.1016/j.
jpedsurg.2009.02.030
Yule S, Rowley D, Flin R, Maran N, Youngson GG,
Duncan J et al (2009) Experience matters: comparing
novice and expert ratings of non-technical skills using
the NOTSS system. ANZ J Surg 79:154–160
 Surgical Problems of Children
with Physical Disabilities
Casey M. Calkins
16.1 Introduction
A person with a disability is defined as an indi-
vidual with a physical or mental impairment that
substantially limits one or more major life activi-
ties, including the ability to work, has a record of
such impairment, or is regarded as having such an
impairment (Perkins and Agrawal 2018). This
chapter will focus on the unique surgical needs of
disabled children with special health care needs
(CSHCN), who are defined as children who have,
or are at, increased risk for a chronic physical,
developmental, behavioral, or emotional condition
and require health and related services of a type or
amount beyond that required by children generally
(Kuo et al. 2016). Based on a national survey of
CSHCN published in 2002, approximately 40,000
children have a special health care need, and one in
five homes in the USA house a child or youth with
a special health care problem (van Dyck et  al.
2002). CSHCN currently represent approximately
10–20% of children in the United States and con-
sume approximately 50% of children’s health
expenditures (Administration 2016).
Children with medical complexity (CMC) are
a subset of CSHCN who are considered the most
medically fragile and utilize the greatest amount
C. M. Calkins (*)
Division of Pediatric Surgery, Medical College of
Wisconsin, Children’s Hospital of Wisconsin,
Milwaukee, WI, USA
e-mail: ccalkins@chw.org
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_16
16
of medical and surgical resources due to subspe-
cialty care needs and functional limitations
(Berry et al. 2015). The prevalence of CMC con-
tinues to grow as advances in health care afford
innovative therapies for once thought-to-be life
ending conditions. Not only are the health care
histories of these children more complex, they
are often ill-defined and further complicated by
participation from multiple providers, and remain
more likely to seek emergency care than their
peers, and often for more complicated issues
related to their unique needs (Kuo et  al. 2016;
Neff et al. 2015; Stille and Antonelli 2004). There
are a host of other definitions that are used in the
medical literature to describe children with medi-
cal complexity such as “children with complex
chronic conditions,” “medically fragile children,”
and “children with severe global developmental
delay” (Allen et al. 2019). This chapter will focus
on some of the most common perioperative
needs that are encountered in these children.
Pediatric surgical medically complex patients
typically have either a genetic malformation
syndrome (i.e., Trisomy 21), cerebral palsy or
neurologic impairment, complex congenital
heart disease, or a spectrum of comorbid condi-
tions as the result of a congenital malformation.
This list is by no means exhaustive, but this
chapter will serve to highlight many of the
unique challenges that must be addressed by the
pediatric surgeon caring for this complex group
of patients (Table 16.1).
185
186
Table 16.1  Common definitions utilized
CSHCN Child with Special Health Care Needs
CMC Child with Medical Complexity
CP Cerebral Palsy
CHD Congenital Heart Disease
16.2 General Considerations
16.2.1 Multidisciplinary
Management, the Medical
Home, and Care Coordination
Much as a requirement of care for CMC is good
communication between the physician and fam-
ily, it is also essential that effective communica-
tion exist between the numerous involved
specialists and the child’s principal provider.
Surgical specialists and generalists need to be
educated on successful means of interaction and
shared management of patients. Lack of useful
exchange between care providers leads to disor-
ganized, lower quality care and reduced satisfac-
tion of both specialist and generalist alike
(Russell and Simon 2014). When numerous spe-
cialists are caring for a CMC, a central provider
or team should serve as the child’s “medical
home” and oversee care coordination and deliv-
ery (Lerner and Klitzner 2017). Current models
of care for CMC can be grouped into 3 main cat-
egories: (1) primary care-centered models, (2)
consultative or co-management-centered models,
and (3) episode-based models (Pordes et  al.
2018). These care delivery models serve to
streamline care coordination and other services
to CMC and are an ongoing focus of health care
and policy initiatives. Indeed, enrolling CHSCN
in a care management program based on one of
these three models is associated with a significant
reduction in ED utilization, inpatient admissions,
and hospital length of stay when compared with
baseline expenditures (Weiss et  al. 2019).
Furthermore, while patients, families, and care-
givers clearly see the value in care coordination,
so to do health care delivery systems and pur-
chasers of care (i.e., insurance companies).
Setting up a care coordination plan for CMC cer-
C. M. Calkins
tainly takes significant time to initiate time, but in
the long run avoids duplication of efforts and pre-
vents delay in meeting the needs of a child
because of assumptions that “someone else will
handle it.” When preparing for surgical interven-
tion in a CMC patient, communication with the
physician or team overseeing the patients’ medi-
cal home is paramount and greatly improves the
quality of care.
16.2.2 Family-Centered Care
Families always play an important role in seeing
through health care directives delineated by pro-
viders. For CMC, these issues go well beyond
administering a course of antibiotics for an ear
infection. In a survey regarding CHSCN, 40%
indicated that the patients’ condition significantly
impacted the family financial situation, and over
13% of respondents stated they spend over
11 hours a week coordinating care for their child
(van Dyck et al. 2002). Furthermore, over 50% of
caregivers either cut back on working or stopped
working altogether to care for their CSHCN. The
American Academy of Pediatrics (AAP) has
rightly determined that family-centered care is
“based on the understanding that the family is the
child’s primary source of strength and support;
and the perspectives and information provided by
families, children, and young adults are impor-
tant in clinical decision making” (Schor 2003).
Care coordination and surgical decision-making
in these patients must be family-centered,
assessment-­driven, continuous, and team-based,
and designed to meet the bio-psychosocial needs
of children, youth, and their families while
enhancing caregiving skills and capabilities that
ensures the best possible surgical outcome
(Antonelli et al. 2008). For CHSCN, the experi-
ence of negotiating an operation in the setting of
a chronic condition is made more bearable by
partnership with a surgeon practicing family-­
centered care. When family-centered care is pro-
vided effectively before and after an operation,
all parties involved benefit—the child, the family,
and the surgeon.
16  Surgical Problems of Children with Physical Disabilities
16.2.3 Medical Complexity
and Polypharmacy
CHSCN are often treated for a host of disorders
with a variety of medicines. The most commonly
used definition of polypharmacy is being on five
or more medicines, and while all of these agents
may indeed be indicated and used with appropri-
ate indications, the perioperative state may alter
the ability for the team to administer the medica-
tions in a manner that proves effective.
Pharmacists have a unique opportunity to
improve health care for pediatric patients by par-
ticipating in the detection of polypharmacy
through clinical pharmacy practice and making
recommendations for appropriate parenteral sub-
stitutes for medications that may not be suitable
for use after surgery when the gastrointestinal
tract is manipulated. We recommend that CHSCN
with polypharmacy undergo medication recon-
ciliation with a member of the pharmacy team
when an operation is being planned that may alter
the normal administration of the patients’ medi-
cal therapy (Al-Jazairi et al. 2017). Traditionally,
medication reconciliation is the process of com-
paring a patient’s medication orders to all of the
medications that the patient has been taking and
is done to avoid medication errors such as omis-
sions, duplications, dosing errors, or drug inter-
actions, but can also be done to deliberately
formulate an appropriate perioperative medica-
tion plan. This should be considered a routine
part of pre-operative planning and should also be
undertaken by an inpatient pharmacist when a
CHSCN is undergoing an unplanned urgent or
emergent operation (Horace and Ahmed 2015).
16.2.4 Contemporary Pain
Management and
Pre-­operative Anesthesia
Evaluation
The underlying pathophysiology of pain in chil-
dren may include acute nociceptive pain, neuro-
pathic pain, psycho–social–spiritual–emotional
pain, and/or chronic persistent pain. For CHSCN
187
undergoing surgery, pain commonly arises from
a combination of these etiologies. The current
best practice of preventing and treating pain in
children involves using multimodal analgesia,
which may include analgesics and adjuvants,
procedural interventions, rehabilitation, psycho-
logical and integrative therapies that act synergis-
tically for more effective pediatric pain control
with fewer side effects than any single analgesic
or modality (Ferland et al. 2018; Friedrichsdorf
2017). Although it may not be possible to entirely
avoid opioid use after major operations, multi-
modal therapy most certainly decreases the needs
for opioids and consultation with a pain manage-
ment specialist benefits the patient and the health-
care team. When able, we recommend
pre-operative planning with a member of the hos-
pital pain team to optimize perioperative man-
agement of the CHSCN undergoing surgery.
While this may not be necessary for simple, out-
patient procedures, complex operations demand
a preemptive assessment of chronic pain and how
best to address the acute pain that will most cer-
tainly accompany a major operation.
In most centers, the pain team typically
includes anesthesiologists with specific expertise
in pain management. A pre-operative visit with
an anesthesiologist not only addresses the need
for implementing a perioperative pain manage-
ment plan, but also allows for a review of the
patients’ medication list which is important for
planning an appropriate anesthetic (Tobias 2018).
A pre-operative visit further informs decisions
regarding laboratory or investigative work-up
that may be needed prior to the procedure, pro-
vides the anesthesiologist an opportunity to meet
the patient and their parents or guardians and out-
line anesthetic risk. Some CHSCN may also have
conditions that may be exacerbated by traditional
nil per os (NPO) times, and the anesthesia team
may assist in developing a more appropriate pre-­
operative plan for perioperative fluid and nutri-
tion management plan. In that regard, many
CHSCN have a history of difficulty in obtaining
peripheral intravenous access and recognizing
this and making appropriate plans prior to a
scheduled operation will prevent unnecessary
188
delays in the operating room. Underlying bleed-
ing disorders, pseudocholinesterase deficiency,
sleep-disordered bleeding syndromes, and pre-­
existing airway abnormalities are also more com-
mon in CHSCN and are best addressed by the
anesthesia team well ahead of time instead of
meeting the patient in the pre-operative holding
area on the date of surgery. Although pre-­
operative anesthesia clinics are common in adult
centers where patients often have a host of
comorbid conditions, CHSCN are equally as
complex and greatly benefit from pre-operative
anesthesia planning.
16.2.5 Ethical Complexity
The principles upon which medical decisions are
made for CSHCN are the same as for any other
child: autonomy, beneficence, justice, nonmalefi-
cence, and fidelity must always be considered. It
is important for the surgeon to discuss her
approach to the treatment of CSHCN with the
entire treatment team and family (Okun 2010).
Agreement on the approach to management prior
to engaging in a discussion with the family is
often helpful to provide the best care, so as to not
to confuse the caregiver with differing opinions
from separate members of the health care team.
For CSHCN who are hopelessly ill or who
have a progressive degenerative neurologic dis-
ease, it is important to determine in advance if
parents want care to be limited. If parents request
Do Not Resuscitate (DNR) status for their child,
it is imperative to have them sign a DNR state-
ment in the medical record, both for clarity and
for legal protection of physicians. In discussing
limitations of care, a distinction should be made
between initiation of life support and prolonga-
tion of life support. A distinction should also be
drawn between DNR and palliative care. DNR
orders generally limit care in circumstances of
cardiopulmonary collapse. The ultimate goal of
treatment prior to such an event is still to prolong
life. Palliative care is the active, all-­encompassing
care of people to provide relief of and prevention
of pain and discomfort. Patients in a palliative
C. M. Calkins
care program may still be resuscitated in the
event of cardiopulmonary arrest. “Comfort mea-
sures only” status is reserved for the patient with
a terminal condition, and the goal of treatment is
no longer to prolong life but to prevent suffering
(i.e., no further diagnostic or therapeutic inter-
ventions) and must be accompanied by a DNR
status.
Some patients with an active DNR status
become candidates for surgical procedures that
may provide them with significant benefit, even
though the procedure may not change the natural
history of the underlying disease. The dilemma
regarding the CSHCN with a DNR order under-
going an operative procedure is that anesthesia
promotes some degree of hemodynamic instabil-
ity that may result in cardiopulmonary arrest, and
many routine anesthetic manipulations can be
considered resuscitative in nature (Fallat and
Deshpande 2004). In addition, cardiopulmonary
resuscitation (CPR) in the operating room carries
a very different medical prognosis than CPR
administered in other hospital areas. While it is
rare for a patient to survive to discharge after
CPR in the hospital ward, some 50–80% of
patients resuscitated in the OR return to their
prior level of functioning. In the OR, the event of
arrest is always witnessed, and the proximate
cause usually known, allowing rapid, effective
intervention which is directed toward the specific
cause of arrest. Also, causes of arrest in the OR
are often reversible effects of anesthesia or hem-
orrhage, and not usually due primarily to the
patient’s underlying disease. Thus, automatic
enforcement of DNR orders without discussion
and clarification may not adequately inform
patients or their authorized representatives about
the new risks associated with surgery and anes-
thesia and may lead to inappropriate periopera-
tive and anesthetic management. Pre-operative
discussion may result in the patient or surrogate
agreeing to suspend the DNR order during sur-
gery and the perioperative period, retaining the
original DNR order, or modifying the DNR order.
In urgent or emergent situations or when the
patient lacks decision-making capacity, the sur-
geon and anesthesiologist should still make an
16  Surgical Problems of Children with Physical Disabilities
attempt to discuss the situation with the patient’s
parent or designated surrogate and come to an
agreement about DNR status. Once a decision is
made, the surgeon must continue his or her
­leadership role in documenting and conveying
the patient’s advance directive and DNR status to
the members of the operating room team (i.e.,
during the pre-operative “time-out”) and if neces-
sary, finding an alternate team member to replace
an individual who has an ethical or professional
conflict with the patient’s advance directive
instructions.
One of the goals of providing care to children
is to advocate for them, and this should be an
essential component of the provision of surgical
care for CSHCN and their families (Goldson
et al. 2006). Advances in technology and in the
application of technology have improved the lon-
gevity of children with even the most severe dis-
abilities. Parent advocacy groups are quick to
point out that denying care to a child because of
the child’s disability is tantamount to discrimina-
tion and is, therefore, a denial of civil liberty.
Many of these same advocates believe that qual-
ity of life considerations should not be taken into
account when making medical decisions about
surgical interventions in CSHCN. However, phy-
sicians do not have an ethical obligation to insti-
tute treatment to try to change the course of a
condition in CSHCN with a hopeless prognosis if
the treatment would be futile in prolonging life.
Presentation of available data regarding the natu-
ral history, the risks of surgery for mortality and
morbidity, and the complications of surgery
should be made empathetically. If, in her experi-
ence, the surgeon considers a child not to be a
surgical candidate because of the futility of sur-
gery to prolong life, one does not have the obliga-
tion to perform surgery, and this may be seen as
advocating on behalf of the child. In this circum-
stance, it is important for the surgeon to explain
that the decision not to perform surgery is not due
to the child’s disability, but rather given the futil-
ity of surgery for the purpose of prolonging life.
For parents who continue to insist on surgery, a
second opinion from another specialist is recom-
mended, and an ethics team consult may prove
helpful in this circumstance.
189
16.2.6 Abuse and Neglect
The National Center on Child Abuse and Neglect
reported that children with special needs or dis-
abilities are at a greater risk of maltreatment and
abuse than their non-disabled peers (Lindberg
1999). This includes all forms of maltreatment:
physical abuse, sexual abuse, neglect, and verbal/
emotional abuse. While the estimated general
incidence of child abuse is between 6 and 14%,
experts estimate that children with special needs
are between 2 and 3.5 times more likely to expe-
rience abuse or maltreatment during their child-
hood than children without special needs
(Sullivan and Knutson 2000). Up to 15% of chil-
dren evaluated by Child Protective Services in
the United States have one or more disabilities.
Shockingly, maltreatment in disabled children
approached a prevalence rate of 31% and dis-
abled children are three times more likely to be
abused than children without special needs (Jones
et al. 2012). Additionally, up to 80% of intellec-
tually impaired children may be sexually abused
at some point during their lifetime. Known risk
factors for abuse in disabled children include
economic disadvantage, single parent homes, and
younger age groups. Although the overwhelming
majority of children with special health care
needs have caretakers who ferociously advocate
and care for them, there remain children within
this population who are abused by care takers
(Nowak 2015; Petska et al. 2017). Pediatric sur-
geons should be vigilant about screening for and
referring suspicious cases in this vulnerable pedi-
atric population.
16.3 Perioperative Organ System
Considerations
16.3.1 Neurologic
Two of the most common primary neurologic
conditions encountered in CHSCN are cerebral
palsy and spina bifida. Neurologic issues in these
populations are commonly pertinent to the pedi-
atric general surgeon as they may add complexity
to the operation to be performed. Cerebral palsy
190 C. M. Calkins

(CP) is an umbrella term covering a group of epileptic treatment. In general, oral doses of
non-progressive, but often changing, motor AEDs should be administered the morning of
impairment syndromes secondary to lesions or surgery with a small sip of water. If the duration
anomalies of the brain. Most cases of CP are of the surgical procedure exceeds the half-life of
caused by abnormal brain development in the the maintenance AED, some AEDs can be admin-
first trimester, not birth asphyxia or prematurity, istered intravenously. Serum drug levels may be
as was previously believed. Cerebral palsy is typ- significantly altered by anesthetics and by the
ically classified according to the pattern of motor physiologic changes resulting from surgery, thus
involvement of extremities (hemiplegia, one side; AED levels should be monitored closely in the
diplegia, lower extremities involved, upper perioperative period. All of these issues are best
extremities only mildly so; quadriplegia, all addressed during the pre-operative medication
extremities involved) and by the type of neuro- reconciliation and pre-operative anesthesia pro-
logic dysfunction (spastic, hypotonic, dystonic, cesses described previously.
athetotic, or a combination) (Velde et al. 2019). Some children with a seizure disorder have
Spina bifida (meningomyelocele) occurs as a had implantation of a vagal nerve stimulator
result of failure of the neural tube to close during (VNS). This device is generally implanted subcu-
the first trimester. The vertebral bodies are taneously in the upper chest and serves to prevent
splayed open, and the spinal cord is malformed. seizures by sending regular, mild pulses of elec-
The defect can occur at any vertebral level, and trical energy to the brain via the vagus nerve.
neurologic consequences depend on the pattern Electrocautery or radio-frequency ablation may
of innervation at and beyond the level of lesion. damage the generator; however, a VNS does not
Surgical closure of the spinal lesion is tradition- necessarily need to be deactivated or inhibited
ally undertaken on the first or second day of life, during surgery. Recommended maneuvers to
however, in select circumstances, the defect may minimize damage to the electrical circuitry from
be closed in-utero. Traditionally, 80–90% of electrocautery include positioning grounding
patients have hydrocephalus that requires place- pads so as to prevent current flow through the
ment of a ventricular drainage shunt. Although system and as far away from the VNS generator
pre-natal repair is associated with a decrease in as possible. Nevertheless, consultation with the
the need for a shunt postnatally (Adzick et  al. neurologist responsible for management of the
2011), many of these patients still require a ven- device is recommended prior to an elective surgi-
tricular drainage procedure and the general sur- cal procedure (Costa and Richman 2016).
geon should have some knowledge of this
procedure as he may be called upon to assist the 16.3.1.2 Spasticity, Body Habitus,
neurosurgeon in complex cases. and Scoliosis
For the surgeon, it becomes essential that patient
16.3.1.1 Seizure Disorders (Epilepsy) body habitus and its effect on the planned surgi-
Seizure disorders are common in CHSCN with cal approach be considered. For instance, in a
CP. Patients with a seizure disorder are often con- patient with severe lower extremity contractures,
sidered at high risk during procedures, based pri- a laparoscopic operation may not be feasible if
marily on the risk of seizures occurring during or body habitus does not allow for adequate extra-
shortly after the procedure, and, to a lesser corporeal clearance of laparoscopic instrumenta-
degree, on the potential interaction between tion. Alternatively, such an operation may be
drugs for the procedure and for seizure control. possible, but the port sites must be altered from
The surgeon must make an appropriate plan for their generally accustomed position. Although
the perioperative management of anti-epileptic treatment for spasticity may be initiated using
drugs (AEDs) that involves the continued use of oral medications, these drugs have limited use
medications right up to the point of a planned because of adverse effects. CHSCN with spastic-
operation and the provision of postoperative anti-­ ity are often treated with injections of botulinum
16  Surgical Problems of Children with Physical Disabilities
toxin (Albavera-Hernández et  al. 2009) or an
intrathecal baclofen pump (IBP) (Lake and Shah
2019). Patients with severe lower extremity spas-
ticity may benefit from remote pre-operative bot-
ulinum toxin injection to facilitate abdominal
access, and a discussion with the treating neurol-
ogist is warranted to determine the possible ben-
efit of this practice. Continuous infusion of
baclofen into the cerebrospinal fluid (CSF) con-
centrates the drug locally where it achieves its
therapeutic effect. An IBP is often implanted in
the subcutaneous or subfascial tissues of the
lower abdomen, and the location of this large
device must be taken into consideration when
planning an appropriate laparotomy incision,
stoma site localization, or port placement in lapa-
roscopic surgery. In addition, consultation with
the physician responsible for the management of
the IBP prior to elective surgery is wise to ensure
that the pump has adequate baclofen.
Surgical correction of scoliosis is one of the
more commonly performed elective surgical pro-
cedures in patients with CP. Although advances
in surgical instrumentation and technique have
allowed for more patients to be treated by poste-
rior fusion alone, the anterior approach for inter-
vertebral discectomy remains valuable in select
circumstances (Cloake and Gardner 2016). The
pediatric surgeon may be called upon to provide
exposure to the thorax or abdomen for the neces-
sary disc spaces to be addressed by the orthope-
dic surgeon when anterior discectomy is required
(Janik et  al. 1997). Pre-operative planning with
the orthopedic surgeon is essential to ensure that
the planned exposure will be adequate to allow
for optimal anterior release and ensure the
expected outcome. Thoracoscopic approaches to
anterior discectomy have also been performed
successfully (Norton et al. 2007). Children under-
going surgical correction of spinal deformities
may occasionally experience the superior mesen-
teric artery syndrome, with rates after corrective
spinal surgery reported between 0.5% and 2.4%.
Symptoms of superior mesenteric artery syn-
drome typically include nausea, bilious emesis,
abdominal pain, early satiety, and anorexia.
Initial treatment focuses on gastric decompres-
sion, maintaining euvolemia and electrolyte bal-
191
ance, and nutritional support. The duodenal
obstruction should be bypassed for feeding by a
nasoduodenal or gastroduodenal tube advanced
beyond the point of obstruction. Gastrostomy
feedings may be necessary after resolution of
gastric distention to prevent recurrence. Other
surgical intervention is not usually indicated,
although parenteral nutrition may also be helpful
(Lam et al. 2014).
16.3.1.3 Hydrocephalus Drainage
and Shunt Considerations
While cerebrospinal fluid (CSF) shunting proce-
dures have significantly lowered the morbidity
and mortality due to hydrocephalus, it has been
estimated that 40–50% of children will experi-
ence shunt failure within the first year after place-
ment. Ventriculoperitoneal (VP) shunt placement
remains the most common site for placement of
the ventricular drainage catheter due to the large
absorptive surface area of the peritoneal lining
and the ease/comfort of insertion. Alternative
drainage sites described include the atrium/supe-
rior vena cava (ventriculoatrial or ventriculove-
nous shunt), the pleural space (ventriculopleural
shunt), or gallbladder (ventriculocholecystic
shunt). Proximal and distal catheter occlusion,
disconnection, and infection are the most com-
mon reasons for shunt malfunction (Stone et al.
2013). Mechanical failures are related to either
improper functioning of the shunt or improper
placement of the device and include shunt
obstruction, fracture, or disconnection of the
device components and migration. Migration of
the shunt can occur from the site of its initial
placement into a position where it can no longer
effectively drain CSF. Cerebrospinal fluid malab-
sorption may lead to abnormal accumulation of
the fluid and may result in functional failure of
the shunt. Infectious complications typically
occur within 6  months of shunt insertion and
occur in up to 10% of patients in some series.
Most are due to inoculation with skin flora at the
time of surgery or seeding from sites of distal
infection. These and other complications may
prompt consultation by a pediatric surgeon.
The general surgeon would be wise to have a
firm understanding of shunt tubing location in a
192
patient with a VP shunt prior to elective or emer-
gent abdominal surgery. The presence of a VP
shunt is not a contraindication to the creation of a
pneumoperitoneum necessary for laparoscopy
(Fraser et al. 2009). Planning for an appropriate
incision or port placement is necessary to avoid
damaging indwelling shunt tubing. Furthermore,
in patients who have peritonitis from another
source neurosurgical consultation is recom-
mended to discuss the possibility for temporary
exteriorization. Abdominal conditions that may
render the peritoneal cavity unsuitable for perito-
neal tubing placement include a history of perito-
nitis, dense adhesions, ascites, or the need for
peritoneal dialysis. Laparoscopy can be utilized
to determine the genesis of distal shunt dysfunc-
tion, to lyze adhesions, or to disrupt an intraperi-
toneal CSF pseudocyst. In addition, some authors
have advocated for the use of laparoscopy rou-
tinely for initial shunt placement citing the
advantage of fewer distal shunt obstructions
(Naftel et  al. 2011). Bacterial peritonitis may
occur as a result of an infected CSF pseudocyst or
perforated viscus from an indwelling shunt. Early
recognition and appropriate antibiotic therapy
may avert major abdominal surgery in selected
cases, but typically requires urgent externaliza-
tion of the shunt.
When the peritoneal cavity is no longer suit-
able to accommodate the distal shunt tubing,
other locations may be utilized. Ventriculoatrial
(more appropriately, ventriculovenous) shunts
can be placed into the great venous system in a
percutaneous manner similar to that utilized for
central line placement. Complications from ven-
triculovenous shunts include tubing migration
with arrhythmia, shunt nephritis, pulmonary
embolism, atrial thrombus, and endocarditis.
Ventriculopleural shunts in children have been
used infrequently in the management of hydro-
cephalus but may be suitable for the older child.
This form of CSF diversion offers a safe and sim-
ple method of drainage and the catheter can be
safely placed percutaneously or with the aid of
thoracoscopy. A small asymptomatic pleural
effusion is typically visible on the chest radiogra-
phy that does not imply malfunction. However, a
symptomatic pleural effusion can occur, and for
C. M. Calkins
that reason it is not recommend it for infants and
toddlers unless there are no reasonable alterna-
tives. Ventriculocholecystic shunting affords a
viable alternative with reasonably good out-
comes. The distal catheter could be conceivably
placed into the gallbladder percutaneously using
retrograde cholangiography or by laparoscopy or
laparotomy. The historical practice of placing a
ventriculovesical shunt is discouraged as this
technique is considerably more difficult and car-
ries a high complication rate (urinary calculi and
electrolyte disturbances are common).
16.3.1.4 Neurologic Impairment
and the Acute Abdomen
The evaluation of abdominal concerns in the neu-
rologically impaired child is more difficult that in
typical children due to a variety of factors.
Patients may have absent or abnormal perception
of abdominal pain, results of diagnostic tests and
procedures may be misleading, the list of possi-
ble etiologies is greater, and kyphoscoliosis may
distort normal anatomical relationships. Three
factors unique to the CSHCN population, present
in many patients, are necessary to keep in mind
when evaluating such a patient: (1) The potential
of a VPS malfunction, infection, or complication
(see above). (2) Neurogenic bowel often results
in constipation and the potential for morbid com-
plications such as sigmoid volvulus. (3) The neu-
rogenic bladder predisposes to both pyelonephritis
and urolithiasis and, in patients who have under-
gone surgical bladder augmentation, to potential
perforation. Abdominal computed tomography
with intravenous contrast is recommended when
the diagnosis is elusive, as it confers the best
chance of securing a diagnosis.
16.3.2 Gastrointestinal
16.3.2.1 Nutrition
The usual causes of malnutrition in children with
CHSCN are inadequate caloric consumption due
to poor oral motor function (neurologic dyspha-
gia), increased metabolic demand due to a hyper-
kinetic movement disorder, food aversion as a
learned response to the discomfort of eating
16  Surgical Problems of Children with Physical Disabilities 193

caused by either gastroesophageal reflux or suggests aspiration. A gag reflex that is both dif-
chronic aspiration (or both), and rarely parental ficult to elicit and has little palatal movement is
neglect. Some patients have combinations of the associated with neurologic dysphagia and aspira-
above causes. The general definition of malnutri- tion. Alternatively, an extremely active gag reflex
tion is a chronic nutritional state resulting from with repeated retching may also be associated
protein-calorie consumption that is inadequate to with dysphagia and aspiration. Repeated pulmo-
meet metabolic needs for health maintenance and nary aspiration leads to chronic coughing, sleep-­
growth over time. Malnutrition is associated with disordered breathing, impaired clearance of
poor somatic growth, diminished bone mineral airway secretions, colonization of the respiratory
density, abnormal progression through puberty, tract by pathogenic bacteria, a high risk of pro-
limited societal participation, and poor health. gressive parenchymal lung damage, and potential
Patients with CP often have an altered body com- mortality.
position such that the fat mass is maintained Malnutrition and concern for aspiration com-
more centrally as compared with able-bodied monly result in surgical consultation to provide
peers (Tomoum et  al. 2010), and growth charts durable enteral access for the patient with a feed-
for adequately nourished children with CP are ing disorder. Videofluoroscopic investigation is
now available as standard weight for height per- helpful in defining the nature of the feeding dis-
centiles perform poorly to identify malnutrition order in order to propose an individualized treat-
in children with CP (Brooks et  al. 2011). ment plan. If weight gain is not achieved in an
Advanced methods for determining nutritional agreed upon time frame, then enteral access
status such as measurement of triceps skinfold should be considered. Parents generally accept
thickness and mid-arm circumference require the need for enteral access if they have exhausted
expertise, expensive measuring tools, availability all resources to improve the ability of the child to
of tables of age-­appropriate norms, substantial take oral feedings. Prior to consideration of sur-
time and are not especially practical. gically placed enteral access, an upper gastroin-
testinal (GI) series is commonly obtained to
16.3.2.2 Neurologic Dysphagia, ensure surgical candidates do not have an esoph-
Aspiration, and Feeding ageal stricture, hiatal hernia, malrotation, or other
Disorders GI anomaly that may alter the plan for a gastros-
The severity of neurologic dysphagia can be tomy. In addition, neurologically impaired chil-
roughly judged by the time it takes to feed a dren have a high prevalence of gastroesophageal
child. Children who require more than 60  min- reflux disease (GERD) compared to typical chil-
utes to be fed a typical meal and who are fed four dren. The etiology of GERD in this population
meals or more per day are rarely adequately may be due to lesions in the neuronal-anatomic
nourished. Calorie counts can be used to confirm swallowing center located in the medulla oblon-
the suspicion of inadequate nutrition. Typical gata leading to an altered vasovagal reflex.
findings associated with neurologic dysphagia Indeed, abnormalities in esophageal motility,
are hypotonic lips, poor lingual function, delayed gastric contraction, and delayed gastric emptying
swallow reflex, and poor esophageal peristalsis. occur much more frequently than in neurologi-
These neurologic abnormalities lead to slow oral-­ cally normal patients with GER (Lauriti et  al.
pharyngeal transit time. The fundamental cause 2018). Other issues germane to the particulars of
of aspiration in children with neurologic impair- enteral access and gastroesophageal reflux will
ment is inability to protect the airway. This is due most assuredly be covered elsewhere in more
to either glossopharyngeal dysfunction, immobil- detail in this textbook. However, there is persis-
ity resulting in inability to roll from supine to tent controversy the optimal treatment for the
side after vomiting, or both. A history of choking neurologically impaired child with GERD who
during meals, especially when drinking liquids requires enteral access.
194
16.3.2.3 Complications of Enteral
Access and Fundoplication
Complications most frequently encountered in
patients with feeding tubes include dislodge-
ment, tube obstruction or leakage, and wound
issues associated with the tube exit site. These are
common problems if CHSCN.  When a child
presents with dislodgement of a feeding tube, the
most important history to gather is the type of
tube and how long ago it was placed. This deter-
mines the level of urgency of tube re-insertion
and whether or not it is safe to do so. When a tube
is dislodged, the stoma should be investigated for
bleeding or obvious disruptions of the tract from
which it emanates. When tubes are dislodged
within six weeks of placement, providers must be
concerned that the stomach has fallen away from
the abdominal wall, making peritoneal contami-
nation from leaking gastric secretions a possibil-
ity. If the tube has been in place for more than
6  weeks, prompt placement of a Foley catheter
into the existing stoma will prevent rapid closure
of the site, allowing for replacement of the appro-
priate tube once it becomes available. It is impor-
tant to verify intraluminal position once the tube
has been replaced. This can be done through
simple physical examination—auscultation of air
in the stomach and aspiration of gastric contents.
If there is any concern about the tube re-­insertion,
we advocate the use of a simple contrast study to
verify tube position.
Tube obstruction is another common compli-
cation associated with feeding tubes, secondary
to kinking of the tube or solidification of formula
within the tube. Gentle flushing with warm water
usually is sufficient to reestablish patency and
use of carbonated beverages to accomplish the
same goal is commonplace in many institutions.
When flushing fails to clear the obstruction,
replacement of the feeding tube is indicated.
Leaking around a feeding tube causes skin irrita-
tion and breakdown, often leading to non-healing
wounds in an already fragile population. Tube
sites and stomas may enlarge over time, promot-
ing drainage of gastric or jejunal contents around
the tube. The answer is not to replace the tube
with a larger diameter tube, as this will only
widen the stoma site. Replacement of the tube
C. M. Calkins
with one of smaller diameter, and the application
of protective skin barrier allows skin constriction
around the tube and healing of the compromised
skin. The skin around a feeding tube can become
irritated due to leaking, tube anchors, and granu-
lomas. Cleansing and drying of the skin are
important in allowing the site to heal. Though
these issues rarely pose an emergency, persistent
cellulitis may require admission for skin care and
intravenous antibiotic therapy.
Occasionally, children with feeding tubes
present with signs and symptoms of bowel
obstruction. These children are at higher risk for
adhesive obstruction due to their past surgical
history; however, it is important to remember that
distal migration of balloons can occur, causing
obstruction of the pylorus and jejunal lumen. A
contrast study can further define the cause of a
child’s obstruction. Jejunostomy balloons should
be filled with the minimal amount of water to
allow for adequate tube placement to prevent this
complication.
Early complications of fundoplication include
pneumonia, wound infection or dehiscence,
obstructive wrap, esophageal or gastric perfora-
tion, bowel obstruction, pancreatitis, splenic
injury, and vagal nerve dysfunction. Late compli-
cations include gas bloat, delayed gastric empty-
ing, retching, dumping syndrome, fundoplication
disruption or migration, para-esophageal hernia,
bowel obstruction, esophageal stricture. These
complications will be discussed in further detail
in the section on gastroesophageal reflux disease.
16.3.2.4 Constipation and Defecation
Disorders
The prevalence of constipation and fecal inconti-
nence is quite high in children—regardless of the
presence of a disability. There is little data about
the prevalence of defecation disorders in CSHCN,
but patients with spina bifida, neurologic impair-
ment, and other congenital anomalies that affect
the gastrointestinal tract with subsequent disabil-
ity commonly have difficulties with fecal elimi-
nation. This may become problematic after major
surgery if it is not appropriately addressed in the
perioperative period. The cause of abnormal
­defecation is often multifactorial and related to
16  Surgical Problems of Children with Physical Disabilities
the underlying etiology for the child’s disability.
Failure of rectoanal coordination, paradoxical
anal contraction or insufficient relaxation of anal
sphincter during defecation, and impairment of
rectal sensation as well as slowing of colonic
transit all may play a role.
CHSCN with neurologic impairment with
intact bulbocavernosal and anocutaneous reflexes
are likely to achieve continence on a simple, con-
sistently timed, reflex-triggered bowel manage-
ment program. Oral agents are typical first-line
agents in patients with a host of disorders when
defecation is impaired. In the spina bifida popula-
tion, bowel regimens are myriad—but generally
begin quite early owing to the underlying spinal
cord anomalies in that disorder. Glycerin supposi-
tories when the child is younger than 3 years of
age results in a reasonably good success rate.
Bisacodyl enemas can be used every other day
beginning at 5 or 6 years of age to help establish a
bowel habit. Large-volume saline enemas, 20 mL
per kg up to 1 L once per day administered by a
non-latex delivery system are an option, effective
in two-thirds of patients with flaccid anal sphinc-
ters. In 1990, Malone and colleagues published a
preliminary report on the use of antegrade “ene-
mas” (more appropriately termed “washouts”)
administered by a catheter passed into the cecum
through an appendicostomy—what is now com-
monly referred to as a MACE (Malone antegrade
continence enema) (Malone et  al. 1990).
Commonly performed variations of that operation
to achieve the same antegrade provision of colonic
washouts include appendiceal disconnection with
reimplantation, appendicostomy with no antire-
flux component, tubularized cecal flap (neo-
Malone), cecostomy or appendico-­ cecostomy
with an indwelling low profile “button” device,
and placement of a Chait tube placed directly into
the cecum. It has generally not been deemed to be
developmentally appropriate to expect a younger
child to demonstrate the rigid discipline required
to achieve adequate results from an antegrade
washout program. However, there are situations
in which younger children may be suitable candi-
dates if parents are highly motivated and the
younger child can engage in an antegrade wash-
out program. The nuances of this procedure and
195
its complications are discussed elsewhere in this
textbook, but it is worth noting that a thorough
understanding of the CSHCN surgical history is
tantamount when planning for an abdominal
operation. A MACE conduit is often hidden
within the umbilicus and may be damaged when
creating a midline laparotomy incision or gaining
trans- or infra-umbilical access to the abdomen
for a laparoscopic procedure if not recognized.
Consideration must be made for the presence of a
Malone conduit (or a Mitrofanoff for bladder
catheterization) and an appropriate plan made for
abdominal access and caring for the blood supply
to the appendiceal vasculature when operating in
the abdomen for another reason. In patients with
pre-existing constipation and defecation disor-
ders, re-institution of the pre-­existing bowel regi-
men as soon as possible after surgery is
recommended to prevent fecal impaction.
16.3.3 Cardiovascular
16.3.3.1 Congenital Heart Disease
The spectrum of congenital heart disease varies
widely, but patients who ultimately require STAT
category 3–5 operations (Table 16.2) often qual-
Table 16.2  Society of thoracic surgery congenital heart
disease cardiac procedural risk categories 3–5
STAT
Category Operation (condition)
Category 3 Aortic coarctation repair
Konno Procedure (Subaortic stenosis)
HLHS (Hemi-Fontan Procedure)
Complete atrio-ventricular canal repair
Rastelli Procedure (Double outlet right
ventricle with pulmonary stenosis)
Pulmonary atresia repair
Category 4 Arterial Switch (Transposition of the
Great Arteries)
Heart transplantation
Ebstein’s anomaly repair
Interrupted aortic arch repair
Truncus arteriosus repair
Pulmonary artery banding
Category 5 Norwood Procedure (HLHS)
Damus–Kaye–Stansel procedure
(Multiple)
Hypoplastic Left Heart Syndrome (HLHS). Modified
from (Jacobs et al. 2013)
196
ify as being considered medically complex.
Indeed, infants with complex congenital heart
disease often require surgical or interventional
procedures in the first year of life, and 11% have
an associated genetic syndrome (Howell et  al.
2019). These patients commonly have a need for
non-cardiac surgical interventions and represent
a population that is inherently unique—espe-
cially with regard to perioperative management.
Despite mounting evidence that fetal, genetic,
immediate postnatal, and pre-operative factors
play a significant role in CHD associated brain
injury, intraoperative factors remain a prime area
of focus and patients with complex CHD lesions
should most definitely be cared for in a special-
ized center with anesthesia services that cater to
this population (Warnes et al. 2008).
Up to 40% of children with CHD require non-­
cardiac surgery by 5  years of age (Sulkowski
et al. 2014). The most common non-cardiac sur-
gical intervention in CSHCN with complex CHD
is the provision of durable feeding access—typi-
cally by way of a gastrostomy (Watkins et  al.
2013). This is most apparent in infants with CHD
that results in pulmonary over-circulation or in
those who have undergone the first stage of pal-
liation (Norwood) for hypoplastic left heart syn-
drome. These children often need 120 to 150 kcal/
kg per day for adequate growth, and many lack
the ability to achieve this goal by oral intake
alone due to a variety of factors including heart
failure, suck and swallow incoordination, postop-
erative vocal cord injury, and airway and struc-
tural or functional neurologic abnormalities.
Anesthesia induction and maintenance are of sig-
nificant concern due to the fact that vasoactive
drugs and positive pressure ventilation represent
a challenge to an already challenged cardiovascu-
lar physiology. In our view, these are operations
that are best reserved to be performed by the most
qualified pediatric surgeon possible. This is one
instance when allowing a neophyte trainee to
serve as the primary surgeon is not advisable, as
minimizing the time under anesthesia is impera-
tive. Although laparoscopic operations have
deemed to be “safe in patients with congenital
heart disease,” not all patients with CHD are
C. M. Calkins
equal, and one must consider the effects that a
pneumoperitoneum has on venous return in those
patients who are highly dependent on preload to
support adequate circulation. A deliberate discus-
sion with the anesthesia team and cardiac inten-
sivist is paramount to decide upon what operative
approach is safest for the patient. Furthermore,
many CHD patients receive anti-thrombotic med-
ications for the presence of a mechanical valve, a
systemic to pulmonary arterial shunt, and those at
risk for thrombosis due to abnormal flow patterns
(i.e., following a Fontan), and the pediatric sur-
geon must make an appropriate perioperative
plan to address perioperative anticoagulation.
Shunt thrombosis in the systemic to pulmonary
artery shunt can occur quite suddenly when
patients are not appropriately anticoagulated, and
this is a life-threatening event during non-cardiac
surgery (Giglia et  al. 2013). Complete shunt
occlusion must be immediately recognized and
managed emergently with anticoagulation, phar-
macologic treatment to increase systemic systolic
blood pressure, and controlled ventilation to
maximize oxygen delivery and minimize oxygen
consumption. Emergent cardiac catheterization
or emergent sternotomy for thrombectomy to
remove thrombus may be necessary and if these
maneuvers are not immediately successful, the
patient will likely require extracorporeal mem-
brane oxygenation.
Ultimately, in our view, any patient with cya-
notic CHD, a Fontan circulation, severe pulmo-
nary hypertension, cardiac insufficiency,
valvulopathy, or an underlying dysrhythmia that
requires intervention of any kind should be
referred to a specialized center with experienced
cardiologists, intensivists, surgical specialists,
and most importantly, specialized pediatric car-
diac anesthesiologists (Brown et al. 2020).
16.3.4 Respiratory
The most common cause of morbidity and mor-
tality in CSCHN with CP is respiratory related.
Early mortality is more common in people with
severe or profound intellectual disability, severe
16  Surgical Problems of Children with Physical Disabilities
motor impairment, epilepsy, spasticity. Patients
with CP, dysphagia, and poor gross motor func-
tion are considered the highest risk and surgical
interventions in these patients should be consid-
ered high risk for postoperative respiratory com-
promise. In addition, scoliosis contributes to
reduced gas exchange (hypoxemia and hypercap-
nia), increased resistive load to breathing, and
increased risk of upper airway obstruction
(UAO), atelectasis, and pneumonia caused by
unequal expansion of lungs. Pediatric respiratory
therapists should be considered a crucial compo-
nent of the perioperative care team when surgical
intervention is considered in these high-risk
patients. The aims of physiotherapy include pre-
vention of atelectasis, improvement in lung com-
pliance, maintenance of chest wall mobility,
airway secretion clearance, and improving cough
effectiveness. Sleep-disordered breathing and
UAO are also common in this population and evi-
dence for these problems should be sought and
considered when planning an operation in this
population (Marpole et al. 2020).
16.3.5 Integumentary
16.3.5.1 Latex Allergy
Twenty to 40% of patients with spina bifida are
allergic to latex. Other pediatric subpopulations
at particular risk include patients with atopia,
children undergoing a surgical procedure during
the neonatal period, and individuals who require
frequent surgical interventions (De Queiroz et al.
2009). Indeed, there is a relationship between the
number of surgical procedures and the incidence
of latex allergy, suggesting that intraoperative
latex exposure sensitizes patients. Although it is
possible that there is some predilection toward
the development of latex allergy that is intrinsic
to patients with spina bifida, many hospitals have
now shifted to an environment that is entirely
latex-free for all patients, and this may ultimately
affect the incidence of latex allergy (Goldberg
et al. 2016). This practice has been shown to pre-
vent latex sensitization as well as to ensure that
no allergic child has an anaphylactic reaction
197
while in the hospital. In years past, one signifi-
cant problem with first-generation latex-free
gloves was that surgeons found them to be pro-
hibitive for use due to limitations in tactile sensa-
tion. But today, latex-free gloves are widely
available which offer equal tactile sensation such
that discomfort should not be considered an
obstacle for using them.
16.3.5.2 Decubitus Ulcers
Body proportions change as children grow.
Through infancy and early childhood, the occiput
is the largest bony prominence and a site of
higher supine pressure than the sacrum. Although
the sacrum, buttock, and heels are the most prev-
alent locations for pressure injuries in adults, the
most common locations of pressure ulcers in
pediatric patients in a tertiary care hospital are
the ears and occiput. Comorbidities in the medi-
cally complex population impact body composi-
tion and among children with cerebral palsy,
those with greater functional impairment have
higher fat percentages and lower lean body mass
than children with less functional impairment.
Although the prevalence of decubitus ulcers in
the pediatric population is lower than that in the
adult population yet has been reported as high as
25% in vulnerable ICU patients (Sullivan and
Schoelles 2013). In addition, medical devices are
an extremely important consideration as they are
associated with 50% of pressure ulcers in pediat-
ric inpatients (i.e., tracheostomy tube sites, adhe-
sives for securing tubes) (Visscher et  al. 2013).
Thankfully the majority of pressure ulcers are
diagnosed during an early stage (partial thick-
ness), and rates have markedly decreased as
awareness of skin fragility has increased. One
must document the presence of pre-existing pres-
sure ulcers prior to surgical intervention and take
appropriate steps to address caring for such a
lesion in the perioperative period. The use of
nursing care “bundles” has been shown to dra-
matically reduce the incidence of hospital-­
acquired pressure ulcers which translates into
significant cost savings and the avoidance of
morbidity associated with treatment (Freundlich
2017).
198 C. M. Calkins

16.3.6 Genetic Syndromes dren qualify as medically complex. A worthy

CSHCN often have an underlying genetic syn-
drome, and although the specifics of all genetic
alterations are beyond the scope of this chapter, the
pediatric surgeon should have an understanding of
the common conditions as many of these children
have surgical needs. As a greater proportion of chil-
dren with once-lethal genetic conditions (Janvier
et al. 2016), a general understanding of general sur-
gical needs is mandatory and many of these chil-
resource of free information about a number of
common genetic syndromes for the pediatric sur-
geon is StatPearls (www.statpearls.com) which
contains thousands of peer-reviewed articles—
many of which cover the common genetic syn-
dromes that will be encountered by the practicing
pediatric surgeon. Table 16.3 contains some of the
more common syndromes in children with medical
complexity and associated surgical issues that may
be encountered (Morrison and Spence 2005).

Table 16.3 Common genetic syndromes of Children with Medical Complexity that often require surgical
intervention
Syndrome Genetic pattern Phenotype Surgical issues
Down Trisomy 21 Hypotonia, flat facies, slanted palpebral Endocardial cushion defect, Duodenal
fissures, mental deficiency, short neck, atresia, Hirschsprung disease, C1-2
hyperflexible joints subluxation, cryptorchidism, immune
dysfunction, feeding disorder
Patau Trisomy 13 Holoprosencephaly, severe mental Cardiac abnormalities, polydactyly
deficiency, narrow convex fingernails,
prominent rocker bottom feet, hypotonia
Edward Trisomy 18 Clenched hands with overlapping fingers, Ventricular or atrial septal defect
short sternum, low-arched dermal ridge,
feeble cry
Beckwith-­ Defect in Overgrowth—macroglossia, Omphalocele, increase risk of
Wiedemann 11p15.5 region hemihyperplasia, hyperinsulinism embryonal tumors (Wilms,
Neuroblastoma, Hepatoblastoma,
Rhabdomyosarcoma)
Cornelia de Nipped-B-­ Characteristic facial features: thin Gastrointestinal dysfunction, cleft lip
Lange homolog downturning upper lip, synophrys, and palate, cardiac septal defects
Autosomal microcephaly, small widely spaced teeth,
dominant oligodactyly, growth restriction, low
pitched cry, developmental delay
Turner 45 XO Short stature, lymphedema, webbed neck, Lymphatic malformation (cystic
horseshoe kidney, amenorrhea, attention hygroma), coarctation of aorta,
deficit hyperactivity disorder bicuspid aortic valve
Noonan Pleomorphic Widely set eyes, low set ears, short Pulmonic stenosis, cryptorchidism,
autosomal stature, intellectual and developmental feeding disorder
dominant delay
Klippel-Feil Mutation in Fusion of cervical vertebrae 2–7, short Spina bifida, scoliosis
GDF6 and GDF3 neck, low hair line, hearing impairment
Prader-Willi Deletion on Hypotonia, obesity, small hands and feet, Early feeding disorder,
paternal mental deficiency, late onset hyperphagia, cryptorchidism
chromosome behavioral disorders
15q11.2-q13
Angelman Deletion of Ataxia, speech deficits, psychomotor Feeding disorders, scoliosis
maternal delay, paroxysms of hand flapping and
chromosome laughter, seizures
15q11-13
16  Surgical Problems of Children with Physical Disabilities 199

16.4 Conclusions and Future Berry JG, Hall M, Cohen E, O'Neill M, Feudtner C (2015)
Ways to identify children with medical complexity
Directions and the importance of why. J Pediatr 167(2):229–237
Brooks J, Day S, Shavelle R, Strauss D (2011) Low
Children with disabilities often present a chal- weight, morbidity, and mortality in children with
lenge to the pediatric surgeon. Adequate pre-­ cerebral palsy: new clinical growth charts. Pediatrics
128(2):e299–e307
operative preparation and consultation with other Brown ML, Dinardo JA, Nasr VG (2020) Anesthesia
healthcare providers is tantamount to achieving in pediatric patients with congenital heart disease
the best possible surgical outcome. As these undergoing noncardiac surgery: defining the risk. J
young patients today are living longer as the Cardiothorac Vasc Anesth 34(2):470–478
Cloake T, Gardner A (2016) The management of scoliosis
result of improved medical and surgical care, it is in children with cerebral palsy: a review. J Spine Surg
important to address the transition from a pediat- 2(4):299–309
ric to an adult provider. Transitional care pro- Costa A, Richman DC (2016) Implantable devices: assess-
grams aim to address the medical, psychosocial, ment and perioperative management. Anesthesiol Clin
34(1):185–199
financial, and educational/vocational needs of De Queiroz M, Combet S, Bérard J, Pouyau A, Genest
adolescents and young adults who seek transfer H, Mouriquand P et  al (2009) Latex allergy in chil-
of care from a familiar, pediatric environment to dren: modalities and prevention. Paediatr Anaesth
the more imposing adult medical world. 19(4):313–319
van Dyck PC, McPherson M, Strickland BB, Nesseler K,
Unfortunately, effective transitional care has Blumberg SJ, Cynamon ML et al (2002) The national
eluded most surgical patients and providers, and survey of children with special health care needs.
large gaps in surgical care exist for adolescents Ambul Pediatr 2(1):29–37
transitioning into young adulthood. The future of Fallat ME, Deshpande JK (2004) Do-not-resuscitate
orders for pediatric patients who require anesthesia
caring for these patients demands that healthcare and surgery. Pediatrics 114(6):1686–1692
systems address the transition for caring for these Ferland CE, Vega E, Ingelmo PM (2018) Acute pain man-
complex patients (2011). agement in children: challenges and recent improve-
ments. Curr Opin Anaesthesiol 31(3):327–332
Fraser JD, Aguayo P, Sharp SW, Holcomb IG, Ostlie
DJ, St Peter SD (2009) The safety of laparoscopy in
References pediatric patients with ventriculoperitoneal shunts. J
Laparoendosc Adv Surg Tech A 19(5):675–678
Administration MaCHBotHRaS.  National Survey of Freundlich K (2017) Pressure injuries in medically com-
Children with Special Health Care Needs 2016. plex children: a review. Children (Basel) 4(4)
Available from: https://www.cdc.gov/nchs/slaits/ Friedrichsdorf SJ (2017) Multimodal pediatric pain man-
cshcn.htm#anchor_1551498339279 agement (part 2). Pain Manage 7(3):161–166
Adzick NS, Thom EA, Spong CY, Brock JW 3rd, Burrows Giglia TM, Massicotte MP, Tweddell JS, Barst RJ,
PK, Johnson MP et  al (2011) A randomized trial of Bauman M, Erickson CC et al (2013) Prevention and
prenatal versus postnatal repair of myelomeningocele. treatment of thrombosis in pediatric and congenital
N Engl J Med 364(11):993–1004 heart disease: a scientific statement from the American
Albavera-Hernández C, Rodríguez JM, Idrovo AJ (2009) Heart Association. Circulation 128(24):2622–2703
Safety of botulinum toxin type A among children with Goldberg H, Aharony S, Levy Y, Sivan B, Baniel J, Ben
spasticity secondary to cerebral palsy: a systematic MD (2016) Low prevalence of latex allergy in children
review of randomized clinical trials. Clin Rehabil with spinal dysraphism in non-latex-free environment.
23(5):394–407 J Pediatr Urol 12(1):52.e1–5
Al-Jazairi AS, Al-Suhaibani LK, Al-Mehizia RA, Goldson E, Louch G, Washington K, Scheu H (2006)
Al-Khani S, Lewis G, De Vol EB et al (2017) Impact Guidelines for the care of the child with special health
of a medication reconciliation program on cardiac care needs. Adv Pediatr 53(1):165–182
surgery patients. Asian Cardiovasc Thorac Ann Horace A, Ahmed F (2015) Polypharmacy in pediatric
25(9):579–585 patients and opportunities for pharmacists’ involve-
Allen J, Molloy E, McDonald D (2019) Severe neurologi- ment. Integr Pharm Res Pract 4:113
cal impairment: a review of the definition. Dev Med Howell HB, Zaccario M, Kazmi SH, Desai P, Sklamberg
Child Neurol FE, Mally P (2019) Neurodevelopmental outcomes of
Antonelli RC, Stille CJ, Antonelli DM (2008) Care coor- children with congenital heart disease: a review. Curr
dination for children and youth with special health care Probl Pediatr Adolesc Health Care 49(10):100685
needs: a descriptive, multisite study of activities, per- Jacobs ML, O'Brien SM, Jacobs JP, Mavroudis C, Lacour-­
sonnel costs, and outcomes. Pediatrics 122(1):e209-16 Gayet F, Pasquali SK et al (2013) An empirically based
200
tool for analyzing morbidity associated with opera-
tions for congenital heart disease. J Thorac Cardiovasc
Surg 145(4):1046–57.e1
Janik JS, Burrington JD, Janik JE, Wayne ER, Chang
JH, Rothenberg SS (1997) Anterior exposure of spi-
nal deformities and tumors: a 20-year experience. J
Pediatr Surg 32(6):852–859
Janvier A, Farlow B, Barrington K (2016) Cardiac surgery
for children with trisomies 13 and 18: Where are we
now?
Jones L, Bellis MA, Wood S, Hughes K, McCoy E,
Eckley L et al (2012) Prevalence and risk of violence
against children with disabilities: a systematic review
and meta-analysis of observational studies. Lancet
380(9845):899–907
Kuo DZ, Berry JG, Glader L, Morin MJ, Johaningsmeir S,
Gordon J (2016) Health services and health care needs
fulfilled by structured clinical programs for children
with medical complexity. J Pediatr 169:291-6.e1
Lake W, Shah H (2019) Intrathecal Baclofen infusion for
the treatment of movement disorders. Neurosurg Clin
N Am 30(2):203–209
Lam DJ, Lee JZ, Chua JH, Lee YT, Lim KB (2014)
Superior mesenteric artery syndrome following sur-
gery for adolescent idiopathic scoliosis: a case series,
review of the literature, and an algorithm for manage-
ment. J Pediatr Orthop B 23(4):312–318
Lauriti G, Lisi G, Lelli Chiesa P, Zani A, Pierro A (2018)
Gastroesophageal reflux in children with neurological
impairment: a systematic review and meta-analysis.
Pediatr Surg Int 34(11):1139–1149
Lerner CF, Klitzner TS (2017) The medical home at 50:
are children with medical complexity the key to prov-
ing its value? Acad Pediatr
Lindberg T (1999) The child’s right to health and treat-
ment. Med Confl Surviv 15(4):336–341
Malone PS, Ransley PG, Kiely EM (1990) Preliminary
report: the antegrade continence enema. Lancet
336(8725):1217–1218
Marpole R, Blackmore AM, Gibson N, Cooper MS,
Langdon K, Wilson AC (2020) Evaluation and man-
agement of respiratory illness in children with cerebral
palsy. Front Pediatr 8:333
Morrison PJ, Spence R (2005) Genetics for surgeons.
Remedica, London
Naftel RP, Argo JL, Shannon CN, Taylor TH, Tubbs RS,
Clements RH et  al (2011) Laparoscopic versus open
insertion of the peritoneal catheter in ventriculoperi-
toneal shunt placement: review of 810 consecutive
cases. J Neurosurg 115(1):151–158
Neff JM, Clifton H, Popalisky J, Zhou C (2015)
Stratification of children by medical complexity
15(2):191–196
Norton RP, Patel D, Kurd MF, Picetti GD, Vaccaro
AR (2007) The use of thoracoscopy in the man-
agement of adolescent idiopathic scoliosis. Spine
32(24):2777–2785
Nowak CB (2015) Recognition and prevention of child
abuse in the child with disability. Am J Med Genet C
Semin Med Genet 169(4):293–301
C. M. Calkins
Okun A (2010) Ethics for the pediatrician: children who
have special health-care needs: ethical issues. Pediatr
Rev 31(12):514–517
Perkins J, Agrawal R (2018) Protecting rights of children
with medical complexity in an era of spending reduc-
tion. Pediatrics 141(Suppl 3):S242–S2S9
Petska HW, Gordon JB, Jablonski D, Sheets LK (2017)
The intersection of medical child abuse and medical
complexity. Pediatr Clin N Am 64(1):253–264
Pordes E, Gordon J, Sanders LM, Cohen E (2018) Models
of care delivery for children with medical complexity.
Pediatrics 141(Supplement 3):S212–SS23
Russell CJ, Simon TD (2014) Care of children with medi-
cal complexity in the hospital setting. Pediatr Ann
43(7):e157–ee62
Schor EL (2003) Family pediatrics: report of the Task
Force on the Family. Pediatrics 111(6 Pt 2):1541–1571
Stille CJ, Antonelli RC (2004) Coordination of care for
children with special health care needs. Curr Opin
Pediatr 16(6):700–705
Stone JJ, Walker CT, Jacobson M, Phillips V, Silberstein
HJ (2013) Revision rate of pediatric ventriculoperi-
toneal shunts after 15 years. J Neurosurg Pediatr
11(1):15–19
Sulkowski JP, Cooper JN, McConnell PI, Pasquali SK,
Shah SS, Minneci PC et al (2014) Variability in non-
cardiac surgical procedures in children with congeni-
tal heart disease. J Pediatr Surg 49(11):1564–1569
Sullivan PM, Knutson JF (2000) Maltreatment and dis-
abilities: a population-based epidemiological study.
Child Abuse Negl 24(10):1257–1273
Sullivan N, Schoelles KM (2013) Preventing in-facility
pressure ulcers as a patient safety strategy: a system-
atic review. Ann Intern Med 158(5 Pt 2):410–416
Tobias JD (2018) Preoperative anesthesia evaluation.
Semin Pediatr Surg 27(2):67–74
Tomoum HY, Badawy NB, Hassan NE, Alian KM (2010)
Anthropometry and body composition analysis in chil-
dren with cerebral palsy. Clin Nutr 29(4):477–481
Velde M, Novak T, Badawi. (2019) Early diagnosis and
classification of cerebral palsy: an historical perspec-
tive and barriers to an early diagnosis. J Clin Med
8(10):1599
Visscher M, King A, Nie AM, Schaffer P, Taylor T, Pruitt
D et  al (2013) A quality-improvement collaborative
project to reduce pressure ulcers in PICUs. Pediatrics
131(6):e1950–e1960
Warnes CA, Williams RG, Bashore TM, Child JS,
Connolly HM, Dearani JA et  al (2008) ACC/AHA
2008 guidelines for the management of adults with
congenital heart disease: a report of the American
College of Cardiology/American Heart Association
Task Force on Practice Guidelines (Writing
Committee to Develop Guidelines on the Management
of Adults with Congenital Heart Disease). Developed
in Collaboration with the American Society
of Echocardiography, Heart Rhythm Society,
International Society for Adult Congenital Heart
Disease, Society for Cardiovascular Angiography and
16  Surgical Problems of Children with Physical Disabilities 201

Interventions, and Society of Thoracic Surgeons. J Am
Coll Cardiol 52(23):e143–e263
Watkins SC, McNew BS, Donahue BS (2013) Risks of
noncardiac operations and other procedures in chil-
dren with complex congenital heart disease. Ann
Thorac Surg 95(1):204–211
Weiss MA, Marchese S, Zhang L (2019) Effective care
management for children with special health care
needs in the era of value-based payment. Clin Pediatr
58(9):949–956
 Surgical Aspects of HIV Infection
in Children
17
Alastair J. W. Millar, Brian Eley, and Sharon Cox

17.1 Introduction and Historical antiretroviral therapy to prevent mother-to-child

Overview transmission, resulting in a substantial reduction
in new childhood HIV infections, but a growing
Paediatric human immunodeficiency virus (HIV) population of HIV-exposed but uninfected (HEU)
infection occurs predominantly in sub-Saharan infants and children (UNAIDS 2019). Despite
Africa (SSA). Approximately 1.55 million HIV-­ the advances in preventing paediatric HIV infec-
infected children under 15 years reside in SSA, tion, many HIV-infected and HEU infants and
representing 91% of the 1.7 million children liv- children can be expected to require surgical pro-
ing with HIV in the world (UNAIDS 2019). cedures for the following reasons:
More than 90% of all HIV-infected children
under 10  years of age acquire HIV through 1. Emergency procedures to deal with a non-­
mother-to-child transmission during the antenatal HIV-­related life-threatening condition.
period, at the time of delivery, or through breast- 2. Emergency procedures to deal with a HIV-­
feeding. The transmission risk can be substan- related complication or disease.
tially reduced by prevention of mother-to-child 3. Non-emergency procedures, where surgery is
transmission (PMTCT) intervention programmes required to assist in the diagnosis of a HIV-­
that include the administration of antiretroviral related condition.
therapy (ART) to all HIV-infected women during 4 . Elective surgery for routine childhood proce-
pregnancy, and throughout breastfeeding and dures.
lifelong antiretroviral prophylaxis, as well as to
their newborn infants during the first 6–12 weeks Most studies on the outcome of HIV-infected
of life. Substantial progress has been made in patients undergoing surgery have been in adults
extending the coverage of PMTCT intervention (Yii et al. 1995; Shelburne et al. 2005; Horberg
programmes. At the end of 2018, 84% of all preg- et  al. 2006; Madiba et  al. 2009). These studies
nant and breastfeeding women with HIV infec- have reported conflicting results, but several have
tion throughout the world were receiving suggested increased morbidity associated with
HIV infection, with little or no impact on mortal-
ity. In contrast, there is limited information on
A. J. W. Millar (*) · B. Eley · S. Cox
surgical presentations and outcomes in HIV-­
University of Cape Town and Red Cross War
Memorial Children’s Hospital, infected children. The initial recognition of the
Cape Town, South Africa entity of female infants and children with HIV
e-mail: alastair.millar@uct.ac.za; infection presenting with recto-vaginal fistulae
brian.eley@uct.ac.za; sharon.cox@uct.ac.za

© Springer Nature Switzerland AG 2023 203

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_17
204 A. J. W. Millar et al.

and the poor outcome of both, whether combined of the HIV-related immunodeficiency and may
with a defunctioning colostomy or not, prompted aid healing and lower the risk of post-operative
surgeons to look more closely at the surgical infection, thus improving the overall surgical
management of children with HIV (Hyde 1994; outcome.
Wiersma 2003). Soon, it was also recognised that
any surgery conducted on infected patients
resulted in poor healing and poor outcomes with 17.2 Incidence
breakdown of wounds, stomas, and any attempt
at gastrointestinal anastomosis. Almost all of Approximately 90% of HIV-infected children
these patients died because antiretroviral therapy will develop muco-cutaneous and soft tissue dis-
was not yet available. ease, which may be infectious or non-infectious
For HIV-infected children presenting with (Stefanaki et al. 2002) (Fig. 17.1). South African
life-threatening diseases for which surgery is studies have shown that 6% of HIV-infected chil-
urgently indicated, the emergent nature of these dren develop BCG complications a median of
events do not allow pre-operative correction of 34  days after commencing ART, and 1% will
co-morbidities, such as malnutrition or the man- develop disseminated BCG infection (Nuttall
agement of the immune deficiency with ART in et al. 2008; Hesseling et al. 2009). HIV-infected
children who are antiretroviral-naïve. However, children are at a higher risk for malignancy than
antimicrobial therapy may be started for those HIV-unexposed children, with tumours repre-
with co-infections and appropriate peri-operative senting 2% of AIDS-defining events.
antibiotic prophylaxis prescribed.
Surgical procedures are frequently undertaken
to assist in the diagnosis of an HIV-related pathol-
ogy or complication. Most of these procedures
allow time for pre-operative treatment of pre-­
morbid conditions and co-infections, and initia-
tion of co-trimoxazole prophylaxis and ART
(Stefanaki et  al. 2002; Shelburne et  al. 2005;
Madiba et al. 2009).
Lastly, there are increasing numbers of HIV-­
infected children who require routine, elective
surgical procedures. The health status and life
expectancy of these children should be taken
into account when considering the timing and
need for surgery. If surgery can be delayed, then
co-­morbidities such as malnutrition and co-­
infections may be treated, co-trimoxazole
prophylaxis commenced, and ART initiated.
Treatment with ART for a period of 2–3 months Fig. 17.1  Extensive suppurative cervical lymphadenitis
or longer, prior to the surgical procedure, will with multiple anterior chest wall and upper limb soft tis-
suppress viral replication, permit partial reversal sue abscesses
17  Surgical Aspects of HIV Infection in Children 205

17.3 Etiopathogenesis Respiratory disease in HIV-infected children can

involve either the upper or lower airway with
HIV infection causes severe immune dysfunc- implications for airway management during
tion, the hallmark being progressive loss and anaesthesia (Bosenberg 2007). Adenotonsillar
dysfunction of CD4+ T-cells (Doitsh et  al. enlargement can cause upper airway obstruction
2014). Although B-lymphocytes are not and difficult endotracheal intubation. There is an
infected by HIV, defects in humoral immunity increased incidence of chronic lung disease in
occur during the course of HIV infection, such HIV-infected children; this may impact on the
as loss of memory B-lymphocytes, attenuated anaesthetic and peri-operative management
immunoglobulin production, and loss of spe- (Leelanukrom and Pancharoen 2007). Lastly, the
cific antibody responses previously gained dur- increased risk of infection may lead to nosoco-
ing routine childhood immunisation. HIV mial pneumonia complicating the post-operative
infection also compromises innate immunity, course.
causing alterations to the function of phago- Tuberculosis (TB) co-infection is common in
cytes, monocytes, natural killer cells, and anti- HIV-infected children in low- and middle-income
gen-presenting cells (Eley 2003). The net effect countries. Drug–drug interactions between
of the HIV-mediated immune deficiency is to rifampicin-containing anti-TB drug regimens and
predispose the HIV-infected child to a wide- other therapies, including antiretroviral drugs,
range of viral, bacterial, fungal, and parasitic create pharmaceutical challenges that may impact
infections, as well as cancers, such as Kaposi on the selection of drugs including anaesthetic
sarcoma and lymphoma. agents and analgesics for children requiring
surgery.
Haematological manifestations of HIV include
17.4 Pathophysiology anaemia, neutropenia, lymphopenia, and thrombo-
cytopenia (Eley et  al. 2002; Calis et  al. 2008).
HIV-infected children have a greater risk of com- Anaemia has been repeatedly identified as a strong,
munity- and hospital-acquired bacterial infec- independent risk factor for HIV disease progres-
tions. These infections may be more severe, sion and death. Severe thrombocytopenia and
follow a protracted course, or result in a worse anaemia correlate with advanced disease and poor
outcome compared to their HIV-unexposed coun- prognosis. Thrombocytopenia can complicate sur-
terparts. Bacterial infection can involve any organ gical procedures by increasing bleeding and the
system or present as bloodstream infection with- need for blood products, leading to increased com-
out a clinical focus. Infections may be polymi- plications in the peri-operative period.
crobial or caused by drug-resistant pathogens, Gastro-intestinal dysfunction including diar-
which has implications for the selection of peri-­ rhoea, nausea and vomiting, dysphagia, or odyn-
operative prophylactic antibiotics (Zar 2004; ophagia causes significant morbidity among
George et al. 2009). HIV-infected children. Abdominal surgery may
Reduced pulmonary reserve and an increase in result in a post-operative ileus. This, together
pulmonary complications are of concern after with pre-existing intestinal dysfunction, can
major surgery (Zar 2004). HIV-infected children make post-operative fluid management and feed-
have a high incidence of pulmonary complica- ing challenging. Furthermore, odynophagia and
tions, which may be infective or non-infective. dysphagia may be secondary to lesions, which
206 A. J. W. Millar et al.

are friable and thus easily traumatised during air-
way instrumentation or insertion of a naso-enteric
tube causing bleeding or perforation.
Malnutrition is one of the most frequent and
severe complications of paediatric HIV infec-
tion, increasing morbidity and mortality.
Analysis of approximately 18,000 HIV-infected
children in five southern African countries
showed that at the start of ART, 50% were under-
weight, 66% stunted, and 16% wasted (Gsponer
et al. 2012). Malnutrition has been reported to be
an independent risk factor for an adverse surgi-
cal outcome.
HIV-infected children are at risk for the devel-
opment of metabolic complications, some of
which may be secondary to HIV infection or
ART.  The surgical and anaesthetic implications
of such complications must be considered in the
overall management of HIV-infected children,
including during the peri operative period
(Bosenberg 2007; Leelanukrom and Pancharoen
2007).
17.5 Pathology
Surgical problems associated with HIV infection
may be classified into four major categories
(Table 17.1):
1. Soft tissue or organ-specific infections requir-
ing drainage or debridement.
2. Gastrointestinal tract disease and complica-
tions.
3. Infections in the perineal area.
4. Malignancies.

Table 17.1  Presentation, aetiology/differential diagnosis, and indications for surgery in HIV-infected children
Clinical presentation Aetiology/differential diagnosis Surgical indications
Surgical infection Abscess Staphylococcus aureus, Drain pus
Necrotising fasciitis Streptococcus species Debride necrotic tissue
Bloodstream infection Candida species Obtain culture to direct antibiotic
Herpes simplex virus (HSV) therapy
Cytomegalovirus (CMV)
Tuberculosis (TB)
Molluscum contagiosum
Gram negative bacteria, e.g.,
Pseudomonas species
Oesophageal Oesophagitis Candida species Contrast swallow to identify
diseases Oesophageal stricture CMV strictures
HSV Endoscopy with biopsy for
Idiopathic ulcers (HIV-related) histology and culture
Malignancy, e.g., Kaposi
sarcoma
Non-HIV-related pathogens
Gastrooesophageal reflux
Intra-abdominal Gastrointestinal bleeding CMV Endoscopy to diagnose GI
problems Gastrointestinal TB bleeding
perforation Mycobacterium Surgery for perforation and
Gastrointestinal avium-intracelularae obstruction
obstruction Candida species
Malignancy, e.g., Non-­
Hodgkin’s lymphoma
Perineal disease Ano-cutaneous fistula CMV Colostomy for sepsis with
Condyloma Human Papilloma virus rectovaginal, rectourethral fistulae
Rectovaginal fistula Cryotherapy or laser for
Rectourethral fistula Condylomata
Malignancy Depend on site and Non-Hodgkin’s lymphoma Biopsy of mass
extent of disease Kaposi sarcoma Surgical excision
Leiomyosarcoma
17  Surgical Aspects of HIV Infection in Children 207

17.6 Diagnosis and Differential cutaneous manifestations, including ecthyma

Diagnosis gangrenosum and a papular rash, often in the
perineal area (Flores et al. 1993). Lethal mucor-
17.6.1 Infections mycosis infection can produce similar features
(Fig. 17.3a, b, c) (Mugambi et al. 2015).
The cutaneous manifestations of HIV are an indi- The principles of management of bacterial
cator of underlying immune status. Bacterial skin skin infection are the same as those for HIV-­
infections are often recurrent, in atypical sites or unexposed children including appropriate antibi-
due to atypical organisms (Prose 1991) otics, and surgical drainage or debridement when
(Fig.  17.2). The most common organisms caus- needed. A high index of suspicion for rapidly
ing skin infections are Staphylococcus aureus spreading infection or necrotising fasciitis must
and Streptococcus species. These usually present be maintained (Pijnenburg and Cotton 2001), as
as cellulitis, ecthyma, erysipelas, furunculosis the immunocompromised patient may not be able
(occasionally of disseminated nature), persistent to mount an effective systemic response, required
and recurrent folliculitis, and impetigo. to contain the infection.
Pyomyositis is also increasingly reported possi- Infection with Molluscum contagiosum
bly associated with an increased risk of (Fig.  17.4), a common viral infection due the
Staphylococcus aureus colonisation. Gram nega- DNA poxvirus is frequent and is identical in
tive organisms may also cause severe, deep-­ appearance to HIV unexposed patients but tends
seated skin infection in HIV-infected children. In to be more extensive and recurrent, usually occur-
particular, Pseudomonas species may produce ring on the face and neck (Prose 1992).

Fig. 17.2  Noma (Cancrum oris) polymicrobial

necrotising soft tissue infection of the face
208 A. J. W. Millar et al.

a b

Fig. 17.3  Mucormycosis: Acute rapidly spreading soft tissue cal resection revealed abdominal wall and bowel necrosis (b);
infection of the anterior abdominal wall with arterial throm- extensive invasion of branched hyphae of mucormycosis in
bosis of the femoral and inferior epigastric vessels (a); surgi- the blood vessels seen on histology of resected tissue (c)

Immunisation with live-attenuated Bacillus

Calmette-Guérin (BCG) vaccine soon after birth
is widely practiced in low- and middle-income
countries. This vaccine is typically administered
by intra-dermal injection on the right upper arm
at the insertion of the deltoid muscle.
Complications after immunisation are common
(Alexander and Rode 2007). In healthy immuno-
competent infants, a local reaction at the site of
inoculation and non-progressive ipsilateral axil-
lary lymphadenopathy may occur. In HIV-­
infected children, site ulceration or abscess
formation, suppurative ipsilateral axillary lymph-
adenopathy with or without fistulation (Fig. 17.5a
and b), and disseminated BCG infection may
develop, especially after ART is commenced due
to immune reconstitution inflammatory syn-
drome (Puthanakit et al. 2006). Vaccine site and
regional lymphadenopathy complications follow
a prolonged course but usually resolve spontane-
ously without specific intervention. Abscess
drainage or surgical excision may be required.
Fig. 17.4  Extensive molluscum contagiosum on the face
of an HIV positive child
17  Surgical Aspects of HIV Infection in Children 209

a b

Fig. 17.5  BCG’osis: Ulceration at the site of the BCG with surgically incised nodal abscess (a), and ulcerating nodal
inflammation requiring surgical curettage (b)

Disseminated BCG infection is treated with com- idiopathic infective ulceration. Due to an over-
bination anti-mycobacterial therapy (Hesseling lap of symptoms, endoscopy and biopsy are
et al. 2006; Nuttall et al. 2008). essential in identifying the pathology (Cooke
et al. 2009). The growing number of effective
antiviral and antifungal agents has mandated a
17.7 Gastrointestinal Tract more goal-directed approach to therapy.
Disease 2. Oesophageal strictures. The end result of
untreated or extensive ulceration is likely to
Esophagitis is a common problem in HIV-­ be stricture formation, occurring in
infected children that can cause prolonged dis- ­approximately 10% of patients (Cooke et al.
comfort and malnutrition, compromise adherence 2009). Strictures can be difficult to treat as
to ART and lead to increased morbidity and mor- they respond poorly to dilation and may
tality (Fantry 2003). Oesophageal symptoms require oesophageal replacement surgery
rank second only to diarrhoea in frequency of (Issa et al. 2004).
gastrointestinal complaints in HIV-infected chil-
dren (Fantry 2003). Oesophageal disease may be
a predictor of poor long-term prognosis, as it 17.8 Intra-abdominal Pathology
reflects severe underlying HIV immunodefi-
ciency. Opportunistic infections are the leading The diagnosis and management of children with
cause of oesophageal complaints. Treatment for intra-abdominal surgical pathology can be chal-
most aetiologies of esophagitis generally has a lenging. Localizing signs and symptoms are fre-
high degree of success, with a resultant improve- quently misleading due to underlying immune
ment in quality of life especially with ART suppression, debilitation, and antibiotic use.
(Cooke et al. 2009). Gastrointestinal (GI) bleeding, distension,
The differential diagnosis for oesophageal obstruction, perforation with abdominal pain,
disease includes: and tenderness are the most common clinical pre-
sentations associated with intra-abdominal dis-
1. Esophagitis or ulceration. The causes are usu- eases in HIV-infected children (Bowley et  al.
ally infective in origin, Candida species being 2007).
the most frequent infection, followed by cyto- Gastrointestinal bleeding represents an impor-
megalovirus (CMV), herpes simplex virus, and tant source of morbidity and can result from
210
opportunistic infections (CMV or Candida infec-
tion), HIV-associated malignancies (Kaposi’s
sarcoma, leiomyosarcoma or lymphoma), or may
be unrelated to HIV infection (Balderas and
Spechler 2006). Lower GI bleeding may be
caused by CMV colitis, tuberculosis (TB), malig-
nancy, or idiopathic colonic ulceration.
Aggressive investigation with endoscopy to find
the source of bleeding is required due to the wide
differential diagnosis (Zanolla et al. 2001).
Abdominal distension may develop secondary
to chronic diarrhoea, ileus, or obstruction. There
are many causes for obstruction including inflam-
matory and infective causes with TB predominat-
ing. Less commonly neoplastic obstruction
develops secondary to lymphoma or Kaposi sar-
coma. The obstruction may present due to inva-
sive bowel infiltration by the tumour or from
more localised disease and intussusception
(Cairncross et al. 2009). In neonates, the use of
ART may mimic functional bowel obstruction.
Gastrointestinal perforations may be second-
ary to CMV (Fig.  17.6), TB, or lymphoma.
Infants with chronic diarrhoea may develop peri-
tonitis due to intestinal perforation (Kahn 1997).
HIV-infected and HEU neonates with necrotising
enterocolitis may have a higher mortality and
develop more extensive disease than unexposed
infants (Karpelowsky et  al. 2011a). Likewise,
infants outside the neonatal period may develop
necrotising enterocolitis type pathology after
Fig. 17.6  Multiple ante-mesenteric small bowel perfora-
tions due to CMV in a child with HIV
A. J. W. Millar et al.
acute gastroenteritis with shock, which may
result in extensive small and large bowel
necrosis.
Abdominal pain as a presenting symptom may
be due to medical or surgical causes. Infective
and neoplastic conditions may present with
abdominal pain. Several unique problems arise,
however. Nucleoside reverse transcriptase inhibi-
tors used in ART regimens may cause abdominal
pain due to pancreatitis or lactic acidosis. Pain in
the HIV-infected child must be fully investigated
for medical, other drug-induced, or surgical
causes.
17.9 Perineal Disease
HIV-infected children have an increased rate of
peri-rectal abscess and ano-cutaneous fistula.
There are several reports of recto-vaginal fistula
and recto-urethral fistulae or multiple fistulae
with an increased rate of sepsis (Banieghbal and
Fonseca 1997; Kahn 1997; Wiersma 2003).
Management includes debridement and antibiot-
ics and on occasion stool diversion with proximal
divided colostomy. Definitive repair has been
reported to have poor results but may be success-
ful if preceded by initiation of ART.
Anal condylomata are rare in children but an
increased incidence occurs in HIV-infected chil-
dren, who may present with extensive and/or
recurrent lesions. Anal condylomata may or may
not be associated with sexual abuse in HIV-­
infected children. Most cases of anal condylo-
mata can be managed with cryotherapy,
electro-coagulation, or ideally CO2 laser ablation
under general anaesthesia, taking care not to
include the whole anal circumference to avoid
anal stricture. A staged approach may be indi-
cated if the lesions are very extensive (Johnson
et al. 1997).
17.10 Malignancy
HIV-infected children are at higher risk of malig-
nancy than HIV-unexposed children, with tumours
representing 2% of AIDS-defining events (Hadley
17  Surgical Aspects of HIV Infection in Children
Fig. 17.7  Mesenteric nodal involvement with Kaposi
sarcoma in a child presenting with intussusception due to
small bowel Kaposi sarcoma lesions
and Naude 2009). Kaposi sarcoma (Fig.  17.7),
non-Hodgkin’s lymphoma (NHL) and cervical
cancer are AIDS-defining events (Stefan et  al.
2011). Although the rates of leiomyosarcoma and
leiomyoma are higher in HIV-infected patients,
they are regarded as non-­AIDS-­defining tumours.
Kaposi sarcoma is the commonest malignancy in
HIV-infected individuals in Africa because human
herpes virus-8 infection is highly prevalent in
Africa and precedes its development. Non-
Hodgkin’s lymphoma is the second commonest
HIV-associated malignancy in Africa but the most
frequent HIV-­ associated malignancy in other
parts of the world. Ebstein-Barr virus infection
often precedes the development of NHL. A recent
South African study confirmed that Kaposi sar-
coma and NHL are the most frequently occurring
malignancies in HIV-infected children.
Furthermore, the risk of developing cancer was
lower on ART, but increased with age, and the
degree of immunodeficiency (Bohlius et al. 2016).
These malignancies are treated with a combina-
tion of ART, chemotherapy and radiotherapy.
Surgery is usually limited to a tissue biopsy or to
the treatment of complications.
Incidental solid tumours must be treated as for
HIV unexposed children, although coexisting
disease, specifically TB, should be considered
during the investigation of solid tumours for met-
astatic spread (Hadley and Naude 2009). This
group of co-infected patients has been reported to
211
have a high mortality independent of the primary
tumour type. A HIV-infected child with malig-
nancy should be started on ART as soon as pos-
sible. The widespread introduction of ART has
lengthened life expectancy, which may result in
an increased lifetime incidence of solid organ
malignancy in HIV-infected children (Biggar
et al. 2000; Caselli et al. 2000).
17.11 Management and Outcomes
Few prospective data exist on the outcomes of
HIV-infected children undergoing surgery
(Mattioli et al. 2009; Nelson et al. 2009). The first
reports of surgical outcomes in children with
HIV infection were brief, focusing on the proce-
dures performed and the risk of transmission of
HIV to health care workers. Subsequently, a few
case series reported HIV-specific surgical condi-
tions in children without describing the outcome
or complications of surgical intervention,
although a neonatal subgroup were reported to
have a high post-operative mortality of 30%.
Most of these children reported on did not receive
ART.
Several case reports or case series of proce-
dures have been published, raising concerns of
poor wound healing and complications
(Kleinhaus et al. 1985; Beaver et al. 1990; Nelson
et  al. 2009). A recent series of 48 HIV-infected
children undergoing minimally invasive surgery
(MIS) for diagnostic and therapeutic procedures
(Banieghbal 2009) concluded that MIS could be
safely performed on HIV-infected children, but
that certain routine procedures, such as fundopli-
cation were more difficult and prone to complica-
tion. The largest series of HIV-infected and
HIV-exposed children undergoing surgery was
published in 2009, but this report focused on the
disease presentations and only alluded to a higher
morbidity, furthermore no control group existed
in that study (Karpelowsky et al. 2009). Only one
prospective cohort study has been undertaken
comparing the outcomes of HIV-infected and
HIV unexposed children, which noted that HIV
infection was the most important risk factor for
development of a complication post-surgery,
212 A. J. W. Millar et al.

associated with an almost 12-fold higher risk. erative total white blood count, postoperative
There was also a significantly higher mortality absolute CD4 count, and postoperative plasma
and longer length of stay (Karpelowsky et  al. viral load were associated with mortality, postop-
2012). erative CD4 count was an independent predictor
of both postoperative infection and other compli-
cations, and a decrease in CD4 percentage was an
17.12 Factors Influencing Post-­ independent predictor of postoperative complica-
surgical Complications tions other than infection (Tran et al. 2000).
Only one paediatric study has assessed predic-
Several factors may impact on the rate of post-­ tors of post-operative complications in HIV-­
operative complications in HIV-infected children infected children undergoing surgery. Although
(Desfrere et al. 2005; Karpelowsky et al. 2012). this study was limited by sample size and a high
Antiretroviral therapy has reduced mortality and proportion of children with advanced HIV dis-
morbidity, improved the quality of life, and ease it found only age less than 1 year and major
slowed disease progression of HIV-infected chil- surgery were predictors of post-operative compli-
dren (Violari et al. 2008; Davies et al. 2009). It cations. Malnutrition, clinical stage of HIV infec-
suppresses viral replication leading to a reduced tion, ART, and type of surgery were not associated
plasma HIV viral load, reversal of CD4 cell attri- with post-operative complications (Karpelowsky
tion, improved overall immune function, and et al. 2011b).
lower risks for infection and HIV-associated
malignancy. Two adult studies have investigated
ART as an independent predictor of surgical 17.13 HIV Exposed But Uninfected
complications. In both studies, 3-drug ART regi- Children
mens were administered for at least 2–3 months
prior to surgery. Neither study found that ART A consequence of the success of PMTCT inter-
reduced post-operative complications. Similarly, vention programmes is that there are now an esti-
in paediatric studies, short-term peri-operative mated 14.8 million HEU infants and children in
ART was not found to be associated with a statis- the world, of whom 13.2 million reside in SSA
tically significant decreased risk of post-­operative (Slogrove et  al. 2020). Compared to HIV-­
complications. Thus, delaying elective surgery unexposed children, HEU infants and young chil-
for the institution of short-term ART may not dren are at an increased risk of morbidity and
seem sensible (Karpelowsky et  al. 2011b). mortality, primarily from infectious causes
However, whether or not a longer period of pre- (Slogrove et al. 2010; Karpelowsky et al. 2011a).
operative ART resulting in a greater degree of One study showed that HEU children are twice as
immune recovery will improve surgical outcomes likely to require hospitalisation for invasive pneu-
has not been subjected to rigorous scientific mococcal disease and those less than 6 months of
evaluation. age are less likely to survive an episode of inva-
A study that assessed the outcome of anorectal sive pneumococcal disease than HIV-unexposed
surgery in HIV-infected adults showed that only children (von Mollendorf et  al. 2015). Reasons
40% of patients had healed their wounds by for the increased risk of infection in HEU infants
3  months post-surgery. Wound healing was sig- are multi-factorial, including adverse social fac-
nificantly delayed when the absolute CD4 count tors associated with being born into an HIV-­
was <50 cells/μL. However, the clinical stage of affected household, often shortened duration of
HIV infection, age of the patient, and serum albu- breastfeeding, and higher risk of malnutrition. In
min concentration were not significant predictors addition, a range of immunological changes have
of wound healing (Lord 1997). In another study been reported in HEU infants predisposing them
that evaluated the outcome of HIV-infected adults to a spectrum of infections (Mofenson 2015).
requiring invasive or surgical procedures, preop- Susceptibility to infections may increase the risk
17  Surgical Aspects of HIV Infection in Children 213

for developing complications in the post-­ Balderas V, Spechler SJ (2006). Upper gastrointesti-
operative period. Poor growth and malnutrition nal bleeding in a patient with AIDS.  Nat Clin Pract
Gastroenterol Hepatol 3(6):349–353; quiz following
may also impact on outcomes post-surgery, 353
increasing the risk for infections and delayed Banieghbal B (2009) Minimally invasive surgery for chil-
wound healing. In one prospective study, it was dren with HIV/AIDS. J Laparoendosc Adv Surg Tech
noted that HEU children had a higher risk of A 19(1):97–101
Banieghbal B, Fonseca J (1997) Acquired rectovaginal
developing postoperative complications and mor- fistulae in South Africa. Arch Dis Child 77(1):94
tality than HIV-unexposed children, but this risk Beaver BL, Hill JL, Vachon DA, Moore VL, Hines SE,
was lower than in HIV-infected children Seiden SW, Stone M, Hutton N, Johnson JP (1990)
(Karpelowsky et al. 2011a). Surgical intervention in children with human immuno-
deficiency virus infection. J Pediatr Surg 25(1):79–82;
Post-operative pain control in HIV-infected discussion 82–74
children can be challenging. For adequate pain Biggar RJ, Frisch M, Goedert JJ (2000) Risk of cancer
control, a combination of medications may be in children with AIDS. AIDS-Cancer Match Registry
needed, increasing the potential for drug-drug Study Group. JAMA 284(2):205–209
Bohlius J, Maxwell N, Spoerri A, Wainwright R, Sawry
interactions in children who are often receiving S, Poole J et  al (2016) Incidence of AIDS-defining
ART or other drugs (Abuzaitoun and Hanson and other cancers in HIV-infected children on South
2000). Africa: record linkage study. Pediatr Infect Dis J
35(6):e164–e170
Bosenberg AT (2007) Pediatric anesthesia in developing
countries. Curr Opin Anaesthesiol 20(3):204–210
17.14 Conclusion Bowley DM, Rogers TN, Meyers T, Pitcher G (2007)
Surgeons are failing to recognize children with HIV
HIV-infected children may present with both infection. J Pediatr Surg 42(2):431–434
Cairncross LL, Davidson A, Millar AJ, Pillay K (2009)
conditions unique to HIV infection and common Kaposi sarcoma in children with HIV: a clinical series
surgical conditions. HIV exposure confers an from Red Cross Children’s Hospital. J Pediatr Surg
increased risk of complications and mortality for 44(2):373–376
all children following surgery, whether or not Calis JC, van Hensbroek MB, de Haan RJ, Moons P,
Brabin BJ, Bates I (2008) HIV-associated anemia in
they are HIV-infected. However, the risk of com- children: a systematic review from a global perspec-
plications is much higher in HIV-infected tive. AIDS 22(10):1099–1112
patients. These findings seem to be independent Caselli D, Klersy C, de Martino M, Gabiano C, Galli L,
of whether patients undergo an elective or emer- Tovo PA, Arico M (2000) Human immunodeficiency
virus-related cancer in children: incidence and treat-
gency procedure, but the risk of an adverse out- ment outcome-report of the Italian Register. J Clin
come is higher after a major surgical procedure. Oncol 18(22):3854–3861
Early treatment with ART is associated with Cooke ML, Goddard EA, Brown RA (2009) Endoscopy
reduced mortality in HIV-infected infants and findings in HIV-infected children from sub-Saharan
Africa. J Trop Pediatr 55(4):238–243
children. However, further studies are required to Davies MA, Keiser O, Technau K, Eley B, Rabie H, van
optimise the duration of ART prior to surgery. Cutsem G, Giddy J, Wood R, Boulle A, Egger M,
Moultrie H, A.  S. A.  C. International Epidemiologic
Databases to Evaluate (2009) Outcomes of the South
African National Antiretroviral Treatment Programme
References for children: the IeDEA Southern Africa collabora-
tion. S Afr Med J 99(10):730–737
Uncategorized References Desfrere L, de Oliveira I, Goffinet F, El Ayoubi M,
Firtion G, Bavoux F, Jarreau PH, Moriette G (2005)
Abuzaitoun OR, Hanson IC (2000) Organ-specific mani- Increased incidence of necrotizing enterocolitis in
festations of HIV disease in children. Pediatr Clin N premature infants born to HIV-positive mothers. AIDS
Am 47(1):109–125 19(14):1487–1493
Alexander A, Rode H (2007) Adverse reactions to the Doitsh G, Galloway NL, Geng X, Yang Z, Monroe KM,
Bacillus Calmette-Guerin vaccine in HIV-positive Zepeda O, Hunt PW, Hatano H, Sowinski S, Munoz-­
infants. J Pediatr Surg 42(3):549–552 Arias I, Greene WC (2014) Cell death by pyroptosis
214
drives CD4 T-cell depletion in HIV-1 infection. Nature
505(7484):509–514
Eley B (2003) The immunology of HIV infection. Curr
Allergy Clin Immunol 16:40–46
Eley BS, Sive AA, Shuttleworth M, Hussey GD (2002)
A prospective, cross-sectional study of anaemia and
peripheral iron status in antiretroviral naive, HIV-1
infected children in Cape Town, South Africa. BMC
Infect Dis 2:3
Fantry L (2003) Gastrointestinal infections in the immu-
nocompromised host. Curr Opin Gastroenterol
19(1):37–41
Flores G, Stavola JJ, Noel GJ (1993) Bacteremia due to
Pseudomonas aeruginosa in children with AIDS. Clin
Infect Dis 16(5):706–708
George R, Andronikou S, Theron S, du Plessis J, Hayes
M, Goussard P, Mapukata A, Gie R (2009) Pulmonary
infections in HIV-positive children. Pediatr Radiol
39(6):545–554
Gsponer T, Weigel R, Davies MA, Bolton C, Moultrie
H, Vaz P, Rabie H, Technau K, Ndirangu J, Eley B,
Garone D, Wellington M, Giddy J, Ehmer J, Egger M,
Keiser O, D. E. A. S. A. Ie (2012) Variability of growth
in children starting antiretroviral treatment in southern
Africa. Pediatrics 130(4):e966–e977
Hadley GP, Naude F (2009) Malignant solid tumour, HIV
infection and tuberculosis in children: an unholy triad.
Pediatr Surg Int 25(8):697–701
Hesseling AC, Rabie H, Marais BJ, Manders M, Lips
M, Schaaf HS, Gie RP, Cotton MF, van Helden PD,
Warren RM, Beyers N (2006) Bacille Calmette-Guerin
vaccine-induced disease in HIV-infected and HIV-­
uninfected children. Clin Infect Dis 42(4):548–558
Hesseling AC, Johnson LF, Jaspan H, Cotton MF,
Whitelaw A, Schaaf HS, Fine PE, Eley BS, Marais
BJ, Nuttall J, Beyers N, Godfrey-Faussett P (2009)
Disseminated bacille Calmette-Guerin disease in HIV-­
infected South African infants. Bull World Health
Organ 87(7):505–511
Horberg MA, Hurley LB, Klein DB, Follansbee SE,
Quesenberry C, Flamm JA, Green GM, Luu T (2006)
Surgical outcomes in human immunodeficiency virus-­
infected patients in the era of highly active antiretrovi-
ral therapy. Arch Surg 141(12):1238–1245
Hyde GS (1994) Acquired Rectovaginal fistula in HIV
positive Children. Pediatrics 6:940–941
Issa RA, Podbielski FJ, Fontaine JP, Connolly AE, Walsh
WV, Fraire AE (2004) Esophagectomy in a patient
with AIDS. Dis Esophagus 17(3):270–272
Johnson PJ, Mirzai TH, Bentz ML (1997) Carbon dioxide
laser ablation of anogenital condyloma acuminata in
pediatric patients. Ann Plast Surg 39(6):578–582
Kahn E (1997) Gastrointestinal manifestations in pediat-
ric AIDS. Pediatr Pathol Lab Med 17(2):171–208
Karpelowsky JS, Leva E, Kelley B, Numanoglu A, Rode
H, Millar AJ (2009) Outcomes of human immunode-
ficiency virus-infected and -exposed children under-
going surgery  – a prospective study. J Pediatr Surg
44(4):681–687
A. J. W. Millar et al.
Karpelowsky JS, Millar AJ, van der Graaf N, van
Bogerijen G, Zar HJ (2011a) Outcome of HIV-exposed
uninfected children undergoing surgery. BMC Pediatr
11:69
Karpelowsky JS, Zar HJ, van Bogerijen G, van der Graaf
N, Millar AJ (2011b) Predictors of postoperative com-
plications in HIV-infected children undergoing sur-
gery. J Pediatr Surg 46(4):674–678
Karpelowsky JS, Millar AJ, van der Graaf N, van Bogerijen
G, Zar HJ (2012) Comparison of in-hospital morbidity
and mortality in HIV-infected and uninfected children
after surgery. Pediatr Surg Int 28(10):1007–1014
Kleinhaus S, Weinberg G, Sheran M, Boley SJ (1985) The
management of surgery in infants and children with
the acquired immune deficiency syndrome. J Pediatr
Surg 20(5):497–498
Leelanukrom R, Pancharoen C (2007) Anesthesia in HIV-­
infected children. Paediatr Anaesth 17(6):509–519
Lord RV (1997) Anorectal surgery in patients infected
with human immunodeficiency virus: factors asso-
ciated with delayed wound healing. Ann Surg
226(1):92–99
Madiba TE, Muckart DJ, Thomson SR (2009) Human
immunodeficiency disease: how should it affect sur-
gical decision making? World J Surg 33(5):899–909
Mattioli G, Avanzini S, Pini-Prato A, Buffa P, Guida E,
Rapuzzi G, Torre M, Rossi V, Montobbio G, Rosati U,
Jasonni V (2009) Risk management in pediatric sur-
gery. Pediatr Surg Int 25(8):683–690
Mofenson LM (2015) Editorial commentary: new chal-
lenges in the elimination of pediatric HIV infection:
the expanding population of HIV-exposed but unin-
fected children. Clin Infect Dis 60(9):1357–1360
von Mollendorf C, von Gottberg A, Tempia S, Meiring
S, de Gouveia L, Quan V, Lengana S, Avenant T, du
Plessis N, Eley B, Finlayson H, Reubenson G, Moshe
M, O’Brien KL, Klugman KP, Whitney CG, Cohen
C, R.  Group for Enteric and A.  Meningeal Disease
Surveillance in South (2015) Increased risk for and
mortality from invasive pneumococcal disease in HIV-­
exposed but uninfected infants aged <1 year in South
Africa, 2009–2013. Clin Infect Dis 60(9):1346–1356
Mugambi MS, Theron A, Cox S, Pillay K, Millar AJW,
Numanoglu A (2015) Disseminated mucormycosis
and necrotizing fasciitis in immune-compromised
patients: two case reports. Ann Ped Surg 11:35–37
Nelson L, Fried M, Stewart K (2009) HIV-infected
patients: the risks of surgery. J Perioper Pract
19(1):24–30
Nuttall JJ, Davies MA, Hussey GD, Eley BS (2008)
Bacillus Calmette-Guerin (BCG) vaccine-induced
complications in children treated with highly active
antiretroviral therapy. Int J Infect Dis 12(6):e99–e105
Pijnenburg MW, Cotton MF (2001) Necrotising fasciitis in
an HIV-1-infected infant. S Afr Med J 91(6):500–501
Prose NS (1991) Cutaneous manifestations of HIV infec-
tion in children. Dermatol Clin 9(3):543–550
Prose NS (1992) Cutaneous manifestations of pediatric
HIV infection. Pediatr Dermatol 9(4):326–328
17  Surgical Aspects of HIV Infection in Children
Puthanakit T, Oberdorfer P, Ukarapol N, Akarathum N,
Punjaisee S, Sirisanthana T, Sirisanthana V (2006)
Immune reconstitution syndrome from nontubercu-
lous mycobacterial infection after initiation of antiret-
roviral therapy in children with HIV infection. Pediatr
Infect Dis J 25(7):645–648
Shelburne SA, Visnegarwala F, Darcourt J, Graviss
EA, Giordano TP, White AC Jr, Hamill RJ (2005)
Incidence and risk factors for immune reconstitution
inflammatory syndrome during highly active antiret-
roviral therapy. AIDS 19(4):399–406
Slogrove AL, Cotton MF, Esser MM (2010) Severe
infections in HIV-exposed uninfected infants: clini-
cal evidence of immunodeficiency. J Trop Pediatr
56(2):75–81
Slogrove AL, Powis KM, Johnson LF, Stover J, Mahy
M (2020) Estimates of the global population of chil-
dren who are HIV-exposed and uninfected, 2000–18:
a modelling study. Lancet Glob Health 8(1):e67–e75
Stefan DC, Wessels G, Poole J, Wainwright L, Stones D,
Johnston WT, Newton R (2011) Infection with human
immunodeficiency virus-1 (HIV) among children
with cancer in South Africa. Pediatr Blood Cancer
56(1):77–79
Stefanaki C, Stratigos AJ, Stratigos JD (2002) Skin
manifestations of HIV-1 infection in children. Clin
Dermatol 20(1):74–86
Tran HS, Moncure M, Tarnoff M, Goodman M, Puc MM,
Kroon D, Eydelman J, Ross SE (2000) Predictors of
operative outcome in patients with human immuno-
deficiency virus infection and acquired immunodefi-
ciency syndrome. Am J Surg 180(3):228–233
UNAIDS (2015) How AIDS changed everything. MDG6:
15 years, 15 kessons of hope from the AIDS response,
2015, from http://www.unaids.org/sites/default/files/
media_asset/MDG6Report_en.pdf
215
UNAIDS (2019) UNAIDS.  UNAIDS Data, 2020,
from http://rstesa.unaids.org/publications/
global-­publications/2019/item/208-­unaids-­data-­2019
Violari A, Cotton MF, Gibb DM, Babiker AG, Steyn
J, Madhi SA, Jean-Philippe P, McIntyre JA, C.  S.
Team (2008) Early antiretroviral therapy and mor-
tality among HIV-infected infants. N Engl J Med
359(21):2233–2244
Wiersma R (2003) HIV-positive African children with
rectal fistulae. J Pediatr Surg 38(1):62–64; discussion
62–64
World Health Organization (January 12, 2011)
Antiretroviral therapy of HIV infection in
infants and children: towards universal access.
Recommendations for a public health approach,
2010 revision. http://whqlibdoc.who.int/publica-
tions/2010/9789241599801_eng.pdf
World Health Organization (2013) Consolidated guide-
lines on the use of antiretroviral drugs for treating
and preventing HIV infection. Recommendations for
a public health approach, June 2013. http://apps.who.
int/iris/bitstream/10665/85321/1/9789241505727_
eng.pdf?ua=1
Yii MK, Saunder A, Scott DF (1995) Abdominal surgery
in HIV/AIDS patients: indications, operative man-
agement, pathology and outcome. Aust N Z J Surg
65(5):320–326
Zanolla G, Resener T, Knebel R, Verney Y (2001) Massive
lower gastrointestinal hemorrhage caused by CMV
disease as a presentation of HIV in an infant. Pediatr
Surg Int 17(1):65–67
Zar HJ (2004) Pneumonia in HIV-infected and HIV-­
uninfected children in developing countries: epidemi-
ology, clinical features, and management. Curr Opin
Pulm Med 10(3):176–182
 Part II
Trauma
 Birth Trauma
Thambipillai Sri Paran and Prem Puri
18.1 Introduction
Delivery room emergencies due to birth injuries
are serious, usually unexpected, and can be dis-
tressing situations that necessitate immediate
action to reduce neonatal morbidity and prevent
neonatal mortality (McKee-Garrett 2019). Every
paediatric surgeon must be familiar with the
management of birth trauma. In order to achieve
the best possible outcomes it is essential that a
paediatric surgeon, obstetrician, and neonatolo-
gist should participate in the evaluation and man-
agement of foetal and birth trauma.
Due to the intrinsic nature of the mechanical
forces involved and the bony boundaries of the
birth passage, the process of birth is naturally
traumatic for the infant. Even under optimal con-
ditions, injuries, such as greenstick fracture of
clavicle and subdural haemorrhages, are seen in
children born by normal spontaneous vaginal
delivery. Furthermore, subdural haematoma has
been documented in utero, even before the initia-
tion of labour. Added to this are the trauma
T. S. Paran
Children’s Health Ireland at Crumlin, Dublin, Ireland
and Trinity College Dublin and University College
Dublin, Dublin, Ireland
P. Puri (*)
Department of Pediatric Surgery, Beacon Hospital,
and University College Dublin, Dublin, Ireland
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_18
18
caused by the mechanical forces applied by the
obstetrician in delivering the baby. It is no sur-
prise that the exact cause of a particular trauma in
a newborn, with associated emotional and
medico-­legal sensitivities, may at times be far
from clear or certain.
The risk factors for birth injuries include mac-
rosomia, prematurity, instrumental delivery,
breech and other abnormal presentations, pro-
longed second stage of labour, and precipitous
delivery (McKee-Garrett 2019). The newborn at
the greatest risk of birth injury is the one in
breech presentation. These risk factors are well
recognised and efforts, such as decreased use of
forceps and more pre-emptive caesarean sections
and close monitoring of the foetus during labour,
have helped in reducing the incidence of birth
trauma over the years (Chaturvedi et  al. 2018).
Furthermore, the use of prenatal ultrasonography
has allowed early identification of the risk factors
for possible birth trauma, including foetal size
and position and enlarged foetal organs or
masses. However, as mentioned above, due to the
nature of labour itself, the overall incidence of
birth injuries still remains at over 1% in most
studies.
Birth trauma caused by intrinsic and applied
mechanical forces during labour are discussed in
this chapter, in order of anatomical location. We
have focussed mostly on the diagnosis and man-
agement of the most common mechanical inju-
ries reported during labour.
219
220
18.2 Head
Head trauma can be classified, into three catego-
ries, according to its location: extracranial, cra-
nial, or intracranial. Extracranial injuries include
skin lacerations secondary to blood sampling,
and haemorrhage in between various extracranial
anatomical layers secondary to rupturing of
blood vessels or bony fractures. Cranial injuries
are usually related to fractures of skull bones,
either linear or depressed. Most skull fractures in
neonates are reported to be linear, usually involv-
ing the parietal bones. However, Cho et  al.
recently reported that greenstick skull fractures
(GSFs) are more common than linear fractures,
but they are difficult to detect with simple skull
radiography (Cho et  al. 2018). Technological
advances in 3D CT have greatly enhanced the
diagnostic accuracy for skull fractures in neo-
nates. No specific treatment is required for linear
fractures; however, skull X-rays should be
repeated when the infant reaches 2–4 months of
age to exclude a ‘growing fracture of the skull’
associated with a leptomeningeal cyst.
Depressed skull fractures are most often
caused by pressure of the foetal head against the
maternal pelvis or delivery forceps. Spontaneous
resolution can be expected with conservative
treatment in neonates with no neurological signs.
Several nonsurgical techniques for the elevation
of depressed skull fracture in the newborn have
been described, including the use of either suc-
tion with a breast pump or vacuum extractor and
by digital manipulation. Surgical intervention
and elevation of the fracture should be consid-
ered when the depression is more than 2  cm,
there is presence of radiographic evidence of
bone fragments within the brain, there is neuro-
logical deficit, there are signs of increased intra-
cranial pressure, or there is failure to elevate the
fracture by closed manipulation (Hung et  al.
2005).
The most common intracranial injuries are
those where veins are ruptured within the sub-
dural space leading to subdural haematoma;
epidural and subarachnoid haemorrhages are
also seen, though less frequently. Subarachnoid
T. S. Paran and P. Puri
haemorrhage occurs when the veins of the sub-
arachnoid villi are torn, leading to collection of
blood in the subarachnoid space (Shah and
Wusthoff 2016).
18.2.1 Extracranial Haematomas
Extracranial haematomas can be classified
according to their anatomical location. These
include caput succedaneum, subgaleal haemor-
rhage and cephalhaematoma.
18.2.1.1 Caput Succedaneum
Caput succedaneum is due to the rupture of blood
vessels within the dense connective tissue under-
neath the skin and can be secondary to birth
trauma itself or due to vacuum or forceps deliv-
ery. Purpura and ecchymosis of the overlying
skin are common, and the suture lines do not
restrict the swelling. The bleeding is usually min-
imal due to the tamponading effect of the dense
connective tissue. These haematomas resolve
spontaneously within a few days.
18.2.1.2 Subgaleal Haemorrhage
Subgaleal haemorrhage is secondary to the rup-
ture of emissary veins within the loose connec-
tive tissue between the galea aponeurotica and
the periosteum. The bleeding can also sometimes
result from a bony fracture. It is seen in up to 6.4
per 1000 forceps deliveries, and the bleeding can
continue for 2–3  days and result in significant
loss of blood due to the lack of resistance by the
loose connective tissue. Other risk factors include
prolonged second stage of labour, foetal distress,
and macrosomia. It presents as a firm to fluctuant
mass that crosses the suture lines. It is estimated
that each 1 cm increase in the head circumference
correlates to 38  ml of blood loss. Treatment is
largely supportive including serial blood tests
and fluid replacement. A full blood count is nec-
essary to establish the extent of blood loss and a
coagulation screen is required following transfu-
sion of blood and large volume of fluids. Bilirubin
levels may increase due to the reabsorption of
large haematomas in neonates, and this needs to
18  Birth Trauma
be monitored closely. The overall mortality is
reported to be in the range of 14–22% when there
is excessive blood loss, and surgery may be
required in severe cases to cauterise the bleeding
vessels.
18.2.1.3 Cephalhaematoma
Cephalhaematoma is by far the most common
extracranial haematoma and is due to the rupture
of blood vessels beneath the periosteum. It is
seen in approximately 2.5% of newborns and is
associated with forceps and breech deliveries.
Repeated buffeting of the foetal skull against the
maternal pelvis during a prolonged labour, and
mechanical trauma caused by the use of forceps
and vacuum extractor in delivery, have been
implicated as important factors. Premature rup-
ture of the membranes was seen and was sug-
gested as an associated factor (Raines and
Whitten 2017). The swelling does not cross the
suture lines due to the periosteal attachments to
the bone, and purpura or ecchymosis is not a
feature of cephalhaematoma. A small percent-
age (5%) is associated with bony fractures.
Rapid expansion and excessive blood loss are
rare, but can occur if the periosteum ruptures.
Cephalhaematoma usually resolves over a few
weeks and no treatment is necessary. However, if
rapid expansion with signs of sepsis is noted,
infection of the haematoma should be suspected;
aspiration of fluid for diagnosis and treatment
may be necessary in such situations (Fig. 18.1).
Fig. 18.1  Large cephalhaematoma
221
18.2.2 Cranial Injuries
Cranial injuries usually consist of linear fractures
of parietal bones. As is the case with most birth
injuries, such fractures can occur even in uncom-
plicated spontaneous vaginal deliveries.
Additional risk factors include forceps delivery
and prolonged labour. Depressed fractures are
seen infrequently, and such depressed fractures
are mostly of the ‘ping-pong ball’ type that is due
to inward buckling of the resilient neonatal cal-
varial bones. Again, most of the cranial injuries
can be managed conservatively. In the absence of
neurological deficits or signs of raised intracra-
nial pressure, surgery is not indicated for the
treatment of depressed bony fractures. However,
nonsurgical interventions, such as the elevation
of the depressed segment with digital pressure or
a breast pump, have been utilised successfully in
the past. When there are signs and symptoms of
raised intracranial pressure such as apnoea, sei-
zures, focal neurological deficits, lethargy, and
hypotonia, neurosurgical intervention is neces-
sary to evacuate any haematoma and to elevate
the fractured segment.
18.2.3 Intracranial Haemorrhage
Intracranial haemorrhage is seen less frequently
than extracranial haemorrhage and the preva-
lence of symptomatic intracranial haemorrhage
in term infants is approximately 5–6 per 10,000
live births. Again, intracranial injuries have been
documented in infants born by normal vaginal
deliveries, but the incidence is higher with risk
factors such as forceps delivery, prolonged
labour, and macrosomia. The most frequent intra-
cranial haemorrhage is subdural haemorrhage
(75%); epidural haemorrhage is rare and sub-
arachnoid haemorrhage is usually associated
with preterm infants and asphyxia.
18.2.3.1 Subdural Haemorrhage
Subdural haemorrhage is seen in 8–10 per 10,000
forceps deliveries and the prevalence has
decreased over the years with the reduction in the
222
use of forceps. It is seen in 6% of all uncompli-
cated vaginal deliveries and has even been docu-
mented in some infants before the initiation of
labour. Subdural haemorrhage commonly results
from the tearing of the tentorial and interhemi-
spheric veins. Studies have shown that in asymp-
tomatic children, the haemorrhage resolves
spontaneously within 4 weeks and results in no
long-term developmental abnormalities.
Significant bleeding can lead to volume depletion
and raised intracranial pressure, and the infant
exhibits the following symptoms: apnoea, sei-
zures, focal neurological deficits, unequal pupils,
eye deviation, bulging fontanelle, lethargy, hypo-
tonia, drowsiness, and coma. Though these
symptoms are usually seen within 24 h of birth,
on occasions it may take up to a few days to fully
develop, as the bleeding can be slow and continu-
ous. Cranial ultrasound is useful in the diagnosis
of subdural haemorrhages, but is operator depen-
dent. The diagnosis is confirmed by a subdural
tap, CT scan, or MRI (Barnes 2001). Although
ultrasonography is a standard practice for detect-
ing germinal matrix haemorrhage in the preterm
neonate, it is less accurate then a CT or MRI scan
in diagnosing peripheral lesions in subarachnoid
or subdural space (Gupta et al. 2009). CT scan is
the diagnostic procedure of choice in suspected
cases, and MRI scan may be useful to gain added
information in selected cases. Most children with
subdural haemorrhage are asymptomatic and
could be treated conservatively following diag-
nostic imaging. However, a small percentage will
have symptoms; posterior fossa bleeds are more
symptomatic and may lead to brain-stem com-
pression. These symptomatic infants will require
surgical evacuation of their haematoma.
Progressive hydrocephalus is a well-known com-
plication of subdural haemorrhage in some, and
symptomatic and asymptomatic children must be
monitored with serial head circumference mea-
surements on a centile chart. When detected, ven-
triculoperitoneal shunt insertion is indicated to
drain the cerebrospinal fluid (CSF) into the peri-
toneal cavity. Though a majority of children with
subdural haemorrhage meet their developmental
milestones well, a significant proportion show
mild to severe developmental delay as they grow
T. S. Paran and P. Puri
older. When the bleeding is within the posterior
fossa, developmental delay is seen in up to 50%
of infants.
18.2.3.2 Subarachnoid Haemorrhage
Subarachnoid haemorrhage is caused by the rup-
ture of the bridging veins of the subarachnoid
space in preterm infants. Though seen in 1  in
10,000 spontaneous vaginal deliveries, the preva-
lence increases to 3 per 10,000 in forceps deliver-
ies. Symptoms are that of raised intracranial
pressure in an infant seen within the first 1–2 days
of life. Cranial ultrasound, CT scan, and CSF
sampling may all help in the diagnosis. As with
subdural haemorrhage, most subarachnoid haem-
orrhages also resolve spontaneously with no
long-term morbidity and do not need surgical
intervention. Progressive hydrocephalus is a
known complication of subarachnoid haemor-
rhage and should be monitored for in the recover-
ing infant. If seen, ventriculoperitoneal shunting
is necessary. Hypoxic injury to the brain tissue
must also be monitored for in those with large
subarachnoid bleed.
18.2.3.3 Epidural Haemorrhage
Epidural haemorrhage is rare and is usually asso-
ciated with cephalhaematoma or skull fracture. It
is usually due to a direct injury to the middle
meningeal artery, which is not protected within a
bony groove in the newborns, and can on occa-
sions be very large. Skull X-rays and CT scan are
useful in the diagnosis of this injury. Unlike those
with subdural haemorrhage, most infants with
epidural haemorrhage are symptomatic and will
need neurosurgical intervention. Aspiration of
the haematoma, without surgery, has also been
reported to be successful in the treatment of epi-
dural haemorrhage.
18.3 Injuries to Peripheral Nerves
These injuries are usually caused by excessive
traction or direct compression of nerves during
the delivery. The nerves most commonly involved
are the brachial plexus, facial nerve, and the
phrenic nerve. The outcome of treatment from
18  Birth Trauma 223

these injuries is usually good. This type of nerve malpresentation. It is believed to occur due to
injury was first classified into three categories by excessive downward traction on the head so as to
Seddon, and was later expanded into the follow- dislodge an impacted shoulder. However, bra-
ing five categories by Sunderland: chial plexus injury has been reported on the
opposite side, in the posterior shoulder, in up to
1. First-degree injury or neuropraxia—involves 40% of affected infants. It has also been reported
a temporary conduction block with demyelin- after normal uncomplicated vaginal deliveries,
ation of the nerve at the site of injury; com- and in infants born by elective caesarean sec-
plete recovery is normal and takes up to tions. The explanation for some of these injuries
12 weeks. may lie in the overall traumatic nature of birth
2. Second-degree injury or axonotmesis—severe and the significant mechanical forces involved
trauma causing proximal and distal axonal during normal labour and may not simply be
degeneration. The endoneurial tubes are intact attributable to the techniques employed by the
and the recovery is complete, with axonal midwife or obstetrician during labour. Brachial
regeneration occurring at a rate of 1 mm per plexus injury has been divided into three main
day. types depending on the site of the injury within
3. Third-degree injury—same as a second-­ the brachial plexus: Erb’s palsy, Klumpke’s palsy,
degree injury but more severe, with disruption and injury to the entire plexus.
to the endoneurial tubes. As with second
degree, the regeneration is complete, but the 18.3.1.1 Erb’s Palsy
regenerating axons may not reinnervate their The upper brachial plexus injury was first
original motor and sensory targets. described by Erb in 1874, and is by far the most
4. Fourth-degree injury—larger area of axon is common type (90%) of birth-related peripheral
damaged and this precludes any axons from nerve injury. The injury is to the C5-C6 nerve
advancing distally during regeneration. roots, resulting in the affected arm hanging
Surgery is necessary to restore neural limply adducted and internally rotated at the
continuity. shoulder, and extended and pronated at the elbow
5. Fifth-degree injury—complete transection of with flexed wrist and fingers in the typical ‘wait-
the nerve, and surgery is again necessary for er’s tip posture (Fig.  18.2). This is the result of
recovery of function. the paralysis of the deltoid, supraspinatus, infra-
spinatus, brachioradialis, and supinator brevis
muscles. On the affected side, the Moro, biceps,
18.3.1 Brachial Plexus Injury and radial reflexes are absent, while the grasp
reflex is preserved. Associated phrenic nerve
Brachial plexus injury could either affect the injury should be excluded.
entire plexus or some part of it. It is by far the Most neonates will make a complete or partial
most common injury and the incidence of bra- recovery on conservative treatment alone, as this
chial plexus birth palsy is estimated to be between type of nerve injury is mostly of the first and sec-
0.4 and 4 per 1000 live births (Foad et al. 2008; ond degree, according to Suderland’s scale.
Hale et al. 2010; Rehm et al. 2019). It has been Initial conservative treatment includes the immo-
reported that approximately 5000 neonatal bra- bilisation of the affected arm underneath the
chial plexus palsies occur every year in the USA, sleeve for 1 week. After 1 week of rest, the arm is
of which over 580–1050 are permanent (Chauhan put through passive range-of-motion exercises at
et al. 2014). The injury is usually caused by trac- the shoulder, elbow, and wrist to prevent contrac-
tion and stretching of the plexus 0.1–0.3% of all tures. The prognosis is excellent when antigrav-
live births. Risk factors include large birth weight ity muscle movements are seen within biceps and
(maternal diabetes), prolonged second stage of shoulder abductors by 3  months of age. If by
labour, forceps delivery, shoulder dystocia, and 3 months no movements are documented within
224
Fig. 18.2  Erb’s palsy with ‘waiter’s tip position of the
right arm
these muscle groups, then specialised neurosur-
gical opinion should be sought, in view to recon-
structive surgery (Abid 2016). The recovery is
somewhat guarded following surgical correction
in children with Erb’s palsy.
18.3.1.2 Klumpke’s Paralysis
Injury to lower plexus including C8-T1 nerve
roots was later described by Klumpke in 1885.
This lower plexus injury is uncommon. It results
in the paralysis of the intrinsic muscles of the
hand and flexors of the wrist and fingers. The
grasp reflex is absent. The same trauma that
causes lower plexus injury may also lead to the
injury of the cervical sympathetic fibres. Injury to
cervical sympathetic fibres may result in ipsilat-
eral Horner’s syndrome. Treatment is again con-
servative, but the recovery rates are poorer than
for Erb’s palsy. Again, surgical exploration must
be sought early, when no recovery is noted after
T. S. Paran and P. Puri
3–6 months of age. In all children considered for
surgery, electrodiagnostic studies and CT or MRI
scan must be carried out preoperatively. Advanced
microsurgical techniques, combined with nerve
grafting from the dural nerve, can improve the
functional outcome in some children.
18.3.1.3 Injury to Entire Brachial
Plexus
Injury to the entire brachial plexus results in flac-
cid arm with the absence of sweating, sensation,
and deep tendon reflexes. X-ray of the upper arm
and shoulder should be performed to exclude the
possibility of bony injuries. Children with injury
to the entire brachial plexus have severe nerve
disruption and get worse on conservative man-
agement. Early neurosurgical opinion and surgi-
cal exploration should be considered for these
children as well.
18.3.2 Facial Nerve Injury
Facial nerve injury is usually unilateral and is
secondary to the compression of the nerve against
stylomastoid foramen or the ramus of the man-
dible. The mechanism of injury is usually either
direct trauma from forceps or compression of the
side of the face and nerve against the maternal
sacral promontory (Rehm et al. 2019). It is seen
in approximately 0.06–0.7% of all live births.
Most injuries are simple swelling of the axons
secondary to the compression, and the disruption
of the axons is rarely seen. The affected side of
the face would have absent or decreased forehead
wrinkling, a persistently open eye, a decreased
nasolabial fold, and flattening of the corner of the
mouth. Spontaneous complete recovery within a
month is seen in >90% of infants (Reddy and
Redett 2015). Therefore, the initial treatment
consists of protecting the affected eye from dry-
ing out with the application of artificial tears,
until full recovery is documented. Because of the
high incidence of spontaneous recovery, surgery
should only be considered for those with no signs
of recovery after 1 year.
18  Birth Trauma
18.3.3 Phrenic Nerve Injury
Phrenic nerve injury arises from C3–C5 nerve
roots and is the motor supply to the ipsilateral
diaphragm. Injury to this nerve usually results
from excessive traction of the neck muscles and
leads to the paralysis of the ipsilateral diaphragm.
Most of these injuries (up to 75%) occur together
with brachial plexus injury, or fracture of the
clavicle, and are unilateral (Bowerson et al. 2010;
Shiohama et  al. 2013; Stramrood et  al. 2009).
The most common cause of phrenic nerve injury
is a difficult breech delivery. Clinically, the infant
may show respiratory distress and decreased air-­
entry on the affected side. Chest X-ray will show
a raised hemidiaphragm with mediastinal shift to
the opposite side (Fig.  18.3). Atelectasis of the
lower lobe of lung on the affected side with pneu-
monia may be present. Ultrasound may confirm
paralysis by demonstrating paradoxical move-
ment of the affected hemidiaphragm. Treatment
is again conservative with oxygen, physiother-
apy, and antibiotics when indicated. If the respi-
ratory distress is significant, continuous positive
Fig. 18.3  Elevated right diaphragm due to phrenic nerve
analysis
225
airway pressure ventilation should be considered.
Surgery is indicated when no improvement is
seen after 2  weeks of mechanical ventilation or
3  months of conservative treatment. Surgical
options include plication of the diaphragm or,
rarely, excision and artificial patch repair of the
diaphragm as employed in the repair of large
congenital diaphragmatic hernias.
18.3.4 Spinal Cord Injury
This is an extremely rare injury and has been esti-
mated to occur in 0.14 per 10,000 live birth
(Uhing 2005). Due to the incomplete mineralisa-
tion of the vertebrae and lax ligaments in a new-
born, the spine could potentially stretch more
than the spinal cord and thereby result in an
injury to the cord. Hyperextension of the neck in
utero has been associated with a 25% incidence
of cord transection when the child delivers vagi-
nally (Caird et  al. 2005). Damage to the upper
cervical cord is common and is associated with
the rotation of the head during vertex delivery
with forceps. Lower cord injury occurs during
breech delivery, when the head is trapped second-
ary to cephalopelvic disproportion. Injury above
the level of C4 will result in the paralysis of the
diaphragm, due to phrenic nerve injury and cause
apnoea. When the injury is below the level of C4,
the following signs and symptoms will be pres-
ent: absent spontaneous movement, absent ten-
don reflexes, and lack of response to painful
stimuli below the level of the injury. The bladder
will be atonic and distended, while the anal
sphincter will be atonic and patulent. The out-
come, in general, is very poor for these infants.
Early management should consist of strict immo-
bilisation of the head, neck, and spine.
Ultrasonography and MRI scan will reveal the
extent of injury and may help in determining the
prognosis. X-rays alone are usually not helpful.
Treatment should be supportive including venti-
lation and passive range of movements to prevent
pressure ulcers and pain relief. The overall out-
come in this group of children is poor, resulting
in early death, severe disability, and ventilator
dependency.
226
18.4 Abdominal Organ Injuries
Trauma to abdominal organs is uncommon dur-
ing birth. The organs most commonly affected
are the liver, spleen, adrenal gland, and kidney.
Risk factors include hepatosplenomegaly, breech
presentation, macrosomia, prematurity, and
coagulation disorders. Proposed mechanism of
injury to the liver and spleen include: direct
trauma by the rib cage secondary to thoracic wall
compression; thoracic wall compression causing
a pulling effect on the ligamentous attachments
to the liver and spleen with consequent tearing of
the parenchyma; and trauma secondary to
instrumental compression of the organs
­ DMSA or MAG3 scan may be necessary to
(Schullinger 1993; Uhing 2005). Hepatic trauma
more commonly results in subcapsular haemor-
rhage than actual rupture of the liver. The infant
with subcapsular haemorrhage usually appears to
be asymptomatic for the first few days of life
when the capsule is intact and haemorrhage lim-
ited by its boundary. Rupture of the capsule, fol-
lowed by extravasation of blood into the
peritoneal cavity, leads to sudden circulatory col-
lapse, abdominal distension, and a rapid drop in
the haematocrit value. Subcapsular bleeds are
common, but organ rupture with intra-abdominal
bleed is also seen. Clinical signs and symptoms
are that of an intra-­abdominal bleeding with pal-
lor, shock, abdominal distension, and abdominal
discoloration. When suspected, abdominal ultra-
sound, CT, and abdominal paracentesis could all
help in the diagnosis. Blood tests should include
full blood count, coagulation screen, and cross
match. Treatment is supportive with fluid replace-
ment and/or transfusion as indicated by the sever-
ity of the blood loss. When supportive treatment
fails or the infant becomes haemodynamically
unstable, explorative laparotomy must be under-
taken. In splenic injury, every attempt must be
made to preserve the spleen before considering
splenectomy to overcome the active bleeding.
Neonatal adrenal haemorrhage occurs most
commonly following a prolonged and difficult
labour, culminating in a traumatic delivery. Risk
factors for adrenal haemorrhage include pro-
longed labour, asphyxia, prematurity, septicae-
mia, renal vein thrombosis, and haemorrhagic
T. S. Paran and P. Puri
disease of the newborn. The right adrenal is
involved in over 70% of cases, with bilateral
involvement in 5–10% (Pond and Haber 1976).
The classical adrenal haemorrhage presents
within the first 4 days of life, as a flank mass with
fever and jaundice or anaemia. Ultrasound is use-
ful in the diagnosis, and a rim of suprarenal calci-
fication may be seen on abdominal radiograph
2–4  weeks later. Management is the same as
above with supportive transfusion followed by
laparotomy in severe cases.
Renal traumas are rare and are usually seen in
the background of congenital renal anomalies.
Investigations, such as renal ultrasound, CT scan,
assess the injury and any underlying congenital
abnormality, following supportive management
for the bleeding. Long-term follow-up may be
necessary as renal trauma could lead to scarring
of the renal parenchyma.
18.5 Fractures
18.5.1 Fracture of Clavicle
Fracture of the clavicle in the newborn usually
occurs during a difficult delivery associated with
large infants, breech presentation, and shoulder
dystocia (Hsu et al. 2002). This is the most fre-
quently fractured bone during birth and is seen in
0.3–3% of newborns. The vast difference in the
reported incidence is due to the fact that a major-
ity of these fractures go unnoticed due to lack of
external findings. A majority of these fractures
are of the greenstick type are not even recognised
at the time of discharge of the infant from hospi-
tal. Risk factors are the same as for most birth
traumas and are sometimes seen in uncompli-
cated normal deliveries as well. The most com-
mon finding is reduced movement of the affected
arm, but discolouration and bony deformity are
also seen infrequently. An X-ray will confirm the
diagnosis (Fig.  18.4). Treatment is usually con-
servative for incomplete fractures. When com-
plete fracture is noted, the arm should be
immobilised for 7–10 days. Differential diagno-
ses include brachial plexus injury, fracture of
18  Birth Trauma
Fig. 18.4  Fractured right clavicle
humerus, and shoulder dislocation. Erb’s palsy
and phrenic nerve injury must be excluded.
Recovery is complete in most infants.
18.5.2 Long Bone Fractures
Fractures of long bones are uncommon and are
usually associated with breech presentation, low
birth weight, and traction during delivery.
Swelling, pain, and decreased movement of the
limb may be present. An X-ray will confirm frac-
ture of long bones. Fractures of the humerus usu-
ally occur in the middle third of the shaft and are
either transverse or spiral. They are usually
greenstick fractures, but occasionally, complete
fracture with overriding of fragments may occur.
The most common mechanisms responsible for
fracture are believed to be traction on the
extended arm in the breech presentations and
axillary traction to disengage an impacted shoul-
der in vertex presentations. Treatment consists of
strapping the arm to the chest. Complete healing
of the fracture fragments usually occurs by
3 weeks.
Femur fracture is a rare condition with an inci-
dence of 0.13 per 1000 live births. Femur fracture
can occur in the context of excessive traction on
the femur; the most common manifestation being
spiral fractures involving femoral shaft
(Chaturvedi et al. 2018). However, a high index
of suspicion of epiphyseal injuries must be borne
227
in mind. These epiphyseal injuries will not show
up on plain X-rays, and when suspected ultraso-
nography for accurate diagnosis must be done.
Treatment consists of immobilisation and splint-
ing of the joint. For displaced fractures, closed
reduction and casting may be necessary. Proximal
femoral fractures need Pavlik harness or spica
cast for a period of time.
18.6 Conclusions
There is a wide spectrum of birth injuries from
minor to major problems. Birth injury is usually
associated with unusual compressive or traction
forces, in association with abnormal presentation
of the foetus. Paediatric surgeons, obstetricians,
and neonatologists must be familiar with these
conditions. Standard protocols for traumatic inju-
ries in pregnancy can help to decrease maternal
and neonatal mortality and morbidity.
References
Abid A (2016) Brachial plexus birth palsy: management
during the first year of life. Orthop Traumatol Surg
Res 102(1 Suppl):S125–S132
Barnes PD (2001) Neuroimaging and the timing of fetal
and neonatal brain injury. J Perinatol 21(1):44–60
Bowerson M, Nelson VS, Yang LJ (2010) Diaphragmatic
paralysis associated with neonatal brachial plexus
palsy. Pediatr Neurol 42(3):234–236
Caird MS, Reddy S, Ganley TJ, Drummond DS (2005)
Cervical spine fracture-dislocation birth injury: pre-
vention, recognition, and implications for the ortho-
paedic surgeon. J Pediatr Orthop 25(4):484–486
Chaturvedi A, Chaturvedi A, Stanescu AL et  al (2018)
Mechanical birth-related trauma to the neonate: an
imaging perspective. Insights Imaging 9(1):103–118
Cho SM, Kim HG, Yoon SH, Chang KH, Park MS, Park
YH et  al (2018) Reappraisal of neonatal greenstick
skull fractures caused by birth injuries: comparison
of 3-dimensional reconstructed computed tomogra-
phy and simple skull radiographs. World Neurosurg
109:e305–ee12
Chauhan SP, Blackwell SB, Ananth CV (2014 Jun)
Neonatal brachial plexus palsy: incidence, prevalence,
and temporal trends. Semin Perinatol 38(4):210–218
Foad SL, Mehlman CT, Ying J (2008) The epidemiology
of neonatal brachial plexus palsy in the United States.
J Bone Joint Surg Am 90(6):1258–1264
228
Gupta SN, Kechli AM, Kanamalla US (2009) Intracranial
hemorrhage in term newborns: management and out-
comes. Pediatr Neurol 40(1):1–12
Hale HB, Bae DS, Waters PM (2010) Current concepts in
the management of brachial plexus birth palsy. J Hand
Surg [Am] 35(2):322–331
Hung KL, Liao HT, Huang JS (2005) Rational manage-
ment of simple depressed skull fractures in infants. J
Neurosurg 103(1 Suppl):69–72
Hsu TY, Hung FC, Lu YJ, Ou CY, Roan CJ, Kung FT et al
(2002) Neonatal clavicular fracture: clinical analysis
of incidence, predisposing factors, diagnosis, and out-
come. Am J Perinatol 19(1):17–21
McKee-Garrett T (2019 Dec) Delivery room emergen-
cies due to birth injuries. Semin Fetal Neonatal Med
24(6):101047
Pond GD, Haber K (1976) Echography: a new approach
to the diagnosis of adrenal hemorrhage of the new-
born. J Can Assoc Radiol 27(1):40–44
Raines DA, Whitten RA (2017) Cephalhematoma.
Treasure Island: StatPearls. https://​www.​ncbi.​nlm.​nih.​
gov/​pubmed/​29262234
T. S. Paran and P. Puri
Reddy S, Redett R (2015) Facial paralysis in children.
Facial Plast Surg 31(2):117–122
Rehm A, Promod P, Ogilvy-Stuart A (2019) Obstetric
neonatal brachial plexus and facial nerve injuries: a 17
years single tertiary maternity hospital experience. Eur
J Obstet Gynecol Reprod Biol 243:57–62
Shah NA, Wusthoff CJ (2016) Intracranial hemorrhage in
the neonate. Neonatal Netw 35(2):67–71
Shiohama T, Fujii K, Hayashi M, Hishiki T, Suyama M,
Mizuochi H et al (2013) Phrenic nerve palsy associ-
ated with birth trauma – case reports and a literature
review. Brain Dev 35(4):363–366
Schullinger JN (1993) Birth trauma. Pediatr Clin N Am
40(6):1351–1358
Stramrood CA, Blok CA, van der Zee DC, Gerards LJ
(2009) Neonatal phrenic nerve injury due to traumatic
delivery. J Perinat Med 37(3):293–296
Uhing MR (2005) Management of birth injuries. Clin
Perinatol 32:19–38
 Pediatric Thoracic Trauma
David E. Sawaya, Michael W. Morris,
and Paul M. Colombani
19.1 Introduction
When treating pediatric trauma patients, many
important differences must be considered when
compared to adults. It may be difficult to estab-
lish emergency vascular access, especially in
cases of shock or cardiac arrest. Interosseous
access is an important option in all children and
is more commonly implemented in the very
young. Children have a smaller blood volume
and a more insidious onset of hemorrhagic shock,
commonly with tachycardia, as the only early
warning sign.
Regarding thoracic injuries, children have a
markedly compliant thorax making them vulner-
able to intrathoracic injury without overlying
bony injury. It is crucial to be vigilant while eval-
uating children with blunt trauma injuries. The
pediatric mediastinum is also very mobile, allow-
ing it to be shifted with much less intrathoracic
pressure. Susceptibility to tension pneumothorax
or hemothorax is much greater, and these injuries
must be treated promptly. This mobility, how-
ever, does make it less susceptible to major vas-
D. E. Sawaya (*) · M. W. Morris
University of Mississippi Medical Center,
Jackson, MS, USA
e-mail: dsawaya@umc.edu; mmorrisjr@umc.edu
P. M. Colombani
The Johns Hopkins Hospital Professor Emeritus,
Baltimore, MD, USA
e-mail: pc@jhmi.edu
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_19
19
cular or airway injuries. Thoracic trauma in
children is a marker for the presence of associ-
ated injuries, found in more than 50% of these
children.
Children are much less likely to have concom-
itant systemic illnesses compromising their respi-
ratory and cardiovascular reserves. This allows
lower morbidity and mortality rates with aggres-
sive medical therapy and faster recovery from
injury.
19.2 Diagnosis
In order to accurately diagnose pediatric thoracic
injuries, obtaining a thorough history of the trau-
matic event, including the mechanism of injury, is
critical. Blunt trauma accounts for 85–90% of
pediatric thoracic injuries. The majority of blunt
thoracic injuries result from motor vehicle
crashes. Children may be passengers in the motor
vehicle or, frequently, may be struck by the motor
vehicle. Key elements in the history for diagnostic
purposes include time of the accident, vehicle
speed, information regarding the vehicle impact,
damage to the vehicle, whether extrication
required, location of the child inside or outside of
the vehicle, restraint system used for the child,
and physical findings noted by the emergency
transport team. Falls from various heights are the
next most common cause of thoracic injuries. Key
elements in the history include the height from
229
230 D. E. Sawaya et al.

which the child fell, surface onto which the child Open pneumothorax
fell, and physical findings noted by the emergency Flail chest
transport team. A coordinated trauma team should Hemothorax
be assessing the airway, breathing, and circulation Cardiac tamponade and commotio cordis
Group II: Potentially life-threatening injuries found
of the pediatric trauma patient (primary trauma
during secondary survey
survey) while obtaining the history, followed by a Tracheobronchial injury
thorough physical exam (secondary trauma sur- Pulmonary contusions
vey). Appropriate imaging is obtained based on Myocardial contusions
the information and findings obtained from the Diaphragmatic injuries
history and examination. All pediatric patients Esophageal rupture
suffering thoracic trauma should have an immedi- Great vessel injury
Group III: Non-life-threatening injuries often found on
ate chest x-ray (CXR). For many critically injured physical exam or chest radiograph
children, computer tomography (CT) of the chest, Simple pneumothorax
and potentially the abdomen, pelvis, and head Small hemothorax
may be required. CT scans are typically obtained Rib fractures
with intravenous contrast and may be further Chest wall laceration
enhanced by obtaining arteriography (CTA) as Traumatic asphyxia (Perthes syndrome)
part of the CT scan. Isolated blunt thoracic trauma
carries a 5% mortality rate. When associated with
blunt traumatic brain injuries, the mortality rate 19.4 Injury Management
rises to 25%. When associated with brain and
abdominal injuries, the mortality rate rises to Immediate intervention for critical thoracic injuries
40%. Rapid diagnosis and intervention are para- cannot be overemphasized. The golden hour fol-
mount for optimal outcomes. lowing traumatic injury represents the highest like-
Penetrating trauma represents only 10–15% of lihood of patient survival with prompt intervention
thoracic trauma but the incidence increases with by the trauma team. This intervention may take
age, especially for children over 12 years of age. place at the scene of the trauma, during transport, in
Nearly 14% of children suffering from penetrat- the emergency department, in the operating suites
ing thoracic trauma die from thoracic injury. or a combination of these locations. Management
Associated injuries with penetrating thoracic of these individual pediatric traumatic injuries will
trauma are less common. For diagnostic pur- be covered in their respective sections.
poses, a quick but thorough history is vital while
performing the same initial trauma evaluation
ensuring that the airway, breathing, and circula- 19.5 Immediately
tion are intact followed by a complete exam. Life-­Threatening Injuries
Imaging includes an initial CXR, which will Found During Primary
guide further imaging. Survey

19.5.1 Airway Obstruction

19.3 Differential Diagnosis
Airway obstruction can prove to be fatal if it is
The differential diagnosis of pediatric thoracic not treated immediately. Securing the airway is
injuries can be divided into three groups based on the first task in patient management. This takes
the severity: precedence over all other needs and requires
Group I: Immediately life-threatening injuries found immediate attention and redirection if a patient’s
during the primary survey airway is lost during evaluation. Complete air-
Airway obstruction way obstruction presents with a total absence of
Tension pneumothorax air movement and breath sounds.
19  Pediatric Thoracic Trauma
Partial obstruction presents with inspiratory stri-
dor if the obstruction is above the vocal cords or
expiratory stridor if the level of obstruction is
below the vocal cords. Other signs and symptoms
include agitation, diaphoresis, chest wall retrac-
tions, cyanosis, and bradycardia. Airway obstruc-
tion can occur from something as simple as a
folding or closure of the normal hypopharynx,
merely requiring a chin-tilt, jaw-thrust maneuver to
reestablish a patient’s airway. More serious causes
of airway obstruction involve tracheal injury or for-
eign body aspiration. Unless an obstructed airway
can be cleared immediately and ventilation reestab-
lished, intubation is required. Non-cuffed endotra-
cheal tubes are used, and the size is determined by
using the formula (age/4 + 4). In certain instances,
a cuffed tube is used, and the size is determined by
the formula (age/4  +  3.5). If a child can be ade-
quately ventilated with a bag-valve-mask, this
allows for a more controlled intubation and repeat
attempts if unsuccessful. If intubation cannot be
achieved, a cricothyroidotomy must be performed.
In children under 12 years, a needle cricothyroid-
otomy is accomplished by placing a large bore
needle/angiocath through the cricoid membrane for
temporary high flow oxygen administration fol-
lowed by an immediate tracheostomy. In children
12  years or older, an open cricothyroidotomy is
done followed by a tracheostomy; after the patient
is stabilized, other life-threatening issues are
addressed and the secondary survey is completed.
Foreign body aspiration requires airway control via
endotracheal intubation followed by removal via
rigid bronchoscopy (Fig. 19.1).
Fig. 19.1  Left bronchial foreign body
231
19.5.2 Tension Pneumothorax
A pneumothorax is caused by air entering the
potential space between the visceral and parietal
pleura resulting in air getting trapped in the hemi-
thorax. This occurs following a tear in the pulmo-
nary parenchyma due to rib fracture and
laceration, deceleration injury, crush injury, or
due to increased intrathoracic pressure causing
pulmonary rupture. Signs and symptoms include
chest pain, dyspnea, increased respiratory rate,
and hyperresonance on the injured side. A pneu-
mothorax can quickly result in cardiovascular
demise if tension within the hemithorax develops
(Fig.  19.2). Findings of hyperresonance, dimin-
ished breath sounds on the affected side, tracheal
deviation, and hemodynamic instability indicate
a potentially lethal tension pneumothorax requir-
ing immediate intervention. Chest x-ray is not
required to make this diagnosis and may waste
valuable time. Needle decompression is the most
efficient therapy for a tension pneumothorax; it is
accomplished by placing a large bore angiocath
in the second intercostal space along the midcla-
Fig. 19.2  Left tension pneumothorax with mediastinal
shift
232
vicular line just above the third rib. Alternatively,
the decompression can be performed in the ante-
rior axillary line just above the sixth rib. This
should be followed by placement of chest tube.
These interventions result in the expansion of the
lung and continued evacuation of air and fluid
until the injured lung seals and air leak resolves.
19.5.3 Open Pneumothorax
An open pneumothorax is caused by a persistent
or continuous communication between the envi-
ronment and the chest cavity and is sometimes
referred to as a sucking chest wound. The negative
intrathoracic pressure generated during inhalation
results in air from the environment being sucked
into the chest cavity and collapse of the lung on the
affected side. Signs and symptoms include short-
ness of breath, chest pain, decreased breath sounds
on the affected side, and sound of air being sucked
into the chest cavity. This can result in respiratory
arrest and cardiovascular collapse if not treated
promptly. Treatment involves sealing or closing
the hole in the chest wall. Vaseline gauze is com-
monly used. Placement of a chest tube evacuates
residual air from the affected hemithorax and
enables lung ­re-­expansion. If there is an associated
parenchymal injury, the chest tube is left in place
until this air leak resolves.
19.5.4 Flail Chest
Flail chest is caused by multiple rib fractures or dis-
articulation of the ribs from the sternum resulting in
an unstable segment of chest wall. The flail segment
demonstrates paradoxical chest wall movement
during the respiratory cycle and physical exam is
diagnostic. Radiographic findings are only support-
ive. Treatment involves adequate analgesia, positive
pressure ventilation, and possible rib fixation stabi-
lizing the flail segment. Placement of a chest tube
may be required for associated pneumothorax or
hemothorax. As with less complicated rib fractures,
pain control and pulmonary toilet are essential.
D. E. Sawaya et al.
19.5.5 Hemothorax
A traumatic hemothorax is the accumulation of
blood in the hemithorax following blunt or pen-
etrating trauma. Bleeding can result from rib
fractures that lacerate the intercostal vessels,
pulmonary parenchymal laceration, or, less
commonly, great vessel injury. A massive hemo-
thorax results from life-threatening exsanguina-
tion into the chest cavity requiring immediate
intervention. Signs and symptoms include
shock, hypotension, diminished breath sounds
and dullness to percussion on the affected side,
chest pain, shortness of breath, and oxygen
requirement. This is a clinical diagnosis, espe-
cially in the unstable child. In the stable child, a
chest x-ray and/or CT scan confirms the diagno-
sis (Fig. 19.3). Standard therapy includes large
bore tube thoracostomy for evacuation of fluid
and air providing for lung re-expansion.
Resuscitation with intravenous fluid and possi-
bly packed red blood cells is required.
Thoracotomy is indicated for hemodynamic
instability, loss of >25% total blood volume
with initial chest tube placement, or persistent
bleeding greater than 2 ml/kg/h. A persistent or
retained hemothorax is an indication for video-
assisted thoracoscopy (VAT) or thoracotomy
with drainage.
Fig. 19.3  Right hemothorax following a gunshot wound
to the chest
19  Pediatric Thoracic Trauma 233

19.5.6 Cardiac Tamponade Commotio cordis is a life-threatening

and Commotio Cordis arrhythmia, resulting from a direct blow to the
precordium. Several cases of precordial blows
Penetrating cardiac trauma quickly results in life-­ have been reported from sports such as baseball,
threatening injuries requiring surgical interven- softball, soccer, lacrosse, and karate. Commotio
tion. These are typically manifested by cordis triggers ventricular fibrillation or other
cardiovascular instability. Traumatic cardiac fatal arrhythmias that quickly deteriorate.
pericardial effusions and tamponade result from Defibrillation within 1–2  min results in
the accumulation of blood, fluid, or, rarely, air 80–100% survival rates. Survivors have mini-
between the heart and the pericardium. This mal to no identifiable laboratory abnormalities
impairs cardiac filling during diastole. Signs and but usually have EKG abnormalities including
symptoms include muffled heart sounds, hypo- S-T elevation or complete heart block.
tension, distended neck veins (Beck’s triad), nar-
row pulse pressure, and pulsus paradoxus.
Echocardiography in the Emergency Department 19.6 Potentially Life-Threatening
(ED) is the radiologic test of choice and can help Injuries Found During
guide therapy. Emergency thoracotomy is Secondary Survey
required for cardiac repair and may be performed
in the trauma bay for hemodynamic instability. 19.6.1 Tracheobronchial Injury
Rib disarticulation from sternum may provide
better exposure. A vertical pericardiotomy is per- Tracheal and bronchial injuries are uncommon in
formed, with identification and repair of cardiac the pediatric population but have been reported
injuries with 5-0 nonabsorbable monofilament following thoracic crush injuries and penetrating
suture and pledgets. During pericardiotomy, care injuries. Signs and symptoms include chest pain,
is taken to avoid the phrenic nerve and the peri- dyspnea, hypoxia, stridor, crepitance in the neck
cardium remains open after cardiac repair. In and chest wall, mediastinal air, tension pneumo-
patients with equivocal signs and symptoms of thorax, and hemoptysis. Placement of chest tube
pericardial effusion or tamponade, a subxyphoid for pneumothorax results in identification of a
pericardial window should be performed. This large non-resolving air leak. Further diagnostic
allows the team to assess the presence of blood or evaluation is best accomplished via rigid bron-
fluid within the pericardium. A note of caution to choscopy for direct visualization of the injury
be ready for thoracotomy or sternotomy is war- and evacuation of blood and secretions. Eighty
ranted. Findings range from decompensation percent of these injuries typically occur within
during pericardiotomy to finding blood in the 2  cm of the carina in the trachea or main stem
pericardium (both necessitating thoracotomy) or bronchi and require early operative repair. Flap
a negative pericardiotomy. Pericardiocentesis is coverage of the repair with pleura or muscle rein-
less commonly indicated but useful during resus- forces and protects the repair.
citation of an unstable patient with profound
hypotension and suspected cardiac tamponade. A
large bore needle/angiocath is inserted in the sub- 19.6.2 Pulmonary Contusions
xyphoid space at a 45° angle while aspirating
with a syringe. If blood is obtained, the angiocath Pulmonary contusions are due to alveolar hemor-
is left in the pericardial space while preparations rhage and parenchymal destruction. Pulmonary
for thoracotomy are made. Pericardiocentesis can contusions, the most common injury found in
be diagnostic and therapeutic but may result in children with thoracic trauma, can result from
further cardiac injury from needle instrumenta- blunt or penetrating trauma. Only 40% of these
tion. Cardiac rupture due to massive blunt trauma patients have associated rib fractures. This low
is typically lethal. percentage is due to increased pediatric chest
234
Fig. 19.4  Left pulmonary contusion due to blunt thoracic
trauma
wall compliance. Signs and symptoms of pulmo-
nary contusions include chest pain, shortness of
breath, hypoxia, and, less commonly, hemopty-
sis. Pulmonary contusions may present insidi-
ously during the first 24  h following thoracic
trauma with worsening pulmonary symptoms
and delayed radiologic findings. Chest x-ray
remains the initial diagnostic exam, and com-
puter tomography is used for complicated cases
when further imaging is required (Fig.  19.4).
Treatment involves supplemental oxygen, fluid
restriction as tolerated, chest physiotherapy to
clear secretions and blood, as well as adequate
pain control. Intubation and mechanical ventila-
tion are infrequently required. The mortality rate
due to isolated pulmonary contusions is low, and
there are no long-term respiratory problems.
19.6.3 Myocardial Contusion
Substernal chest pain following blunt chest
trauma is indicative of myocardial contusion.
Other signs and symptoms include cardiac
arrhythmias and hypotension. Myocardial contu-
sion is the most common pediatric cardiac injury,
accounting for 95% of blunt cardiac injuries.
Motor vehicle accidents are the leading cause.
Diagnostic evaluation reveals elevated cardiac
enzymes, and EKG shows S-T changes, arrhyth-
mias, or heart block. Echocardiography may
show wall motion changes consistent with contu-
D. E. Sawaya et al.
sion or hematoma. Most injuries are self-limiting
but can result in aneurysm formation and possi-
bly cardiac rupture. Patients with myocardial
contusions should be continuously monitored
until free of arrhythmias for 24 h.
19.6.4 Diaphragmatic Injuries
Diaphragmatic injuries are more commonly
caused by penetrating trauma. Signs and symp-
toms include dyspnea, abdominal pain, vomit-
ing, and decreased breath sounds. These injuries
can be diagnostic dilemmas. The initial radio-
logic exam consists of chest x-ray with a naso-
gastric tube (NGT) in place. With a left
diaphragmatic injury, the stomach may be herni-
ated into the chest and the NGT confirmatory of
the injury. Helical computer tomography is cur-
rently the most useful exam with 70–100% accu-
racy. MRI is as accurate but commonly not
available in the acute setting and requires much
more time. This is costly especially during the
golden hour. Up to one-half of diaphragmatic
injuries are not found upon initial diagnostic
evaluation. Nearly one-­third are found intraop-
eratively, either incidentally or with minimally
invasive surgical techniques. Isolated diaphrag-
matic rupture from blunt trauma is more com-
mon in children and occurs more frequently on
the left side (Fig. 19.5).
Primary surgical repair with nonabsorbable
suture is most common; a patch is used if the
Fig. 19.5  Left diaphragmatic rupture
19  Pediatric Thoracic Trauma
defect is too large and not amenable to primary
closure. Closure of larger defects with an inter-
costal muscle flap has also been accomplished.
19.6.5 Esophageal Rupture
Esophageal rupture is a very rare injury, resulting
from rapid elevation of intraluminal pressure. In
cases of esophageal rupture, this is relieved by
laceration of the wall rather than intraluminal dis-
sipation. Signs and symptoms include chest pain,
fever, subcutaneous emphysema, and dysphagia.
X-rays may reveal subcutaneous air (Fig. 19.6).
The resulting mediastinal sepsis can prove to be
fatal if not diagnosed and treated promptly.
Esophageal debridement, closure, and drainage
remain the primary course of treatment.
19.6.6 Great Vessel Injury
Great vessel injury in the pediatric population is
uncommon but can be devastating when the
injury is not promptly recognized and treated.
Aortic rupture is suspected with rapid decelera-
tion injuries from motor vehicle crashes and falls
from great heights. The most common site is at
the level of the ligamentum arteriosum. Signs and
symptoms include chest pain and hemodynamic
Fig. 19.6  Cervical soft tissue air and mediastinal air
from esophageal rupture
235
Fig. 19.7  Mediastinal widening and loss of aortic knob
due to aortic rupture
instability. Chest X-ray may reveal a widened
mediastinum, depressed left bronchus, loss of the
contour of the aortic knob, first rib fracture, and
deviation of the esophagus demonstrated with a
nasogastric tube in place (Fig. 19.7).
The historic diagnostic test of choice is arteri-
ography but is now largely replaced by helical
computer tomography angiography (CTA) of the
chest. Transesophageal echocardiography may
also be useful. Treatment consists of aortic repair,
commonly requiring cardiopulmonary bypass.
Primary repair may be accomplished with nonab-
sorbable monofilament sutures with pledgets and
vascular graft interposition may also be required.
Branches of the thoracic aorta can usually be
repaired without cardiopulmonary bypass.
19.7 Non-Life-Threatening
Injuries Often Found
on Physical Exam or Chest
Radiograph
19.7.1 Simple Pneumothorax
A simple pneumothorax occurs when air gets
accumulated in the chest cavity without putting it
in any tension. Decreased breath sounds may be
observed as well as chest pain and shortness of
breath. Chest x-ray is diagnostic and placement
of chest tube is therapeutic (Fig.  19.8). In rare
cases, a pneumothorax may remain very small
236
Fig. 19.8  Simple left pneumothorax
due to a transient air leak from the parenchyma
that readily stops. In this situation, conservative
management may prevail without the placement
of chest tube.
19.7.2 Small Hemothorax
A small hemothorax is caused by minor bleeding
in the chest cavity due to pulmonary parenchy-
mal injury or chest wall injury. Only the place-
ment of a chest tube is required for drainage if the
lung volume is compromised.
19.7.3 Rib Fractures
Rib fractures in children are less common than in
adults due to the increased compliance of the
chest wall. Therefore, rib fractures are felt to be a
marker of increased injury severity. Rib fractures
are also commonly associated with child abuse.
Nearly two-thirds of rib fractures in children
under three years of age result from abuse. The
most common cause of rib fractures in older chil-
dren are motor vehicle crashes. Signs and symp-
toms include chest pain, dyspnea, and splinting
on the side of the fracture which can result in
atelectasis and pneumonia. Chest x-ray will con-
firm the diagnosis and more importantly evaluate
for an underlying pulmonary injury, pneumotho-
rax, or hemothorax. Rib fractures typically
require no surgical therapy. Pain control and pul-
D. E. Sawaya et al.
monary toilet significantly reduce the risk of
complications from these fractures. Children
with first rib fractures should be evaluated for
vascular injury.
19.7.4 Chest Wall Laceration
Chest wall lacerations occur from multiple dif-
ferent mechanisms but the treatment remains
the same. Wound exploration, foreign body
removal, control of bleeding, debridement, and
wound closure are the basics of laceration
management.
19.7.5 Traumatic Asphyxia (Perthes
Syndrome)
Traumatic asphyxia is the inability to breathe due
to compression of the thorax. Removal of the
compressive force allows resumption of breath-
ing and recovery. Commonly reported causes are
compression from garage doors, motor vehicles,
furniture, and other heavy objects. This results in
global hypoxia most often rendering the patient
unconscious followed by cerebral injury and
death if the compression is not promptly relieved.
Signs and symptoms include subconjunctival
hemorrhage; facial, neck, and chest petechiae;
and facial edema (Fig. 19.9). In hospitals, mortal-
ity rates are very low if these children survive the
actual event.
Fig. 19.9  Subconjunctival hemorrhage, facial petechiae,
and edema from asphyxia
19  Pediatric Thoracic Trauma
19.8 Conclusion
Prevention of pediatric trauma should be our ulti-
mate goal and can be realized through safety and
education. Most thoracic injuries heal without the
need for surgical intervention. The golden hour
of trauma is realized when pediatric thoracic
injuries requiring surgical intervention are
diagnosed.
Further Reading
Bliss D, Silen M (2002) Pediatric thoracic trauma. Crit
Care Med 30:s409–s415
Coran AG, Adzick NS, Krummel TM, Laberge JM,
Shamberger RC, Caldamone AA (2012) Pediatric sur-
gery, 7th edn. Elsevier Saunders
237
Haxhija EQ, Nores H, Schober P, Hollwarth ME (2004)
Lung contusion—lacerations after blunt thoracic
trauma in children. Pediatr Surg Int 20:412–414
Mendelson KG, Fallat ME (2007) Pediatric injuries pre-
vention to resolution. Surg Clin N Am 87:207–228
Oldham KT, Colombani PM, Foglia RP, Skinner MA
(2005) Principles and practice of pediatric surgery.
Lippincott, Williams & Wilkins, Philadelphia, PA
Puri P, Höllwarth M (2006) Pediatric surgery. Springer,
Berlin, Heidelberg
Shehata S, Shabaan B (2006) Diaphragmatic injuries in
children after blunt abdominal trauma. J Pediatr Surg
41:1727–1731
Wesson DE et al (2017) Pediatric trauma, 2nd edn. Taylor
and Francis Group, New York
Westra SJ, Wallace EC (2005) Imaging evaluation of pedi-
atric chest trauma. Radiol Clin North Am 43:267–281
 Abdominal and Genitourinary
Trauma
Claire D. Gerall, Vincent P. Duron,
and Steven Stylianos
20.1 Introduction
Trauma is the leading cause of morbidity and
mortality in children with most traumatic injuries
secondary to blunt mechanisms. Blunt trauma
accounts for up to 90% of abdominal injuries
with the liver being the most commonly injured
organ followed by the spleen (Dalton et al. 2015).
Children are at increased risk of sustaining
intraabdominal injuries after high-energy trauma,
as their intraabdominal organs are proportionally
large and in close proximity to one another, and
the smaller size of pediatric patients results in a
greater degree of force per body surface area
(Stylianos and Ford 2008). Mechanisms of blunt
abdominal trauma vary by age, with the most
common causes being motor vehicle collisions
(MVC), falls, and sports-related activities, such
as cycling and skateboarding. In younger chil-
dren, nonaccidental trauma should be considered
when the history or mechanism of injury is not
clear (Carter and Moulton 2016).
Management of blunt abdominal trauma has
evolved significantly over the last few decades,
C. D. Gerall · V. P. Duron · S. Stylianos (*)
Division of Pediatric Surgery, Columbia University
Vagelos College of Physicians & Surgeons,
NYP-Morgan Stanley Children’s Hospital,
New York, NY, USA
e-mail: gerall@uthscsa.edu;
vd2312@cumc.columbia.edu;
ss128@cumc.columbia.edu
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_20
20
initiated by James Simpson’s publication in
1968 on the successful nonoperative treatment of
select children presumed to have a splenic injury.
During the first half of the nineteenth century,
surgical intervention for blunt solid organ injury
was common, as clinicians depended solely on
vitals, physical exam, and abdominal radio-
graphs to direct management. Successful nonop-
erative management was first demonstrated in
the 1960s in the Hospital for Sick Children in
Toronto; however, adoption of this practice was
slow to follow (Upadhyaya and Simpson 1968;
Aronson et  al. 1977). Advances in imaging
modalities, most notably, computed tomography
(CT), resulted in the ability to delineate injury
severity. In 1989, a solid organ injury grading
system was developed by the American
Association for the Surgery of Trauma (AAST),
allowing for treatment specific to injury grade,
further promoting nonoperative management
(Moore et  al. 1989). Most recently, treatment
guidelines based on the patient’s physiologic
response to injury and medical intervention have
been established, recommending nonoperative
management in hemodynamically stable patients
despite injury grade. With this change in prac-
tice, nonoperative treatment is currently utilized
in more than 90% of solid organ injuries in chil-
dren (Gaines 2009). This chapter will focus on
common blunt abdominal injuries and unique
aspects of care in children.
239
240
20.2 Diagnosis
Accurate diagnosis of intraabdominal injury in
children after blunt abdominal trauma remains
challenging and controversial (Calder et  al.
2017). The primary concerns in the initial evalu-
ation of the acutely injured child include airway,
breathing, and circulation. Following a quick, but
thorough physical exam, imaging modalities are
then utilized in hemodynamically stable patients
to include plain radiography of the C-spine,
chest, and pelvis. Other plain abdominal films
offer little guidance in the acute evaluation of the
pediatric trauma patient. As a result of improved
imaging modalities, prompt identification of
potentially life-threatening intraabdominal inju-
ries is possible in the majority of children. With
advances in treatment and the resulting improve-
ment in survival of pediatric trauma patients,
risks of long-term effects of radiation exposure
must be balanced with the efficacy of the imaging
source when determining the best diagnostic
modality.
20.2.1 Computerized Tomography
CT is well known as the imaging study of choice
for the evaluation of injured children as it is non-
invasive and readily accessible in most health
care facilities. Additionally, it is a very accurate
method of identifying and qualifying the extent
of intraabdominal solid organ injury and results
in reduced incidence of nontherapeutic explor-
atory laparotomy (Stawicki 2017). Intravenous
contrast and utilization of dynamic methods of
scanning have optimized vascular and parenchy-
mal enhancement in patients with intraabdominal
injury. It is important to note that in patients with
an additional concern for head trauma, a head CT
without contrast should be performed first to
avoid contrast concealing a hemorrhagic brain
injury.
Not all trauma patients with potential abdomi-
nal injuries are candidates for CT evaluation.
Hemodynamic stability is a key determining fac-
tor; CT can be used in patients who are stable;
however, hemodynamically unstable patients
C. D. Gerall et al.
should not be taken out of an appropriate resusci-
tation room for imaging. Patients with hemody-
namic instability may benefit from an alternative
diagnostic study, such as Focused Abdominal
Sonography for Trauma (FAST), or may neces-
sitate urgent operative intervention. Additionally,
patients with obvious penetrating injury may
necessitate immediate operative intervention
without prior imaging. Although evaluation with
CT has greatly advanced our understanding and
treatment of blunt abdominal injuries, it contin-
ues to lack the accuracy to detect a hollow viscus
injury (Peterson et  al. 2015). CT scans also
expose patients to high levels of radiation, an
important consideration, as improved treatment
modalities for significantly injured trauma
patients have led to increased survival and the
potential for long-term effects of radiation
exposure.
20.2.2 Focused Abdominal
Sonography for Trauma
The standard FAST exam consists of four views,
including Morrison’s pouch, the pouch of
Douglas, the left flank to include the perisplenic
anatomy, and a subxiphoid view to visualize the
pericardium (Fig. 20.1). Prior to the use of bed-
side sonography, diagnostic peritoneal lavage
(DPL) was used to screen hemodynamically
unstable adult patients for intraabdominal
injury. FAST was established as a rapid screen-
ing study for patients too unstable to undergo
abdominal CT and a noninvasive alternative to
DPL. Although DPL is not commonly performed
in the pediatric population, the role of FAST has
been investigated for screening of hemodynami-
cally unstable patients as well as an alternative to
CT in stable patients to prevent unnecessary radi-
ation exposure in children. Several studies have
demonstrated the high sensitivity and specificity
of FAST in adults. However, the utility of FAST
is controversial in the pediatric trauma popula-
tion. Multiple studies have assessed the use of
FAST in pediatric trauma patients with high
specificity, ranging from 88.5% to 95.0%, and
low sensitivity, ranging from 27.8% to 44.4%
20  Abdominal and Genitourinary Trauma
Fig. 20.1  Schematic of FAST exam noting probe place-
ment for subxiphoid. RUQ/Morrison’s pouch, LUQ/left
para-colic, and pelvic/Pouch of Douglas views. Source:
Original illustration by Mark Mazziotti, MD
(Calder et al. 2017; Liang et al. 2019). Although
sensitivity is slightly increased in children that
are more critically ill, a negative FAST result
alone does not exclude a significant intraabdomi-
nal injury and cannot preclude further diagnostic
workup (Liang et al. 2019). As most intraabdomi-
nal injuries in children are now managed
nonoperatively, clinical correlation is essential
­ intraabdominal injury, as well as a higher degree
for surgical decision-making.
Correlating an FAST exam with laboratory
results may also be beneficial. Sola et al. found
that a positive FAST, combined with AST or ALT
levels higher than 100 IU/L, was associated with
a sensitivity of 88%, specificity 98%, PPV 94%,
NPV 96%, and accuracy 96% in detecting an
intraabdominal injury in children (Sola et  al.
2009). We recommend using FAST as an adjunct
to history, physical examination, and laboratory
studies to help determine whether further evalua-
tion is required. As ultrasound does not expose
241
children to radiation, ideally, FAST could reduce
the number of CT scans performed. However,
this was not observed in a recent multiinstitu-
tional prospective trial that noted no significant
decrease in the use of CT scan after the FAST
exam, with 41% of patients undergoing CT scan
after FAST versus 46% who had a CT scan and
no FAST (Calder et  al. 2017). Although FAST
may obviate the need for CT scans in a small sub-
set of low-risk pediatric blunt abdominal trauma
patients, a CT scan with IV contrast remains the
gold standard for the diagnosis of intraabdominal
injuries in blunt trauma.
20.2.3 Laparoscopy
A CT scan has 96% specificity for detecting
injury to the liver, spleen, and kidney. However,
injury to the intestine, pancreas, mesentery, and
diaphragm are more challenging to diagnose,
with up to 21% of patients with bowel or mesen-
teric injuries initially having negative studies
(Gaines and Rutkoski 2010). Unequivocal signs
of intestinal injury are often not seen, with signs
suggestive, but not diagnostic, of bowel or mes-
enteric injury being more common. Such find-
ings include bowel wall thickening, free
intraperitoneal fluid without evidence of solid
organ injury, and mesenteric stranding. These
findings lead to a diagnostic dilemma, which pre-
viously was resolved with exploratory laparot-
omy (Gaines and Rutkoski 2010). High rates of
exploration without findings of significant
of morbidity with laparotomy versus laparos-
copy, have led to the use of diagnostic laparos-
copy in patients with blunt abdominal trauma.
Additionally, hemodynamically stable patients
with penetrating abdominal trauma can benefit
from diagnostic laparoscopy, as CT scans are
often inconclusive and local wound exploration
is challenging in the pediatric population. Recent
reports of the use of diagnostic laparoscopy in
pediatric trauma patients have demonstrated an
increased diagnostic accuracy with no reported
missed injuries, avoidance of laparotomy in up to
242
62% of patients, and a significant decrease in the
total number of both intensive care unit and hos-
pital days (Gaines and Rutkoski 2010).
In addition to the use of laparoscopy in the
evaluation of trauma-related intraabdominal inju-
ries, the utility of laparoscopy for definitive man-
agement of such injuries is well documented in
the pediatric population (Molinaro et  al. 2009).
Although most intraabdominal injuries in chil-
dren can be successfully managed nonopera-
tively, repairs of intestinal perforations, bladder
ruptures, liver lacerations, diaphragmatic inju-
ries, and splenic injuries have all been reported
(Streck et al. 2006; Huang et al. 2012). The utility
of laparoscopy may be highest in hemodynami-
cally stable children with equivocal imaging;
however, clear indications for laparoscopy versus
laparotomy in pediatric trauma patients have yet
to be established. As minimally invasive surgery
becomes more commonplace for pediatric sur-
geons, the use of laparoscopy will be more widely
used and accepted in trauma, with more out-
comes studies to report.
20.2.4 The “Seat Belt Sign”
Children with lap belt restraints while passengers
are in high-speed motor vehicle collisions may
present with visible seat belt signs (SBS), ecchy-
mosis, and/or abrasions in the pattern of a lap belt
restraint (Fig.  20.2). This finding is associated
with increased abdominal injuries to both solid
Fig. 20.2  Seatbelt sign across lower abdomen of a pedi-
atric patient
C. D. Gerall et al.
and hollow organs. A triad of injuries has been
noted in these patients, including abdominal wall
contusions/herniation, chance fractures of the
lumbar spine, and intraabdominal injury, often
isolated jejunal/ileal perforations. Recent litera-
ture shows that the predictive factors for the need
for abdominal exploratory surgery in patients
presenting with an SBS vary between the adult
and pediatric trauma population. Several studies
suggest that in adult patients, findings on CT
scan, specifically increased thickness of abdomi-
nal wall contusion, location of contusion above
the anterior superior iliac spine (ASIS), as well as
the presence of free fluid, have significantly
improved predictive capacity of needing opera-
tive treatment over clinical evidence of SBS
(Johnson and Eastridge 2017). Additionally, mul-
tiple studies have noted a 100% negative predic-
tive value for CT scan in identifying adult patients
who need laparotomy (Delaplain et  al. 2020).
This has recently raised the question if admission
and observation are needed in adult patients with
negative CT findings.
Conversely, multiple reports have found that a
positive SBS is associated with an increased risk
for intraabdominal injury in pediatric patients
with high false-negative rates of CT scan
(Drucker et  al. 2018; Kopelman et  al. 2018). A
single-center experience looking at 857 children
who were involved in an MVC revealed that 17
patients (2.0%) required abdominal exploration
due to intraabdominal injury (Drucker et  al.
2018). Vitals, laboratory values, physical exam
findings, and CT scan results were compared
between patients with blunt abdominal trauma
requiring operative intervention and those not
requiring operative intervention. Physical exam
findings including seat belt sign, abdominal
bruising, abdominal wound, and abdominal ten-
derness were present significantly more fre-
quently in those requiring abdominal surgery
(p < 0.0001), with each finding having a negative
predictive value for the need for operative explo-
ration of at least 0.98, as seen in Table  20.1
(Drucker et al. 2018). This chapter concludes that
a lack of seatbelt sign, abdominal bruising,
abdominal wounds, or abdominal tenderness in
pediatric patients who have been involved in an
Table 20.1  Physical examination findings in pediatric patients after motor vehicle collision
Present in Present in Negative
20  Abdominal and Genitourinary Trauma

Operative Nonoperative Sensitivity, % Specificity, % PPV, % NPV, % (95% Positive Likelihood Likelihood Ratio
Finding Group (%) Group (%) p Value (95% CI) (95% CI) (95% CI) CI) Ratio (95% CI) (95% CI)
Seat belt sign 12 (70.6) 147 (17.6) <0.0001 70.6 82.4 7.6 99.3 (98.5–99.7) 4.0 (2.9–5.7) 0.4 (0.2–0.8)
(44.0–89.7) (79.7–85.0) (5.5–10.3)
Abdominal bruising 7 (41.2) 77 (9.2) <0.0001 41.2 90.8 8.2 98.7 (98.1–99.1) 4.5 (2.5–8.2) 0.7 (0.4–0.9)
(18.4–67.1) (88.7–92.7) (4.7–14.3)
Open abdominal 13 (76.5) 8 (1.0) <0.0001 76.5 99.1 61.9 99.5 (98.9–99.8) 80.3 (38.4–168.0) 0.2 (0.1–0.6)
wound (50.1–93.2) (98.1–99.6) (43.7–
77.3)
Abdominal 13 (76.5) 106 (12.6) <0.0001 76.5 87.4 10.9 99.5 (98.7–99.8) 6.06 (4.41–8.33) 0.27 (0.11–0.63)
tenderness, (50.1–93.2) (84.9–89.6) (8.2–14.4)
peritonitis
243
244
MVC are individually predictive of patients who
will not require surgical intervention. Although
the free fluid is used as a CT finding signifying
hollow viscous injury in adults, this finding is
nondiagnostic in children, with free fluid noted in
36.5% of patients not requiring operative inter-
vention and in only 29.4% of those requiring
operative treatment in this chapter (Drucker et al.
2018). Additionally, multiple studies note a rate
of false-negative CT scan for diagnosis of hollow
viscus injury in children that ranges from 4–26%,
highlighting the need for admission and serial
examination for delayed manifestation in pediat-
ric patients with SBS (Kopelman et al. 2018).
20.3 Differential Diagnosis
Abdominal trauma in pediatric patients can result
in varying degrees of injury to multiple organs
including, but not limited to, the spleen, liver,
intestines, pancreas, and kidneys. Late presenta-
tion and an inability to communicate pain within
the pediatric patient population increase the risk
of missed injury. In all pediatric abdominal
trauma cases, each organ system should be con-
sidered in the differential diagnosis with a high
index of suspicion for injury to multiple organs.
Additionally, in any patient in which the history
and/or mechanism of trauma described are incon-
sistent with the injury, nonaccidental trauma
should be investigated. Nonaccidental trauma is
most common in patients younger than 4  years
old and often includes multiple injuries, includ-
ing injuries to abdominal organs, bone fractures,
subdural hematoma, bruising, and burn marks.
20.3.1 Liver and Spleen Injury
The liver is the most common organ injured in
blunt abdominal trauma, followed by the spleen
(Stylianos and Ford 2008). Children are at
increased risk for injury to upper abdominal
organs due to their underdeveloped rib cage and
often lack of soft-tissue padding as protection
from external forces. The most common mecha-
C. D. Gerall et al.
nisms of injury are falls and sports injuries; how-
ever, motor vehicle collisions and nonaccidental
trauma can also lead to liver and splenic injury.
The most reliable indicator of abdominal injury
in children is abnormal physical exam findings,
such as tenderness, abrasions, contusion, “handle
bar mark”, and “seat belt sign”. Additionally,
referred shoulder pain is common with liver
injury (Kehr sign). In severe injury with signifi-
cant bleeding, patients may present with signs of
shock, including poor perfusion, tachycardia, and
mental status changes.
As mentioned previously, the FAST exam can
be used to detect injury in hemodynamically
unstable patients with abdominal trauma; how-
ever, the utility in pediatric patients is less clear.
FAST should be used as an adjunct to history,
physical exam, and laboratory studies to deter-
mine the need for further evaluation. The gold
standard for diagnosing liver and splenic injury
in hemodynamically stable patients is CT with
intravenous contrast.
20.3.2 Bile Duct Injury
Persistent bile leak is seen in up to 6% of pediat-
ric patients with a blunt liver injury who undergo
nonoperative management (Temiz et  al. 2018).
Radionuclide scanning is recommended when
biliary tree injury is suspected. Delayed views
may show a bile leak, even if early views are nor-
mal. Endoscopic retrograde cholangiopancrea-
tography (ERCP) can also be used to aid in the
diagnosis, as well as provide treatment.
20.3.3 Abdominal Compartment
Syndrome
Abdominal compartment syndrome (ACS)
describes the deleterious effects of increased
intraabdominal pressure. ACS can lead to respi-
ratory insufficiency from worsening ventilation/
perfusion mismatch, hemodynamic compromise
from preload reduction due to IVC compression,
impaired renal function from renal vein compres-
20  Abdominal and Genitourinary Trauma
sion, as well as decreased cardiac output, intra-
cranial hypertension from increased ventilator
pressures, splanchnic hypoperfusion, and abdom-
inal wall overdistention. Trauma can lead to
hemoperitoneum, visceral, abdominal wall and/
or retroperitoneal edema, and use of abdominal/
pelvic packing which can cause ACS. The combi-
nation of tissue injury and hemodynamic shock
creates a cascade of events, including capillary
leak, ischemia–reperfusion, and release of
­vasoactive mediators, and free radicals, which
combine to increase extracellular volume and tis-
sue edema. Unfortunately, the risk of ACS is
increased in trauma patients as acidosis and
hypothermia as well as coagulopathy lead to a
worsening capillary leak and need for resuscita-
tion, respectively (Kirkpatrick et al. 2013).
In recent years, the use of high-volume resus-
citation protocols has led to a marked increase in
ACS (Vatankhah et al. 2018). Although some cli-
nicians rely on physical exams to diagnose ACS,
many studies have shown that clinical assessment
of IAP based on abdominal diameter or abdomi-
nal palpation is unreliable (Kirkpatrick et  al.
2013). More objective measures are recom-
mended, most commonly measuring bladder
pressure. This involves instilling 1 ml/kg of ster-
ile saline into the patient’s Foley catheter, with a
minimum instillation volume of 3 mL and a max-
imum volume of 25 mL. A pressure transducer or
manometer is connected via a three-way stop-
cock and zeroed at the level of the midaxillary
line. IAP is expressed in mmHg and measured at
end expiration in the supine position. In 2013, the
World Society of the Abdominal Compartment
Syndrome (WSACS) established new pediatric
guidelines for diagnosing IAH and ACS
(Kirkpatrick et al. 2013). IAP in critically ill chil-
dren is approximately 7  ±  3  mm Hg, with IAH
defined as a sustained or repeated pathological
elevation in IAP > 10 mmHG. ACS is diagnosed
when a patient has a sustained elevation in
IAP  >  10mmHG with new or worsening organ
dysfunction that can be attributed to elevated
IAP, at which point abdominal decompression
should be considered (Kirkpatrick et al. 2013).
245
20.3.4 Duodenal and Pancreatic
Injury
Duodenal and pancreatic injuries often cause a
diagnostic dilemma as they are frequently associ-
ated with other intraabdominal and/or multisys-
tem injuries, which can mask subtle physical and
radiographic diagnostic signs. Additionally,
although the retroperitoneal location is beneficial
as it lowers the chance of injury, it further
increases the difficulty of early diagnosis.
20.3.4.1 Duodenum
Due to its retroperitoneal location, duodenal inju-
ries are rare, occurring in less than 5% of all pedi-
atric patients with trauma-related abdominal
injuries (Peterson et al. 2015). Unlike in adults in
which penetrating trauma is the most common
mechanism, >70% of pediatric duodenal injury is
a result of blunt abdominal trauma, with MVC,
handlebar injury, and nonaccidental trauma
(NAT) being the most frequent causes (Peterson
et al. 2015). Nonaccidental trauma, secondary to
a direct assault to the epigastrium, is unfortu-
nately common in younger patients. Multiple
studies report that up to 100% of patients less
than 4 years old who present with traumatic duo-
denal injury are victims of nonaccidental trauma
(Carter and Moulton 2016). Duodenal injuries in
younger patients should raise suspicion if the his-
tory and/or mechanism of injury described are
inconsistent.
Injuries are most often seen in the second por-
tion of the duodenum and consist of duodenal
hematoma and duodenal perforation, with differ-
entiation between the two essential, as manage-
ment differs. Shilyansky et al. demonstrated the
similarity in the clinical presentation of duodenal
hematoma and perforation in a review of 27 chil-
dren who sustained blunt duodenal injury, 13
with duodenal perforations and 14 with duodenal
hematomas (Shilyansky et al. 1997). The clinical
presentation, laboratory evaluation, and radio-
graphic findings of those with duodenal hema-
toma versus perforation are summarized in
Table 20.2. These data demonstrate that the clini-
246 C. D. Gerall et al.

Table 20.2  Presenting symptoms and signs in children major vascular (41.3%), duodenal (19.3%),
with duodenal hematoma and duodenal perforation splenic (28%), and renal (23.4%) injuries
Duodenal Duodenal (Almaramhy and Guraya 2012). The mechanism
Patient Hematoma Perforation of injury is similar to that of duodenal injuries,
Number 14 13
with MVC, handlebar injury, NAT, and falls
Age (yr) 5 9a
ISS score 10 25
being the most common causes. The majority of
Seat belt worn: 6 (100) 5 (71) blunt pediatric pancreatic trauma occurs in the
number (%) pancreatic body, followed by the tail and head.
Presentation Diagnostic modalities, as well as management
Pain or tenderness: 10 (71) 12 (92) strategy of traumatic pancreatic injury in pediat-
number (%)
ric patients, remain controversial. Until recently,
Bruising: number 6 (43) 11 (85)
(%) a critical determinant of the management of pan-
GCS score 15 15 creatic trauma was pancreatic ductal integrity.
Associated injuries Traditionally, in hemodynamically stable
Pancreatic: number 7 (50) 3 (23) patients, a multidetector CT (MDCT) is used first
(%) to screen for pancreatic injury; however, the diag-
Lumber spine: 1 (7) 4 (31)
number (%)
nostic sensitivity of a CT scan varies widely,
Total: number (%) 11 (79) 8 (62) ranging from 47 to 100% (Rosenfeld et al. 2017).
Laboratory evaluation Magnetic resonance cholangiopancreatography
Hgb: mh %Hct 12.3/0.36 12.2/0.37 (MRCP) is often used when ductal disruption is
Amylase: units (%) 678 (64) 332 (46) unclear on CT.  Conflicting data exist regarding
GCS Glasgow Come Scale, Hct hematocrit, Hgb hemo- the utility of MRCP vs CT, with the previous
globin, ISS Injury Severity Scale studies noting the limitations of CT technology
Source: Shilyansky et al. Diagnosis and management of
duodenal injuries in children. J Pediatr Surg. 1997; in identifying pancreatic ductal injuries. More
32:880–886 recently, Panda et al. compared CT and MRCP to
a
 Statistically significant difference intraoperative findings among pediatric and adult
patients with high-grade pancreatic injuries, with
cal presentation is strikingly similar in both a reported 92.9% accuracy for identifying pan-
groups, with only age and ISS statistically differ- creatic duct injury in 14 patients who underwent
ent. CT scans were performed with IV and oral MRCP compared to 91.7% in 24 patients who
contrast and provided diagnostic capability as all had MDCT only (Panda et  al. 2015). Similarly,
patients with perforation had findings of extrava- Rosenfeld et  al. conducted a multiinstitutional
sation of air or enteral contrast into the retroperi- retrospective review of data collected from 11
toneal, periduodenal, or prerenal space, findings pediatric trauma centers from 2010 to 2015, com-
not seen in any of the patients with duodenal paring CT to MRCP for diagnosing ductal injury.
hematoma. Although the use of oral contrast in A total of 21 patients were included; mean age of
the trauma setting remains controversial and a 7.8 years. Although the pancreatic duct was visu-
delayed radiographic presentation of duodenal alized more frequently on MRCP (48% vs 5%,
rupture is documented in pediatric trauma p < 0.01), confirmed or suspected duct disruption
patients, free air or extravasation of enteral con- did not significantly differ between MRCP and
trast is diagnostic of perforation. CT, suggesting that MRCP may not be superior
for confirmation of duct integrity and therefore
20.3.4.2 Pancreas may not be a necessary adjunct for CT for visual-
Injuries to the pancreas are also uncommon and ization of pancreatic injuries in children
are diagnosed in less than 5% of children admit- (Rosenfeld et  al. 2018). The best modality to
ted for blunt abdominal trauma (Koh et al. 2017). evaluate the duct is ERCP. ERCP is the gold stan-
Concomitant abdominal visceral injuries are fre- dard for diagnosing pancreatic ductal disruption
quent, including hepatic (46%), gastric (42.3%), and allows for treatment via stent placement;
20  Abdominal and Genitourinary Trauma 247

however, the invasive nature and lack of wide risk for significant renal damage. Most major
availability for the pediatric patient population blunt renal injuries occur in association with
limit its utility. other major injuries of the head, chest, and abdo-
men; therefore, the urologic investigation should
be considered in patients with these injuries. It
20.3.5 Renal Injury should also be undertaken in all crush injuries to
the abdomen or pelvis when the patient has sus-
The kidney is the most commonly injured organ tained a severe deceleration injury. Since a renal
in the urogenital system. Children are more sus- pedicle injury or ureteropelvic junction (UPJ)
ceptible to major renal trauma secondary to less disruption may not be associated with one of the
cushioning from perirenal fat, weaker abdominal classic signs of renal injury, radiologic evaluation
musculature, and a less well-ossified thoracic of the urinary tract should always be considered
cage. In children, the kidney occupies a propor- in patients with a mechanism of injury that could
tionally larger space in the retroperitoneum than potentially injure the upper urinary tract.
in adults, another potential variable leading to CT is widely accepted as the gold standard
increased rates of injury. Major deceleration and imaging modality for hemodynamically stable
flexion injuries are common and can lead to renal children presenting with blunt renal trauma.
artery or vein injuries due to stretching forces on Spiral CT, including noncontrast phase, arterial
a normally fixed vascular pedicle. This type of contrast phase, and late phase, is used in the pedi-
injury may be more common in children because atric population as it can visualize the renal
of their increased flexibility and renal mobility. parenchyma, vascular structures, and collecting
Posttraumatic thrombosis of the renal artery system while simultaneously viewing other
occurs secondary to an intimal tear, which pro- organs. It is both sensitive and specific for dem-
duces turbulence, thrombosis, and eventual ves- onstrating parenchymal laceration, urinary
sel occlusion leading to renal ischemia. A high extravasation, delineating segmental parenchy-
index of suspicion must be maintained in order to mal infarcts, determining the size and location of
identify these injuries. surrounding retroperitoneal hematoma, and/or
After adequate resuscitation and life-­associated intraabdominal injury (Kartal et  al.
threatening injuries have been addressed, evalua- 2020). With efforts to decrease radiation expo-
tion of the genitourinary system can be sure in children, the investigation into the effi-
undertaken. Following any blunt injury, urologic cacy of renal ultrasound to evaluate for blunt
evaluation should be obtained if the patient has renal trauma is ongoing. In a retrospective review,
microscopic or gross hematuria, abdominal by Edwards et  al., of 24 children with 25 renal
ecchymosis, distention and significant drop in injuries (6 grade I–II and 19 grade III–V), the
hemoglobin, a palpable flank mass, or flank sensitivity of ultrasound for detection of grade
hematomas. Gross hematuria is the most reliable III–V injuries ranged from 79 to 100% with a
indicator for serious urologic injury and should negative predictive value (NPV) between 97 and
warrant investigation with imaging. Protocols for 100% (Edwards et al. 2020). Sensitivity and NPV
imaging in patients with blunt trauma and micro- were determined based on the review of ultra-
scopic hematuria, however, are not clear. sound images by four pediatric physicians
Traditionally, all children with any degree of (patients also had CT confirmation of injury
microscopic hematuria after blunt trauma under- which reviewers were blind to), with three out of
went renal imaging. More recently, observation four reviewers identifying all 19 injuries on ultra-
in stable patients with microscopic hematuria sound, while one only identified 15 out of 19.
<50 RBC/HPF has been noted. It is important to Ultrasound was less accurate for diagnosing
note that the degree of hematuria does not always grade I–II injuries; however, it was noted that all
correlate with the degree of injury. The presence of these injuries were treated nonoperatively with
of multisystem trauma significantly increases the CT scan confirmation not altering the clinical
248
course. This chapter suggests that ultrasound can
safely be used instead of CT for the evaluation of
blunt renal injury in hemodynamically stable
trauma patients; however, a prospective study is
needed for confirmation.
20.4 Management
20.4.1 Spleen and Liver Injury
The concept of nonoperative management for
pediatric blunt abdominal trauma patients was
first investigated in blunt spleen injury. In April
1968, Updhyaya and Simpson published their
landmark report on spleen injury in children, sug-
gesting that some children with suspected spleen
injury could be treated nonoperatively.
Nonoperative management was highly debated
until 2000 when the American Pediatric Surgical
Association (APSA) Trauma Committee devel-
oped evidence-based guidelines based on the
treatment of 856 children with isolated blunt
spleen and liver injury (Stylianos 2000). These
guidelines introduced the concept that grade of
injury based on CT imaging should determine
treatment and length of stay in stable children,
Updated APSA Blunt Liver/Spleen Injury Guidelines 2019
Admission
• ICU Admission Indicators
• Abnormal vital signs after initial volume resuscitation
• ICU
• Activity - Bedrest until vitals normal
• Labs − q6hour CBC until vitals normal
• Diet − NPO until vital signs normal and hemoglobin stable
• Ward
• Activity - No restrictions
• Labs − CBC on admission and/or 6 hours after injury
• Diet − Regular diet
Set Free
• Based on clinical condition NOT injury severity (grade)
• Tolerating a diet
• Minimal abdominal pain
• Normal vital signs
Fig. 20.3  Updated Liver/Spleen Injury APSA Guidelines
C. D. Gerall et al.
with nonoperative management a recommended
option. Soon after the establishment of these
guidelines, multiple studies were published sug-
gesting a pathway based on hemodynamics with-
out regard for the grade of injury in children with
isolated spleen or liver injury. In 2015, the
Arizona–Texas–Oklahoma–Memphis–Arkansas
Consortium (ATOMAC) guidelines were estab-
lished, which outlined management based on
physiologic parameters rather than the grade of
injury, as well as outlined clinical parameters to
define endpoints for nonoperative treatment.
These guidelines were applied to 1007 children
with blunt liver and spleen injury, with findings
indicating that nonoperative treatment was suc-
cessful in 93–97% of patients (Notrica et  al.
2015). In 2019, APSA acknowledged the shift in
the treatment strategy of these injuries, and
offered a new set of guidelines for the manage-
ment of blunt liver and spleen injuries, based on
the patient’s physiologic response to injury and
medical intervention (Fig.  20.3). Adapted from
the APSA guidelines, our institution developed
an algorithm that incorporates physiologic
parameters to guide the treatment of blunt splenic
and liver injury (Fig. 20.4). Although imaging is
still utilized to evaluate the abdomen, diagnose a
Procedures
• Transfusion
• Unstable vitals after 20 cc/kg bolus of isotonic IVF
• Hemoglobin < 7
• Signs of ongoing or recent bleeding
• Angioembolization • Operative exploration with
- Signs of ongoing bleeding Control of Bleeding
despite pRBC transfusion - Unstable vitals despite pRBC
- Not indicated for contrast transfusion
blush in admission CT - Consider massive transfusion
withuout unstable vitals protocol
Aftercare
• Activity Restriction
• Restricting activity to grade plus 2 weeks is safe
• Shorter restrictions may be safe but there is inadequate
data to support decreasing these recommendations
• Follow up Imaging
• Risk of delayed complications following spleen and liver
injuries is slow
• Consider imaging for sympomatic patients with prior
high grade injuries
20  Abdominal and Genitourinary Trauma 249

PEDIATRIC BLUNT LIVER/SPLEEN INJURY GUIDELINES

*Normal VS listed
CONFIRMED LIVER/SPLEEN in Pediatric
INJURY ON CT SCAN Trauma General
Care Admission
Orders

Yes Abnormal VS* after No

Admit to PICU,
initial volume Admit to Floor
Consider procedure
resuscitation

CBC 6hr after

CBC q6h until VS normal admission. Repeat only
if change in VS
observed

Bedrest and NPO until

VS normal
No activity restrictions,
Transfer to the Clear Liquid Diet,
Floor Advance as Tolerated
Yes

VS normal and Hb
stable?
No Tolerating diet, minimal
abdominal pain,
No Remain as inpatient ambulating, & normal
Yes VS?
Yes

Remain in the PICU,

Consider procedure
Discharge to home

Medications Procedures

Pain Control: Transfusion: Unstable VS after 20mg/kg bolus of isotonic IVF,

Tylenol 15mg/kg q6h PRN
Oxycodone 0.05-0.1 mg/kg q6h PRN(severe pain, on diet)
Morphine 0.05-0.1 mg/kg q6h PRN(severe pain,
when NPO)
Bowel Regimen:
Colace or Miralax daily PRN for constipation
Hb < 7, signs on ongoing/recent bleeding
Angioembolization: Signs of ongoing bleeding despite
pRBC transfusion (not indicated for contrast blush on
admission CT w/out unstable VS)
Operative exploration with control of bleeding: Unstable
VS despite pRBC transfusion; consider Mass Transfusion
Protocol (MTP)

Post Discharge Care:

• Non-contact activity restriction for [Grade + 2] weeks
• Full-contact activity to be determined at trauma clinic follow-up
• Return to school as tolerated with activity restrictions
• Follow-up clinic visit 2 weeks post-discharge
• Consider imaging in symptomatic patients with prior high-grade level of injury
• No Motrin x 1 month

Fig. 20.4  Pediatric blunt liver/spleen injury guidelines used at Morgan Stanley Children’s Hospital, adapted from the
new APSA guidelines
250 C. D. Gerall et al.

blunt splenic or liver injury, and rule out other be repeated frequently and in accordance with
injuries, the patient’s clinical status is used to the 2019 APSA guidelines, and a complete blood
guide therapy instead of the anatomic grade of count (CBC) should be obtained on admission
injury. and/or 6 hours after injury. If there is a concern
for ongoing bleeding or if vitals are abnormal,
20.4.1.1 Management of the Stable then subsequent hemoglobin checks are war-
Pediatric Patient with Blunt ranted. If vitals are stable, the patient should be
Spleen or Liver Injury able to ambulate and have a diet as deemed
Nonoperative management of hemodynamically appropriate.
stable children and adolescents with isolated Studies suggest that very few patients with
blunt liver or spleen injury, as well as select mul- blunt solid organ injury who have stopped bleed-
tiply injured patients is now the standard of care. ing will rebleed in the short term after trauma
In patients with other intraabdominal injuries, (Notrica et  al. 2015; Davies et  al. 2009).
these may take priority over the liver or spleen Additionally, the incidence of delayed bleeding is
injury, and require operative intervention, such as extremely rare (Davies et al. 2009). Factors iden-
certain hollow viscus, bladder, ureter, or pancre- tified that might increase the risk of failure of
atic injuries. Determining hemodynamic stability nonoperative management include a contrast
in a child suspected of having recent or ongoing blush on CT scan (van der Vlies et al. 2010), han-
bleeding is challenging. Unlike in adult trauma dlebar injury, and more than one abdominal solid
patients with hemorrhage, in which blood pres- organ injury, especially the pancreas (Holmes
sure, heart rate, and lactic acid parameters have Holmes et al. 2005). Discharge after an abbrevi-
been proposed to determine hemodynamic stabil- ated period of hospitalization (less than 24 hours
ity, these are not universally accepted in pediatric in some cases) has been demonstrated to be safe
trauma patients. The recently published pediatric-­ in patients with isolated low-grade injuries. The
specific shock index may provide some guidance 2019 APSA guidelines advocate early discharge
in defining hemodynamic stability; however, it is of patients when they are tolerating a regular diet,
not widely used in the initial management of have minimal abdominal pain, are ambulating,
these patients (Acker et al. 2015). and have normal vital signs. Since delayed bleeds
Clinical judgment of hemodynamics in pedi- have been reported, often outside of the APSA
atric patients remains the primary driver of suc- guideline period, standardized education for the
cess or failure. Current APSA guidelines using family and patient remains an important compo-
physiologic parameters to guide resuscitation nent of discharge instructions. Patients should be
label a patient as a candidate for nonoperative coached specifically on the signs and symptoms
management if the patient has normal vital signs of delayed bleeding and other complications.
and does not have signs of ongoing bleeding after
the initial resuscitation, defined as 20 cc/kg iso- 20.4.1.2 Management of the Pediatric
tonic intravenous fluid bolus. No single clinical Patient with Blunt Spleen or
parameter or test can predict that a child is at risk Liver Injury and Ongoing
of needing a blood transfusion or operation, so Bleeding
current guidelines rely mostly on physician eval- Children who show signs of ongoing bleeding
uation (Acker et al. 2015). and may require intervention or transfusion
If a patient is deemed to be a fluid responder should warrant admission to an intensive care
and/or does not have signs of ongoing bleeding, unit (ICU). Patients who are categorized as non-
they may be admitted to a nonICU level floor for responders, secondary to abnormal vital signs
nonoperative management, regardless of injury after an initial 20 mL/kg bolus of isotonic fluid,
grade. In this subset of patients, vital signs should should be administered a 10–20  mL/kg packed
20  Abdominal and Genitourinary Trauma
red blood cell transfusion. Other sources of shock
and bleeding should be ruled out and the surgeon
should decide if surgery or interventional radio-
logic treatment is indicated (Notrica et al. 2015).
Vital signs, admission hemoglobin, and pediatric-­
specific shock index on admission have shown
reliability to identify children who require an
intervention (Acker et  al. 2015). If the patient
responds to the blood transfusion, he/she should
remain in the ICU and continue to have serial
hemoglobin checks until stable. During this
time, the patient remains on bedrest and NPO. A
recurrent episode of hypotension should warrant
further procedure, including surgery or angioem-
bolization. Although 40  mL/kg PRBC is a vol-
ume at which the risk of death has been found to
increase significantly and is considered a reason-
able threshold for the failure of nonoperative
management by consensus studies, it is best to
begin mobilizing the operating room or IR per-
sonnel as soon as blood transfusion begins (Neff
et al. 2015).
Very few children fail nonoperative manage-
ment for blunt solid organ injury; however, early
failure is characteristic in the pediatric popula-
tion (Acker et  al. 2015; Holmes Holmes et  al.
2005). One large multiinstitutional retrospective
review of 1800 pediatric solid organ injuries
found that among over 1000 blunt liver and
spleen injuries in the series, only 19 patients
failed nonoperative management secondary to
shock or bleeding. The median time to operate
for all patients was three hours with 87% failing
by 24  hours. Only 0.6% of the total number of
patients failed after 24  hours (Holmes Holmes
et  al. 2005). Large database studies have con-
firmed that those children who require operative
intervention are usually identified within the first
6–12  hours of their hospital course (Holmes
et al., 2015; Nance et al., 2006). Risk factors for
failure at pediatric trauma centers include higher
Injury Severity Scale (ISS), lower Glasgow Coma
Scale (GCS) at presentation, higher organ injury
grade and/or multiple organs injured, and older
age (Wisner et al. 2015).
251
20.4.2 Interventions for Blunt Spleen
or Liver Injury
20.4.2.1 Angioembolization
Selective angioembolization may be a highly
effective adjunct to the nonoperative treatment of
blunt liver and spleen injury, with increasing
reports of use in the pediatric population.
Although historically most commonly used in
adult trauma patients with contrast blush on CT,
controversy exists in regard to the presence of
contrast extravasation as a sign of ongoing bleed-
ing requiring intervention in the pediatric trauma
population. Several small series have demon-
strated no association between contrast blush and
the need for intervention or failure of nonopera-
tive management (Davies et  al. 2010). In con-
trast, a recent systematic review of nine studies
including 117 pediatric patients with spleen or
liver injury with contrast extravasation found a
failure rate of 22.5% for nonoperative treatment
without angioembolization versus only 6.5% for
nonoperative treatment with angioembolization
(van der Vlies et al. 2010). The authors advocate
the use of contrast blush as a trigger for early
consideration of angioembolization in these
patients. Significant complications can occur
after angioembolization of liver and spleen
injury including post-procedure pain, contrast-
induced kidney injury, hemorrhage, intraab-
dominal abscess, bile leaks, gallbladder
ischemia, hepatic necrosis, and femoral artery
injury. The decision to use angioembolization
versus operative intervention in a patient who is
failing nonoperative management will depend
on institutional resources and how quickly they
can be mobilized.
20.4.2.2 Damage Control Surgery
In patients requiring emergent laparotomy for an
actively bleeding complex liver or spleen injury,
damage control surgery is the established stan-
dard of care. Damage control surgery includes an
abbreviated operation with control of obvious
vascular injuries and perihepatic packing (stop
252
the bleeding and close), ICU resuscitation (i.e.,
correcting hypothermia, coagulation, and oxygen
delivery issues), and a return to the operating
room for the definitive operation once the hemo-
dynamic and metabolic status is optimized (usu-
ally 24–72  hours after the trauma). The three
phases of damage control are detailed in
Table 20.3. It is important to note that control of
major vasculature should be considered if easily
accessible. Searching for an elusive or retrohe-
patic source of bleeding should not take prece-
dence over well-placed packs. The aim of damage
control surgery is to interrupt and avoid the lethal
triad of hypothermia, acidosis, and coagulopathy.
This triad creates a vicious cycle in which each
derangement exacerbates the others and the
physiologic and metabolic consequences of the
triad often preclude completion of the procedure.
Lethal coagulopathy from dilution, hypothermia,
and acidosis can occur rapidly; therefore, the
decision to proceed with damage control surgery
must be made quickly. In unstable patients with
active bleeding, this technique is superior in post-
operative outcome to more invasive procedures
such as clamping, hepatectomy or splenectomy,
direct coagulation, and ligation.
Proper packing is essential. The key to pack-
ing the liver is placing packs above and below the
liver, with packs placed strategically between the
liver and diaphragm, abdominal sidewall, and
infra-hepatic structures. The aim of the packs is
to compress the parenchyma with care to avoid
critical compression of the inferior vena cava. If
bleeding stops, it is best to leave the packs in
place without further manipulation. It is impor-
Table 20.3  Damage control strategy in the exsanguinat-
ing trauma patient
Pahase 1 Abbreviated laparotomy for exploration
Control of hemorrhage and contamination
Packing and temporary abdominal wall
closure
Pahase 2 Aggressive ICU resuscitation
Core rewarming
Optimize volume and oxygen delivery
Correction of coagulopathy
Pahase 3 Planned reopration(s) for packing change
Definitive repair of injuring
Abdominal wall closure
ICU Intensive care unit
C. D. Gerall et al.
tant not to dissect deeper into the injured paren-
chyma nor attempt mass ligatures or excessive
cautery as this can often lead to increased blood
loss and wasted time. The fascia should be left
open with the use of temporary abdominal clo-
sure techniques, including Vacuum-Assisted
Closure® appliances, Silastic® sheeting, or sur-
gical towels. This will expedite getting the patient
back to the ICU for further resuscitation. Signs of
continued hemorrhage should prompt a return to
the operating room for further inspection and
packing and/or angiography.
Although establishing an exact timeframe for
reexploration is controversial, once patients are
rewarmed, coagulation factors replaced, and oxy-
gen delivery optimized the patient can be returned
to the operating room for pack removal and defin-
itive repair of injuries. Alternatively, once ade-
quately resuscitated with blood and coagulation
products, angioembolization can be considered
for ongoing arterial bleeding. Extreme caution
and effective resource mobilization are required
when taking an actively bleeding pediatric trauma
patient to the angiography suite remote from the
operating room should there be further decom-
pensation. It is essential to emphasize that the
success of the abbreviated laparotomy and
planned reoperation depends on an early decision
to employ this strategy prior to irreversible shock.
Intraabdominal packing for control of exsangui-
nating hemorrhage is a life-saving maneuver in
highly selected patients in whom coagulopathy,
hypothermia, and acidosis render further surgical
procedures unduly hazardous. Early identifica-
tion of patients likely to benefit from abbreviated
laparotomy techniques is crucial for success.
Although extremely rare in the pediatric popu-
lation, the use of Resuscitative Endovascular
Balloon Occlusion of the Aorta (REBOA) has
been successfully described in adult trauma
patients. REOBA describes the rapid retrograde
insertion of a balloon catheter through the femo-
ral artery, advanced and inflated at the level of the
diaphragmatic descending aorta with the goal of
controlling inflow and allow time for hemostasis.
In severe liver trauma, endovascular balloon
catheters in the aorta and cava can be useful for
temporary vascular occlusion to allow access to a
20  Abdominal and Genitourinary Trauma
juxtacaval injury. The extremely rare need for
this intervention in children, however, makes it
challenging to develop the expertise and rapidly
assemble the resources necessary to use REBOA
in pediatric trauma centers.
20.4.3 Bile Duct Injury
ERCP with the placement of transampullary bili-
ary stents coupled with percutaneous drainage is
becoming the standard of care for treating biliary
duct injury following blunt hepatic trauma
(Aljahdali and Murphy 2018). Endoscopic b­ iliary
stents provide a less invasive alternative to surgi-
cal treatment; however, they may migrate or clog,
requiring further treatment. ERCP with sphinc-
terotomy can be used to decrease intrabiliary
pressure and encourage internal decompression
in patients with persistent bile leak. Although
ERCP is invasive and requires conscious seda-
tion, it is an accurate diagnostic modality that
that allows treatment of the injured ducts without
open surgery.
20.4.4 Abdominal Compartment
Syndrome
Medical management strategies, such as the use
of neuromuscular blockade, evacuation of any
intraluminal contents, and aggressive diuresis,
can be considered based on the patient’s clinical
status; however, decompressive laparotomy is the
gold standard treatment for abdominal compart-
ment syndrome. Several authors have found that
abdominal decompression resulted in decreased
airway pressures, increased oxygen tension, and
increased urine output in children with abdomi-
nal compartment syndrome as well as reduced
mortality by up to 37% (Popescu et  al., 2018).
These studies highlight the widely known fact
that early recognition and timely intervention are
essential for lower mortality rates in pediatric
patients. Prevention with early hemorrhage con-
trol and avoidance of excessive fluid resuscita-
tion, however, is the most beneficial. Using
balanced blood products and damage control
253
resuscitation protocols prevents excessive fluid
resuscitation and therefore ACS (Cotton et  al.
2011). In patients that are at high risk for devel-
oping IAH, serial or continuous IAP monitoring
should be considered. It is important to note that
ACS can not only occur after abdominal trauma
but also in extremity trauma.
After abdominal decompression, temporary
patch abdominoplasty is performed. Traditionally,
materials, such as silastic sheeting, Goretex®
sheeting, ostomy appliances, silo appliances, and
various mesh materials, have been used. Recently,
negative pressure wound treatment (NPWT) has
gained popularity in the pediatric population
(Fig. 20.5). The benefits of NPWT include reduc-
ing edema, recruiting fibroblasts, and inducing
inflammatory cell migration, directly reducing
bacterial populations by impairing bacterial
enzymatic processes and modifying the wound
environment, and reducing exposure to effluent
(Popivanov et al. 2017). Guidelines for pressure
settings and mode of suction for pediatric patients
are largely adapted from adult parameters and
require further studies to establish accepted
parameters. Enteric fistula with the use of NPWT
remains a concern; however, those reported in the
literature are mostly in newborns with open abdo-
men after complex surgical cases (Gutierrez and
Gollin 2012). Although techniques for temporary
wound closure have improved, facial closure
should be accomplished as soon as possible.
Fig. 20.5  Vacuum-pack closure technique after decom-
pressive laparotomy for abdominal compartment syndrome
254 C. D. Gerall et al.

20.4.5 Duodenal and Pancreatic

Injury

In contrast to the liver and spleen, injuries to the

duodenum and pancreas are much less frequent
with blunt abdominal trauma. Operative versus
nonoperative management is determined by the
severity of the injury and other associated
injuries.

20.4.5.1 Duodenum
The management of duodenal hematoma is
expectant in most cases, including bowel rest,
nasogastric decompression, and peripherally
inserted central catheter (PICC) placement for
parenteral nutrition. Multiple recent studies note
the success of nonoperative management; nota-
bly, Peterson et al. described the successful non-
operative management in 19 pediatric blunt
abdominal trauma patients with duodenal hema-
tomas (Peterson et al. 2015).
Duodenal perforation is treated operatively
with multiple possible strategies, as well as both
open and laparoscopic techniques depending on
the severity of the injury and the surgeon’s pref-
erence. Traditionally, the primary closure of the
duodenal injury was combined with duodenal
drainage and either pyloric exclusion with gastro-­
jejunostomy (Fig. 20.6) or gastric drainage with
feeding jejunostomy. These surgical options have
been shown to decrease the incidence of duode- Fig. 20.6  Lateral duodenal injury treated by primary
nal fistula, reduce the time to GI tract alimenta- duodenal repair and pyloric exclusion consisting of clos-
tion, and shorten hospital stay (Ladd et al. 2002). ing the pylorus with an absorbable suture and gastrojeju-
nostomy. Closed suction drainage of the repair is not
An effective surgical approach, when faced with
depicted in this drawing. Source: Original illustration by
complicated duodenal trauma, is the “three tube Mark Mazziotti, MD
technique”: duodenal closure (primary repair,
serosal patch, or anastomosis) with duodenal Previously, pyloric exclusion for severe duo-
drainage tube for decompression (tube 1), pyloric denal trauma was encouraged as it was thought to
exclusion with an absorbable suture via gastrot- result in less morbidity and a shorter hospital stay
omy and gastric tube placement (tube 2), and (Ladd et  al. 2002). However, recently, multiple
feeding jejunostomy (tube 3). Several closed suc- studies have looked at the treatment with primary
tion drains are placed adjacent to the repair. closure and drain placement without diversion
When the duodenum is excluded (via an absorb- (Smiley et al. 2012). Smiley et al. reported on 7
able suture for the temporary closure of the pylo- children with traumatic duodenal injuries ranging
rus), complete healing of the injury routinely from Grade I to IV, in which all patients under-
occurs prior to the spontaneous reopening of the went primary closure and drain placement with-
pyloric channel (Fig. 20.7). out exclusion and no complications were noted,
20  Abdominal and Genitourinary Trauma
Fig. 20.7 Duodenal diverticularization for combined
proximal duodenal and pancreatic injury. Resection and
closure of the duodenal stump, tube duodenostomy, tube
cholecystectomy, gastrojejunostomy, and multiple closed
suction drains are depicted. A feeding jejunostomy should
be strongly considered (not depicted). Source: Original
illustration by Mark Mazziotti, MD
suggesting that primary repair with drain place-
ment alone is safe even in more complex injuries
(Smiley et  al. 2012). Although previously
approached as a laparotomy, recent case reports
also noted successful laparoscopic repair of iso-
lated traumatic duodenal perforation in hemody-
namically stable patients (Huang et al. 2012). In
rare cases of severe duodenal injury, pancreatico-
duodenectomy (Whipple Procedure) and pan-
creas preserving duodenectomy are surgical
options. The Whipple Procedure should be
reserved for injuries to the duodenum and pan-
creas in which the common blood supply is
destroyed and any possibility of reconstruction is
impossible. When possible, pancreas preserving
duodenectomy is preferred as this technique
includes preservation of the pancreas, reduction
255
in the number of anastomoses, and avoidance of
manipulation of the biliary tree. It is important to
note that although pancreas preserving duode-
nectomy has the above-mentioned advantages,
studies show no benefit in the reduction of mor-
bidity or mortality when compared to the Whipple
Procedure (Isik et al. 2007). Although less widely
used, there are multiple reports of the use of tube
duodenostomy and feeding jejunostomy with or
without pyloric exclusion for large duodenal per-
forations in hemodynamically unstable patients
with a hostile abdomen (Isik et  al. 2007;
Bhattacharjee et al. 2011). In a hemodynamically
unstable trauma patient, long procedures such as
Whipple Procedure or pancreas preserving duo-
denectomy result in increased morbidity and
mortality. Tube duodenostomy follows the dam-
age control principles of trauma surgery with
recent studies showing decreased morbidity and
rates of the leak (Isik et al. 2007; Bhattacharjee
et al. 2011).
20.4.5.2 Pancreas
Although it is widely accepted that low-grade
injury that does not involve the pancreatic duct
can safely and effectively be managed with non-
operative management, controversy persists
regarding the optimal management of high-grade
injuries in the pediatric population. Traditional
management of high-grade injuries with duct
transection has previously been operative; how-
ever, recent reports of successful nonoperative
management, including ERCP with stent place-
ment, have highlighted this strategy as an option
for treating high-grade pancreatic injuries (Koh
et  al. 2017). Nonoperative management avoids
potential injury to the spleen during distal pan-
createctomy as well as complications associated
with surgery; however, data remain inconclusive
as to which management strategy leads to supe-
rior outcomes. Multiple retrospective multicenter
studies have tried to compare outcomes; how-
ever, the results are variable. Management strate-
gies utilized by institutions in these studies range
from observational care for all injury grades to
aggressive surgical management with debride-
ment and/or resections, and variation in treatment
is not only noted between but also within institu-
256
tions, with operative and nonoperative manage-
ment strategies reported equally (Naik-Mathuria
2016). The lack of a standard management path-
way likely plays a role in this variability and has
led to attempts to establish a standard clinical
pathway for nonoperative management of blunt
pancreatic injury.
In a systematic review of 674 pediatric patients
from 23 separate studies, 83% of patients with
low-grade injuries were managed nonoperatively
with a 96% success rate, while 46% of patients
with high-grade injuries were managed nonoper-
atively with an 89% success rate (Koh et  al.
2017). Pseudocysts occurred in 18% of patients
treated nonoperatively compared to only 4% in
patients treated with operative intervention
(p < 0.01). As nonoperative management becomes
more widely accepted, the incidence of peripan-
creatic fluid collections and pseudocysts have
increased, noted as high as 44% incidence in
children with grade III injuries (Rosenfeld et al.
2019). Although current clinical management
varies significantly, multiple studies support
observation compared to percutaneous or trans-
papillary drainage, noting no significant differ-
ence in total parenteral nutrition use, hospital
length of stay, time to tolerate regular diet, or
requirement of a definitive procedure for persis-
tent pseudocyst (Rosenfeld et al. 2019; Koh et al.
2017). Prospective trials are underway to validate
nonoperative management of both high-grade
pancreatic injuries, as well as the common
sequela of peripancreatic fluid collections.
20.4.6 Renal Injury
Similar to the management of traumatic injury to
all other abdominal solid organs, management of
renal injury is becoming more commonly nonop-
erative. Nonoperative management consists of
supportive care with bed rest, hydration, and anti-
biotics, as well as serial hemoglobin and blood
pressure monitoring. In 2019, evidence-based
pediatric blunt renal trauma practice manage-
ment recommendations were established through
a collaboration between the Eastern Association
C. D. Gerall et al.
for the Surgery of Trauma and the Pediatric
Trauma Society (Hagedorn et al. 2019). From a
literature review of 51 related articles, data sug-
gested that there is a reduced rate of renal loss
and blood transfusions in patients managed non-
operatively, angioembolization had decreased
rate of renal loss compared to surgical interven-
tion in patients with grade III–V injury and ongo-
ing or delayed bleeding, and the rate of
posttraumatic renal hypertension was 4.2%
(Hagedorn et al. 2019). From these findings, the
following evidence-based recommendations
were made: nonoperative management for pedi-
atric patients with blunt renal trauma of all
grades, angioembolization for hemodynamically
stable children with high-grade renal trauma and
ongoing or delayed bleeding, and routine blood
pressure check on the follow-up to diagnose post-
traumatic renal hypertension in children.
Similarly, in March 2019, a systematic review
published by Sujenthrian et al. (2019) reports that
nonoperative management is safe and effective in
stable patients with a higher renal preservation
rate, shorter length of stay, and comparable com-
plication rate to open surgical exploration.
Early complications of nonoperative treat-
ment can occur within the first 4  weeks of the
injury, and include delayed bleeding, abscess,
sepsis, urinary fistula, urinary extravasation, uri-
noma, and hypertension. Delayed hemorrhage is
a potentially life-threatening complication that
most often occurs within the first 2  weeks of
injury. Immediate intervention with either surgi-
cal exploration or angioembolization is indicated.
Persistent urinary extravasation has successfully
been managed by percutaneous drainage.
Hypertension in the early post-trauma period is
uncommon; however, hypertension may develop
in the ensuing months, and in most instances, it is
treated with medical management.
20.4.6.1 Renal Interventions
Although surgical exploration has traditionally
been the gold standard for patients requiring
intervention for renal trauma, advanced imaging
technology and better understanding of trauma
pathophysiology have prompted a more signifi-
20  Abdominal and Genitourinary Trauma 257

cant role for interventional radiology in trauma If an operation is required, early control of the
management. Endovascular strategies have vessels increases the rate of renal salvage. When
proven successful in the management of all proximal vascular control is initially achieved
grades of renal injury and can often obviate a before any renal exploration, nephrectomy is
lengthy operation and renal loss. This is sup- required in less than 12% of cases (Buckley Jill
ported by Jacobs et al. who reported lower rates and McAninch 2006). When primary vascular
of nephrectomy in patients treated conservatively control is not achieved and massive bleeding is
compared to those undergoing initial surgical encountered, in the rush to control bleeding, a
intervention, adjusted for age, renal injury grade, kidney that could have been salvaged may be sac-
and injury mechanism (Jacobs Micah et al. 2012). rificed unnecessarily. The surgeon must carefully
Transcatheter angioembolization is a minimally identify the kidney’s relationships with the poste-
invasive treatment method gaining wide accep- rior abdomen and the posterior parietal perito-
tance for the treatment of active bleeding due to neum. The colon is lifted from the abdomen and
renal parenchymal lacerations, contrast extrava- placed on the anterior chest in order to allow
sation, perirenal hematoma, arterial pseudoaneu- mobilization of the small bowel. The inferior
rysms, arteriovenous fistulas, and delayed mesenteric vein and the aorta are identified at this
bleeding following conservative treatment. point, and the posterior peritoneum is incised
Complications associated with transcatheter medial to the inferior epigastric vein. The aorta is
angioembolization are lower in children com- dissected above the level of the ligament of
pared to adults. Potential complications include Treitz, where the left renal vein is found crossing
the formation of hematoma in the arterial access anterior to the aorta. Retraction of the left renal
site, arterial injury, contrast nephropathy, renal vein exposes both renal arteries beneath, which
ischemia, and embolization of a nontargeted may now be isolated and controlled with vessel
organ. Although renal artery dissection is reported loops. Once vessel isolation is complete, an inci-
in up to 7.5% of adult patients, this complication sion is made in the peritoneum just lateral to the
has not been reported in children (Kartal et  al. colon. The colon is reflected medially to expose
2020). In a literature review published by Kartal the retroperitoneal hematoma in its entirety and
et  al. in 2020, the overall success rate of trans- the kidney may be exposed. If significant bleed-
catheter angioembolization in pediatric patients ing is encountered, the ipsilateral renal vessels
with blunt renal trauma is 89.5% with improved may be occluded. Warm ischemia time should
preservation of ipsilateral kidney function. not surpass 30 min. Renal vascular injuries must
When necessary, the goals of surgical renal be addressed promptly. Major lacerations to the
exploration are to either definitively treat major renal vein are repaired directly by venorrhaphy.
renal injuries with preservation of renal paren- Repair of renal arterial injuries may require a
chyma when possible, or to thoroughly evaluate a variety of techniques, including resection and
suspected renal injury. Absolute indications for end-to-end anastomosis, bypass graft with autog-
renal exploration include persistent life-­enous vein or synthetic graft, and arteriorrhaphy.
threatening bleeding, an expanding, pulsatile, or Traumatic renal artery occlusion requires many
uncontained retroperitoneal hematoma, or sus- of the same techniques for repair.
pected renal pedicle avulsion. A relative indica-
tion for exploration includes substantial
devitalized renal parenchyma. Urinary extravasa- 20.5 Conclusion
tion is no longer a relative indication, as patients
with major renal injury and urinary extravasation Continued advancements in the diagnosis and man-
can successfully be managed conservatively, with agement of children with blunt abdominal trauma
endoscopic management for persistent extravasa- have resulted in improved outcomes following
tion (Kartal et al. 2020). major injuries. New diagnostic modalities, includ-
258
ing improved cross-sectional imaging and diagnos-
tic laparoscopy, have allowed a more accurate
assessment and management of injury. Although
the management of pediatric trauma patients with
blunt abdominal injury has become increasingly
nonoperative, pediatric surgeons must be familiar
with current treatment guidelines as the decision not
to operate is always a surgical decision.
References
Acker SN, Ross JT, Partrick DA, Tong S, Bensard DD
(2015) Pediatric specific shock index accurately
identifies severely injured children. J Pediatr Surg
50(2):331–334
Aljahdali AH, Murphy JJ (2018) Bile duct injury in
children: is there a role for early endoscopic ret-
rograde cholangiopancreatography? Surg J (N Y)
4(3):e119–ee22
Almaramhy HH, Guraya SY (2012) Computed tomogra-
phy for pancreatic injuries in pediatric blunt abdomi-
nal trauma. World J Gastrointest Surg 4(7):166–170
Aronson DZ, Scherz AW, Einhorn AH, Becker JM,
Schneider KM (1977) Nonoperative management of
splenic trauma in children: a report of six consecutive
cases. Pediatrics 60(4):482
Bhattacharjee HK, Misra MC, Kumar S, Bansal VK
(2011) Duodenal perforation following blunt abdomi-
nal trauma. J Emerg Trauma Shock 4(4):514–517
Buckley Jill C, McAninch JW (2006) Selective manage-
ment of isolated and nonisolated grade IV renal inju-
ries. J Urol 176(6):2498–2502
Calder BW, Vogel AM, Zhang J, Mauldin PD, Huang
EY, Savoie KB, et al (2017) Focused assessment with
sonography for trauma in children after blunt abdomi-
nal trauma: a multi-institutional analysis. J Trauma
Acute Care Surg 83(2)
Carter KW, Moulton SL (2016) Pediatric abdominal injury
patterns caused by “falls”: a comparison between
nonaccidental and accidental trauma. J Pediatr Surg
51(2):326–328
Cotton BA, Reddy N, Hatch QM, LeFebvre E, Wade CE,
Kozar RA et al (2011) Damage control resuscitation is
associated with a reduction in resuscitation volumes
and improvement in survival in 390 damage control
laparotomy patients. Ann Surg 254(4):598–605
Dalton BGA, Dehmer JJ, Gonzalez KW, Shah SR (2015)
Blunt spleen and liver trauma. J Pediatr Intens Care
4(1):10–15
Davies DA, Fecteau A, Himidan S, Mikrogianakis A,
Wales PW (2009) What’s the incidence of delayed
splenic bleeding in children after blunt trauma? An
institutional experience and review of the literature. J
Trauma Acute Care Surg 67(3)
Davies DA, Ein SH, Pearl R, Langer JC, Traubici J,
Mikrogianakis A et al (2010) What is the significance
C. D. Gerall et al.
of contrast “blush” in pediatric blunt splenic trauma? J
Pediatr Surg 45(5):916–920
Delaplain PT, Barrios C, Spencer D, Lekawa M, Schubl
S, Dosch A et al (2020) The use of computed tomog-
raphy imaging for abdominal seatbelt sign: a single-­
center, prospective evaluation. Injury 51(1):26–31
Drucker NA, McDuffie L, Groh E, Hackworth J, Bell TM,
Markel TA (2018) Physical examination is the best
predictor of the need for abdominal surgery in chil-
dren following motor vehicle collision. J Emerg Med
54(1):1–7
Edwards A, Hammer M, Artunduaga M, Peters C, Jacobs
M, Schlomer B. (2020) Renal ultrasound to evaluate
for blunt renal trauma in children: a retrospective com-
parison to contrast enhanced CT imaging. J Ped Urol
16(5):557.e1–557.e7.
Gaines BA (2009) Intra-abdominal solid organ injury in
children: diagnosis and treatment. J Trauma Acute
Care Surg 67(2)
Gaines BA, Rutkoski JD (2010) The role of laparoscopy in
pediatric trauma. Semin Pediatr Surg 19(4):300–303
Gutierrez IM, Gollin G (2012) Negative pressure
wound therapy for children with an open abdomen.
Langenbeck’s Arch Surg 397(8):1353–1357
Hagedorn JC, Fox N, Ellison JS, Russell R, Witt CE,
Zeller K et al (2019) Pediatric blunt renal trauma prac-
tice management guidelines: collaboration between
the Eastern Association for the Surgery of Trauma and
the Pediatric Trauma Society. J Trauma Acute Care
Surg 86(5)
Holmes JH, Wiebe DJ, Tataria M, Mattix KD, Mooney
DP, Scaife ER et al (2005) The failure of nonoperative
management in pediatric solid organ injury: a multi-­
institutional experience. J Trauma 59(6):1309–1313
Huang C-L, Lee J-Y, Chang Y-T (2012) Early laparo-
scopic repair for blunt duodenal perforation in an ado-
lescent. J Pediatr Surg 47(5):e11–ee4
Isik B, Yilmaz S, Kirimlioglu V, Sogutlu G, Yilmaz M,
Katz D (2007) A life-saving but inadequately dis-
cussed procedure: tube duodenostomy. Known and
unknown aspects. World J Surg 31(8):1616–24; dis-
cussion 25–6
Jacobs Micah A, Hotaling James M, Mueller Beth A,
Koyle M, Rivara F, Voelzke BB (2012) Conservative
management vs early surgery for high grade pediat-
ric renal trauma—do nephrectomy rates differ? J Urol
187(5):1817–1822
Johnson MC, Eastridge BJ (2017) Redefining the
abdominal seatbelt sign: enhanced CT imag-
ing metrics improve injury prediciton. Am J Surg
214(6):1175–1179
Kartal I, Durmaz H, Cimen S, Cimen S, Ersoy H (2020)
Transcatheter embolixation in pediatric blunt renal
trauma: case report and review of the literature. Ulus
Travma Acil Cerrahi Derg 26(3):475–481
Kirkpatrick AW, Roberts DJ, De Waele J, Jaeschke R,
Malbrain MLNG, De Keulenaer B et al (2013) Intra-­
abdominal hypertension and the abdominal compart-
ment syndrome: updated consensus definitions and
clinical practice guidelines from the World Society
20  Abdominal and Genitourinary Trauma
of the Abdominal Compartment Syndrome. Intensive
Care Med 39(7):1190–1206
Koh EY, van Poll D, Goslings JC, Busch OR, Rauws EA,
Oomen MW et al (2017) Operative versus nonopera-
tive management of blunt pancreatic trauma in chil-
dren: a systematic review. Pancreas 46(9):1091–1097
Kopelman TR, Jamshidi R, Pieri PG, Davis K, Bogert J,
Vail SJ et al (2018) Computed tomographic imaging in
the pediatric patient with a seatbelt sign: still not good
enough. J Pediatr Surg 53(2):357–361
Ladd AP, West KW, Rouse TM, Scherer LR, Rescorla FJ,
Engum SA et al (2002) Surgical management of duo-
denal injuries in children. Surgery 132(4):748–753
Liang T, Roseman E, Gao M, Sinert R (2019) The utility
of the focused assessment with sonography in trauma
examination in pediatric blunt abdominal trauma: a
systematic review and meta-analysis. Pediat Emer
Care. 9000; Publish Ahead of Print
Molinaro F, Kaselas C, Lacreuse I, Moog R, Becmeur
F (2009) Postoperative intestinal obstruction after
laparoscopic versus open surgery in the pediat-
ric population: a 15-year review. Eur J Pediatr Surg
19(3):160–162
Moore EE, Shackford SR, Pachter HL, McAninch JW,
Browner BD, Champion HR et  al (1989) Organ
injury scaling: spleen, liver, and kidney. J Trauma
29(12):1664–1666
Naik-Mathuria B (2016) Practice variability exists in
the management of high-grade pediatric pancreatic
trauma. Pediatr Surg Int 32(8):789–794
Nance ML, Holmes JH, Wiebe DJ (2006) Timeline to
operative intervention for solid organ injuries in chil-
dren. J Trauma 61(6):1389–1392
Neff LP, Cannon JW, Morrison JJ, Edwards MJ, Spinella
PC, Borgman MA.  Clearly defining pediatric mas-
sive transfusion: cutting through the fog and fric-
tion with combat data. J Trauma Acute Care Surg
2015;78(1):22–8; discussion 8–9
Notrica DM, Eubanks JW III, Tuggle DW, Maxson RT,
Letton RW, Garcia NM et  al (2015) Nonoperative
management of blunt liver and spleen injury in chil-
dren: evaluation of the ATOMAC guideline using
GRADE. J Trauma Acute Care Surg 79(4)
Panda A, Kumar A, Gamanagatti S, Bhalla AS, Sharma R,
Kumar S et al (2015) Evaluation of diagnostic utility
of multidetector computed tomography and magnetic
resonance imaging in blunt pancreatic trauma: a pro-
spective study. Acta Radiol 56(4):387–396
Peterson ML, Abbas PI, Fallon SC, Naik-Mathuria BJ,
Rodriguez JR (2015) Management of traumatic duode-
nal hematomas in children. J Surg Res 199(1):126–129
Popivanov G, Kjossev K, Mutafchiyski V (2017) The
open abdomen  - still a challenge for the surgeons.
Which is the best technique for temporary abdominal
closure? A focus on negative pressure wound therapy.
G Chir 38(6):267–272
Rosenfeld EH, Mazzolini K, DeMello A, Yu YR, Lee TC,
Naik-Mathuria B et  al (2017) Postoperative feeding
regimens after laparoscopic gastrostomy placement.
J Laparoendosc Adv Surg Tech A 27(11):1203–1208
259
Rosenfeld EH, Vogel A, Russell RT, Maizlin I, Klinkner
DB, Polites S et al (2018) Comparison of diagnostic
imaging modalities for the evaluation of pancreatic
duct injury in children: a multi-institutional analysis
from the Pancreatic Trauma Study Group. Pediatr
Surg Int 34(9):961–966
Rosenfeld EH, Vogel AM, Jafri M, Burd R, Russell R,
Beaudin M et  al (2019) Management and outcomes
of peripancreatic fluid collections and pseudocysts fol-
lowing non-operative management of pancreatic inju-
ries in children. Pediatr Surg Int 35(8):861
Shilyansky J, Pearl RH, Kreller M, Sena LM, Babyn PS
(1997) Diagnosis and management of duodenal inju-
ries in children. J Pediatr Surg 32(6):880–886
Smiley K, Wright T, Skinner S, Iocono JA, Draus JM
(2012) Primary closure without diversion in manage-
ment of operative blunt duodenal trauma in children.
ISRN Pediatrics 2012:298753
Sola JE, Cheung MC, Yang R, Koslow S, Lanuti E, Seaver
C et al (2009) Pediatric FAST and elevated liver trans-
aminases: an effective screening tool in blunt abdomi-
nal trauma. J Surg Res 157(1):103–107
Stawicki SPA (2017) Trends in nonoperative management
of traumatic injuries – a synopsis. Int J Crit Illn Inj Sci
7(1):38–57
Streck CJ, Lobe TE, Pietsch JB, Lovvorn HN 3rd. (2006)
Laparoscopic repair of traumatic bowel injury in chil-
dren. J Pediatr Surg 41(11):1864–1869
Stylianos S (2000) Evidence-based guidelines for
resource utilization in children with isolated spleen or
liver injury. J Pediatr Surg 35(2):164–169
Stylianos S, Ford HR (2008) Outcomes in pediatric
trauma care. Semin Pediatr Surg 17(2):110–115
Sujenthiran A, Elshout PJ, Veskimae E, MacLennan S,
Yuan Y, Serafetinidis E et  al (2019) Is nonoperative
management the best first-line option for high-grade
renal trauma? A systematic review. Eur Urol Focus
5(2):290–300
Temiz A, Ezer SS, Gedikoğlu M, Serin E, İnce E, Gezer
HÖ et  al (2018) Management of traumatic bile duct
injuries in children. Pediatr Surg Int 34(8):829–836
Upadhyaya P, Simpson JS (1968) Splenic trauma in chil-
dren. Surg Gynecol Obstet 126(4):781–790
Vatankhah S, Sheikhi RA, Heidari M, Moradimajd P
(2018) The relationship between fluid resuscita-
tion and intra-abdominal hypertension in patients
with blunt abdominal trauma. Int J Crit Illn Inj Sci
8(3):149–153
van der Vlies CH, Saltzherr TP, Wilde JC, van Delden
OM, de Haan RJ, Goslings JC (2010) The failure
rate of nonoperative management in children with
splenic or liver injury with contrast blush on com-
puted tomography: a systematic review. J Pediatr Surg
45(5):1044–1049
Wisner DH, Kuppermann N, Cooper A, Menaker J,
Ehrlich P, Kooistra J, et al (2015) Management of chil-
dren with solid organ injuries after blunt torso trauma.
J Trauma Acute Care Surg 79(2):206–14; quiz 332
 Surgical Treatment of Severe
Head Trauma
Hans G. Eder
21.1 Introduction
Head injuries are very common in children and
account for most neurosurgical admissions to
hospitals. More than 1.7 million traumatic brain
injuries occur in adults and children each year in
the United States, with approximately 30%
occurring in children aged less than 14  years
(Morrissey and Fairbrother 2016). The majority
are children with mild head injuries with a
Glasgow coma scale (GCS) score of 13–15, with-
out loss of consciousness or brief loss of con-
sciousness for 5 min or less. About 10% of head
injuries are severe, as defined by a GCS score of
8 or less, and the remaining are moderate head
injuries with a GCS score of 9–13. The mortality
rate of head injuries is five times higher than that
of leukemia, which is the second most common
cause of death in childhood and nearly 18 times
higher than that of brain tumors. Only 20–30% of
children with severe head injury require a neuro-
surgical operation (Albright et  al. 1999; Choux
et al. 1999; Quayle et al. 2014).
Naturally, the etiology of traumatic brain
injury varies according to the age-specific activi-
ties of the child. As such, the most common
causes of head injuries and skull fractures in
small children are accidental falls, whereas inju-
H. G. Eder (*)
Department of Neurosurgery Medical University of
Graz, Graz, Austria
e-mail: hans.eder@medunigraz.at
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_21
21
ries in children older than 5  years are mostly a
consequence of motor vehicle accidents. In some
series, children younger than 2 years of age are
disproportionately affected by brain injuries from
shaken baby syndrome (Nnadi et al. 2014; Joyce
and Huecker 2019).
In general, children have lower mortality and
better functional outcome than adults with simi-
lar injuries. The outcome in adolescents equals
that of young adults, whereas children of elemen-
tary school age do far better than very young
ones.
The primary role of medical therapy is to pre-
vent secondary brain injury from ischemia,
increased intracranial pressure, focal brain com-
pression, and brain edema. This therapy should
begin with the initial resuscitation and continue
throughout the child’s hospital stay (Albright
et al. 1999; Bauer and Kuhn 1997; Choux et al.
1999). Timely resuscitation during early care is
associated with lower in-hospital mortality and
favorable discharge Glasgow outcome scale
score in severe pediatric traumatic brain injury
(Kannan et al. 2018).
21.2 Principles of Treatment
The primary injury occurs within a few millisec-
onds in which the trauma directly damages the
brain, which remains untreatable. Secondary
injuries like brain ischemia are a significant and
261
262
often preventable result of systemic hypotension
or hypoxemia. Early identification and manage-
ment of severe traumatic brain injury are crucial
in decreasing the risk of secondary brain injury
and optimizing the outcome. Therefore, adequate
prehospital care is crucial for these patients to
avoid such secondary injuries to the already
injured brain and begins with the first responder
at the accident scene. Hypoxemia is present in
30–45% of patients with severe head injuries and
even in 15% of patients with moderate head
injuries.
When prehospital hypoxia and hypotension
are present in patients with traumatic brain inju-
ries, the mortality rate is at least doubled.
Conversely, normotensive normoxic patients are
twice as likely to make a good recovery.
In addition to ischemia, increased intracranial
pressure (ICP) due to epidural or subdural hema-
tomas can lead to secondary injuries or focal
swelling of the brain, which often results in her-
niation syndrome without surgical intervention.
Initial clinical assessment, using either the
GCS score or one of the modifications available
for younger patients (children’s coma score), is
most important to identify the extent of head
injury and the neurological status. The GCS score
still remains a critical tool for grading the sever-
ity of neurological injury after traumatic brain
injury and appears to correlate strongly with the
outcome in children (NICE 2014; Nnadi et  al.
2014).
21.3 Management
21.3.1 Mild Head Trauma
Mild head trauma is defined as an injury to the
head with a GCS score of 13–15. Depending on
the age of the child and the time after mild head
trauma, most infants and small children will
arrive alert or with a minor alteration of con-
sciousness. Young children are usually able to
walk and respond well to verbal or painless stim-
uli. Vomiting is common. Neurological findings
are typically normal. Older children complain of
H. G. Eder
posttraumatic amnesia and intense headache.
Clinical examination should include the cervical
spine, chest, abdomen, and limbs looking for
other injuries. Vital signs, such as pulse rate, res-
piration, temperature, and blood pressure, are
recorded.
Most children with relatively mild injuries and
a normal clinical examination can be discharged
from the hospital and observed at home by reli-
able family members (Greenberg et  al. 2014;
Lindholm et al. 2019).
For a child who has suffered the loss of con-
sciousness or amnesia, most neurosurgical
departments will recommend admission for
observation between 1 and 2 days, and children
with a skull fracture should be observed for
around 3  days. Children with an intracranial
lesion are admitted to the hospital for 5–7 days
(NICE 2014).
The appropriate evaluation of a child with an
apparent concussion is debatable. Computed
tomography (CT) scans should be considered in
all children with persistent symptoms or neuro-
logical findings. In the presence of localized
scalp or penetrating injury, unconsciousness, or
significant amnesia for over 10  min, a CT scan
should be obligatory. Follow-up CT for children
with mild or moderate head injuries should be
performed prior to hospital discharge if the initial
CT showed intracranial lesions. However, if pos-
sible, magnetic resonance imaging (MRI) follow-­
ups are recommended to reduce irradiation. A
normal CT scan more than 6 h from the time of
injury is sufficient unless the patient demon-
strates neurological deterioration. An urgent fol-
low-­up CT or MRI should be performed for
neurological deterioration, persistent vomiting,
and seizures. If the initial CT scan reveals a non-
operative hematoma repeat, imaging should be
performed within 24 h to ensure stabilization of
the hematoma size (Holmes et al. 2011).
CT scan is preferable to skull radiography
because it provides more valuable information
regarding anatomic evidence of brain injury and
intracranial hemorrhage or raised intracranial
pressure. Therefore, skull radiography is inade-
quate for assessing intracranial injuries because
21  Surgical Treatment of Severe Head Trauma
normal skull X-ray may occur despite a signifi-
cant intracerebral injury. Skull radiography was
normal in 75% of minor head injury patients
found to have intracranial lesions on CT scans
and was only valuable in the treatment in 0.4–2%
of patients. It is possibly useful for localizing a
penetrating foreign body or in clinical signs sug-
gesting an underlying fracture. Still, the increased
probability of an intracranial injury in case of a
skull fracture finally requires a CT scan.
Skull X-rays are only recommended if a CT
scan is not available. MRI scans in trauma are
usually not appropriate for acute head injuries.
MRI may be helpful after the patient is stabilized
to evaluate brain stem injuries or small white
matter changes in the setting of a diffuse axonal
injury. If the neurological status cannot be
explained by CT scans alone, MRI is indicated
for its higher sensitivity (Beier and Dirks 2014;
Buttram et al. 2015). However, in the acute phase
of trauma, there were no surgical lesions demon-
strated on MRI that were not evident on CT
(Vázquez et al. 2014). An exact clinical examina-
tion before performing CT is most important,
because sedation in very young children for neu-
roimaging may disguise clinical assessment.
Once the child’s clinical course demonstrates
a return to baseline behavior and neurological
function, hospital discharge can be considered.
Mild headaches or attention deficits for several
weeks after injury may be expected. Guidelines
for return to sports or other activities have not
been standardized, but most authors suggest that
the patients should at least be fully asymptomatic
before returning to such activities.
Deterioration of neurological findings of a
child who looks quite well initially includes sev-
eral differential diagnoses such as iatrogenic
sedation effects, seizures, expanding mass
lesions, and brain swelling. Focal seizures espe-
cially followed by focal neurological deficits
should be evaluated with neuroimaging. Early
seizures after head injury are more common in
children than in adults. However, posttraumatic
epilepsy and delayed intellectual impairment
occur with a very low incidence after mild head
injury in children.
263
Contusions, intracranial hemorrhage, or extra-­
axial lesions can appear as a delayed complica-
tion, although the majority of these are visualized
within the first 6 h after injury.
Posttraumatic hyponatremia commonly
appears in moderate-to-severe injuries but has
also been observed after mild head injury in rare
cases. Untreated it can lead to an impairment of
consciousness and seizures due to cerebral
swelling.
An unusual cause of neurological deteriora-
tion after relatively minor trauma is a carotid or
vertebral dissection. Most children with this
diagnosis have focal neck pain after an injury.
Deterioration is due to infarction in the affected
artery distribution. The mechanism of injury is
thought to stretch the artery over the transverse
processes of the upper cervical spine. Initially,
there may be no neurological sequel; however,
progressive thrombosis or embolic events may
develop in a delayed fashion.
21.3.2 Severe Head Injuries
Only 7–10% of head injuries are severe as defined
by a GCS of 8 or less. These patients should be
intubated and ventilated as soon as safely possi-
ble. Intubation in patients with severe head inju-
ries reduces the rate of aspiration and therefore
the risk of pneumonia and helps to prevent respi-
ratory insufficiency. Artificial ventilation should
be adjusted to achieve an arterial saturation
greater than 95%, and pCO2 should be kept
between 30 and 35 mm Hg. Analyses have shown
that the greatest reduction in ICP seems to occur
when the pCO2 is lowered to 35  mm Hg.
Therefore, hyperventilation should only be used
during actual ICP elevations and should not be
lowered to 30 mm Hg. Recent data suggest that
uncritical hyperventilation increases cerebral
ischemia and worsened the patient’s outcome
(NICE 2014).
Early and frequent measurements of the
child’s blood pressure are vital as hypotension is
profoundly harmful to the child with a severe
traumatic brain injury. Treatment of low blood
264
pressure is needed to maintain a cerebral perfu-
sion pressure (CPP) of at least 60  mm Hg. It
should be kept in mind that the most common
cause of hypotension is an extracranial injury.
Adequate sedation and analgesia are essential
in patients with head injuries, especially if venti-
lated. At this point, it should be stated that the
neurological assessment should ideally occur
prior to the administration of these medications.
Up to now, no reliable drugs have been found to
interfere with the molecular secondary injury or
demonstrated any benefit (NICE 2014).
CT of the head and usually upper spine to C3
and plane cervical spine radiographs are usually
obtained. CT of the abdomen is done if there is
any suspicion of abdominal trauma. Other radio-
graphs will depend on the history and the exami-
nation. Approximately 20% of severely
head-injured children have mass lesions from
hematomas that require evacuation.
The need and timing of repeat imaging depend
on the findings on the initial CT scan and con-
tinuous clinical examinations. If the initial CT
scan reveals a nonoperative hematoma, repeat
imaging should be performed within 24  h to
ensure stabilization of the hematoma size. For
stable patients, a follow-up CT or MRI is usually
obtained between 3 and 5  days after the injury
and again between 10 and 14  days. An earlier
MRI is recommended in patients presenting clin-
ical deterioration which cannot be explained by a
CT scan (Vázquez et al. 2014). Delayed imaging
is indicated in children with a diffuse axonal
injury which occasionally results in posttrau-
matic hydrocephalus. External ventricular drain-
age may be necessary, but permanent shunt
placement is rarely required.
21.3.3 Intensive Care Unit
The aim of intensive care management is to mini-
mize the degree of secondary brain injury by a
tailored treatment strategy for each patient based
on continuous monitoring of intracranial pres-
sure, cerebral perfusion pressure, and hemody-
namic parameters. To achieve this optimal care
and treatment, a multidisciplinary team of neuro-
H. G. Eder
surgeons, critical care physicians, pediatricians,
anesthesiologists, and specially trained nursing
staff is required (NICE 2014; Spaite et al. 2014;
Belisle et al. 2018).
All children with a GCS of 8 or less should
receive an ICP monitor to guide the management
of intracranial hypertension as well as to detect
an expanding mass lesion in an early stage. ICP
monitoring by ventricular catheter is still consid-
ered the most accurate and reliable method (He
et al. 2019). Besides ICP measurement, the ven-
tricular catheter allows the therapeutic benefit of
cerebral spinal fluid drainage for ICP control.
Other acceptable methods include parenchymal
fiber-optic and microtransducer systems.
Subarachnoid, subdural, or epidural monitors do
not provide comparable accuracy and are not
recommended.
Delayed rises in ICP are not uncommon
3–10  days after injury. Therefore, a longer
period of observation may be warranted when
the initial diagnosis is an acute subdural hema-
toma or multiple contusions. Intraoperative ICP
monitoring is also generally recommended in
patients with moderate head injuries undergo-
ing general anesthesia for extracranial injuries
(Yang et al. 2014).
The risks of a ventricular catheter for ICP
monitoring are quite small compared to the sig-
nificant damage that could occur in patients with
elevated ICP and include intraparenchymal hem-
orrhage during placement and development of
bacterial ventriculitis (McCarthy et  al. 2010;
Soleman et  al. 2012). The rate of intracerebral
hemorrhage varies from 0 to 2% of placements.
The placement of ventricular catheters should be
avoided in patients with known coagulopathy or
thrombocytopenia. The infection rate is approxi-
mately 7–11%. Ventricular catheters usually need
not be changed routinely before 10  days after
placement. Periodic analysis of cultures and cell
counts are recommended by most trauma centers,
whereas the benefit of prophylactic antibiotics of
an implanted ventricular catheter is uncertain.
Evidence suggests that antibiotic-impregnated
catheters reduce the incidence of shunt infection,
although more well-designed clinical trials test-
ing the effect of antibiotic-impregnated shunts
21  Surgical Treatment of Severe Head Trauma
are required to confirm their benefit. The catheter
should be replaced if there is a positive culture or
increasing pleocytosis in the CSF.
21.4 Surgical Management
of Increased Intracranial
Pressure
Drainage of the Cerebral Spinal Fluid (CSF) is
the most physiological treatment of ICP eleva-
tion, can produce rapid solution of elevated ICP,
and is always worth trying as the initial treatment
rather than more intensive therapy. Either inter-
mittent or continuous drainage can be performed.
In the case of intermittent drainage, the ventricu-
lar catheter is opened when ICP exceeds the
treatment threshold. Another option is continu-
ous ventricular drainage with a ventricular cham-
ber at 5 cm above the level of the lateral ventricle
to avoid sudden ICP fluctuations. Tunneling the
catheter subcutaneously as remotely as possible
from the borehole to minimize a possible ven-
triculitis is recommended (McCarthy et al. 2010;
Soleman et al. 2012).
Besides CSF drainage through a ventricular
catheter, the insertion of a lumbar drain is possi-
ble in selected cases to reduce ICP. However, this
technique is quite controversial and is of unproven
safety and efficacy (Levy et al. 1995; Murad et al.
2012).
Decompressive craniectomy may result in a
decrease of ICP and good outcomes in selected
patients with extremely refractory intracranial
hypertension. However, specific recommenda-
tions or guidelines for this procedure have not
been developed. Aggressive surgical intervention
is recommended only in severely head-injured
children with progressive edema if the brain stem
does not appear to be irreversibly damaged (Beier
and Dirks 2014).
21.5 Skull Fracture
Fractures are diagnosed either by plain radio-
graphs or by CT with the use of bone window
techniques. However, there has been conflicting
265
literature concerning the usefulness of plain
radiographs, because children with a neurologi-
cal deficit or with a GCS less than 13 should have
a CT scan as the examination of choice. If a lin-
ear skull fracture is present, the occurrence of
other lesions such as contusions and epidural
hematomas is greatly increased. CT scan will
detect not only skull fractures but also intracra-
nial lesions. Therefore, if an acutely head-injured
patient is found to have a skull fracture on the
plain radiography, CT scan of the brain should be
obtained to exclude possible intracranial lesions.
Head-injured children with linear skull frac-
tures, normal CT scan of the brain, and normal
neurological examinations do not need any treat-
ments, because these children carry only a low
risk of subsequent neurological deterioration
(Nnadi et al. 2014). These children can usually be
discharged from the hospital after an observation
period of about 72  h. In infants or very young
children, subgaleal hematomas due to subperios-
teal hemorrhage can result in symptomatic blood
loss. Some children require blood transfusion.
21.6 Depressed Skull Fractures
Depressed skull fractures are usually the result of
a very focal impact and represent approximately
one-quarter of all childhood skull fractures. Dural
lacerations are reported in about 10% of patients.
In newborns with a developing very thin skull,
the under-surface of the depression is usually
quite smooth resulting in the so-called ping-pong
fractures. Other common fractures are greenstick
fractures where the bone remains connected and
is fixed in its new abnormal position. Three-­
quarters of depressed skull fractures are located
in frontal or parietal bones.
Location, degree of depression, and integrity
of the scalp are important for the assessment of
depressed skull fracture. If the depression of frac-
tures is at least as deep as the thickness of the
skull, surgical elevation should be considered,
because fractures depressed more than 5  mm
commonly tear the dura (Fig. 21.1a, b). In case of
obvious compression of the brain or if CSF or
brain tissue come through a laceration over the
266
a b
Fig. 21.1 (a) 8-year-old boy with laceration of the skin,
CSF coming through the wound, and a CT scan (bone
window technique) revealing a frontal left depressed skull
fracture, the fracture should be elevated and any
dural tear repaired. Fractures without brain com-
pression or tear of the dura have an excellent out-
come with conservative therapy alone. Especially
in the young infant, it has been suggested that
there is no need for elevation because the skull
will remodel itself (Albright et al. 1999; Choux
et al. 1999).
In some children, the elevation of the skull
fracture is necessary for cosmetic reasons, e.g., if
the fracture is located on the forehead. This can
be performed on an elective basis in the case of
complete integrity of the scalp. More important
than prophylactic antibiotics in open skull frac-
tures is the early closure of the wound. If perma-
nent closure is delayed for more than 48  h, the
incidence of infection increases from approxi-
mately 5–37%. Regarding seizures, there are no
clinical data that support elevating depressed
skull fractures for seizure prophylaxis. Several
investigators have concluded that if an epilepto-
genic focus develops, it occurs at the time of
injury and not from depressed skull fracture
(Albright et al. 1999; Choux et al. 1999).
Depressed bone fractures involving the dural
sinus present a very high risk of death and mor-
H. G. Eder
fracture. (b) After reconstruction of the bone fragments
with absorbable sutures and closure of the dura
bidity to the patient undergoing surgery because
uncontrollable bleeding might occur and result in
hypovolemic shock. The surgeon should keep in
mind that the anterior one-third of the superior
sagittal sinus tolerates legation without neuro-
logical deficits, whereas the posterior two-thirds
do not.
Surgical techniques used for the elevation of
depressed skull fractures vary from a simple
borehole near the fracture to a more extensive
craniotomy. In newborns, it might be possible to
use a small borehole to lift the depressed frag-
ment. In older children with more complex frac-
tures, we prefer a more extensive craniotomy to
have safe access to any dural and cortical injury
(Albright et  al. 1999; Bauer and Kuhn 1997;
Choux et al. 1999).
Basilar skull fractures appear to occur more
commonly in adults than in children. CSF otor-
rhea carries the same diagnostic importance as
does CSF rhinorrhea, and it is only present in
10%. If the tympanic membrane remains intact,
hemotympanum will be observed. The natural
history of traumatic CSF leaks is that almost all
will cease spontaneously with conservative
management. CSF rhinorrhea is rare in children
21  Surgical Treatment of Severe Head Trauma
with basilar skull fractures, which sometimes
requires lumbar CSF drainage. Another option
is the detection of the CSF leak by fluoroscopy
with the following endoscopical sealing of the
leak. The use of prophylactic antibiotics is con-
troversial. The only one randomized prospec-
tive study has not proven any beneficial effect
of prophylactic antibiotics based on either the
incidence or the severity of posttraumatic
meningitis.
21.7 Leptomeningeal Cysts
Diastatic fractures to the skull in a growing child
can in less than 1% result in the formation of a
leptomeningeal cyst or growing skull fractures.
The leptomeningeal cyst is believed to develop
due to a tear in the dura with subsequent enlarge-
ment of the bone caused by brain pulsation. CT
or MR images will show both brain and menin-
ges within the defect which avoids the healing of
the bone (Fig. 21.2a, b). The hallmark of a grow-
ing fracture is a palpable nontender swelling in
the area of a previous skull fracture. Surgical
treatment includes water-tight repair or replace-
ment of the dura and closure of the skull defect
a b
Fig. 21.2 (a) Skull film 8 months after linear skull fracture revealing enlargement of the bone defect. (b) MRI confirms
the diagnosis of growing skull fracture, showing brain and meninges within the bone defect
267
whenever possible with the autogenous skull
bone. Children under the age of 3  years with a
linear or diastatic skull fracture need a follow-up
skull film at 6 months post injury to detect a pos-
sible growing skull fracture early enough (Prasad
et al. 2015).
21.8 Penetrating Cerebral Injuries
CT scan remains the preferred initial imaging
modality of penetrating cerebral injuries. In
selected cases, plane films may help to delineate
skull fractures and intracranial foreign bodies;
however, CT scan alone is generally sufficient.
Magnetic resonance imaging is not indicated in the
early evaluation period. In younger children, the
side of injury is often around the face, especially
through the orbita. In cases of deep head penetra-
tion by sharp instruments and location of the for-
eign body to the major cerebral vessels, intradural
and extradural exploration by craniotomy will
maximize the comfort of safe removal of the for-
eign body. Rapid exploration should be possible if
bleeding is to be expected during the removal of the
foreign body (Choux et  al. 1999). Every effort
should be made to remove the entire foreign body
268
and to close the dura. Short courses of prophylactic
antibiotics are generally indicated.
Gunshot wounds to the brain are operated on
only if there is a chance of survival and recovery
(Bizhan et al. 2015). GCS score of 3–5 combined
with nonreactive pupils or refractory hypotension
has a mortality rate up to 100%. The decision to
withhold surgery seems warranted because
­surgery will not alter the outcome. However, the
decision to perform surgery or not is often com-
plex. The higher post-resuscitative GCS score of
5 is an indication for debridement of the wound
and closure of the dura and the eventual place-
ment of a monitoring device for ICP. There is no
evidence that extensive parenchymal debride-
ment or removal of deeply penetrating fragments
of bullets or bone does prevent infection. Delayed
complications include brain abscess, traumatic
aneurysm formation, and wandering intracranial
fragments (Rosenfeld et al. 2014).
21.9 Mass Lesions After Head
Injury
Until proven otherwise, a child with an altered
level of consciousness, pupillary dysfunction,
and laterally extremity weakness should be sus-
pected of having a mass lesion that may require
surgery.
21.10 Epidural Hematomas
Approximately 60% of patients with epidural
hematomas (EDHs) are below the age of 20.
EDHs present differently from one age group to
the other. Neonates and infants present with a
decrease in hematocrit, bradycardia, pallor,
change in the level of consciousness, bulging
fontanelles, and focal neurological deficits
(Fig. 21.3). These patients may be neurologically
normal or have mild neurological abnormalities
and then rapidly decompensate within hours or
days after injury. The more common venous eti-
ology accompanied with large cerebral spinal
fluid spaces and open fontanelles allows a longer
compensatory phase. In older children, EDHs
H. G. Eder
Fig. 21.3  Acute parietal EDH in a 6-month-old infant 4 h
after falling 50 cm from the bench. Child presented ini-
tially without deficit but at admission with vomiting, pal-
lor, and bulging fontanelles. Surgery due to visible mass
lesion and increased intracranial pressure is indicated
usually originate from arterial bleeding which
reduces the ability to compensate and lead to a
more rapid progression of events. The lucid inter-
val is shorter than in neonates. Another peculiar-
ity of EDHs in children is their quite common
location in the posterior fossa; however, the pari-
etal and temporal regions present the most likely
location (Irie et al. 2011).
Not all EDHs are considered emergencies and
need surgical intervention. There are some chil-
dren with EDHs who are asymptomatic or mini-
mally symptomatic with a relatively small
hematoma without compression to the brain. In
these patients, a nonsurgical approach is proba-
bly justified (Khan et  al. 2014). Some of these
hematomas show spontaneous regression fol-
lowed by absorption within 4–6  weeks post
injury. However, all patients with acute EDHs
should be admitted to the intensive care unit for
observation (Teichert et al. 2012).
Appropriate preoperative planning is required
in patients who are considered surgical candi-
dates because of the low incidence of finding the
mass lesion only with the use of multiple explor-
atory boreholes. Therefore, there is rarely a rea-
son for blind surgery unless a CT scan is
21  Surgical Treatment of Severe Head Trauma
unavailable and the child is deteriorating. In this
case, boreholes are performed where the fracture
is seen on radiography and the location of which
is compatible with the clinical picture of a possi-
ble EDH.  Otherwise, the bone flap should be
large enough to entirely expose and evacuate the
blood clot. In children who were awake after
trauma and then deteriorate to a GSC less than 8,
ICP monitoring may not be required after rapid
clot removal. On the other hand, with children
who have been unconscious from the time of
injury, ICP monitoring should be placed after clot
removal to allow appropriate treatment of the dif-
fuse brain injury. In these children, EDHs may be
associated with parenchymal injuries that are
contributing to the coma. If preoperative CT only
reveals the EDH, opening of the dura should be
avoided to minimize the risk of injury to the
underlying cortex.
a b
Fig. 21.4 (a) MRI reveals subdural effusions over the
right hemisphere due to shaken baby syndrome in a
2-month-old infant. Patient presented with signs of
269
21.11 Subdural Hematomas
Neonatal subdural hematomas (SDHs) are
uncommon and can be caused by birth trauma
related to vacuum extraction or forceps injury. In
general, SDHs large enough to require surgery
are also rare in children. SDHs in infants and
young children are mostly associated with shaken
baby syndrome (Fig.  21.4a, b). These children
are admitted with significant neurological injury
and do not show much evidence of external dam-
age. Every child who has minimal external signs
of trauma with neurological damage, acute SDH,
and retinal hemorrhages should be assumed to
have been abused until proved otherwise (Mian
et al. 2015).
SDHs are usually described as acute, sub-
acute, or chronic in order to relate them to the
time of initial hemorrhage. The etiology gener-
increased intracranial pressure. (b) MRI 1  month after
subdural drainage for 5  days through a borehole on the
right side
270
ally involves bridging veins from the cortical sur-
face to the various venous sinuses that are
disrupted secondary to force.
Acute SDHs usually present with sudden
deterioration and compression of the brain or cer-
ebellum and appear within few hours after the
injury. Patients present with varying decreases of
loss of consciousness, pupillary dilatation, cra-
nial nerve deficits, motor deficits, vomiting, and
headaches.
Subacute SDHs develop days following the
initial hemorrhage. During these days, more than
one hemorrhage event may occur as the hema-
toma enlarges and causes rebleeding at the origi-
nal site.
Three weeks to several months later, the SDH
becomes a chronic SDH. This appearance often
has multiple layers because of membrane forma-
tion and rebleeding. Clinical symptoms in a
chronic SDH are typically nonlocalizing and
associated with vomiting, irritability, failure to
thrive, and seizures. Chronic SDHs occur more
frequently in children than acute SDHs.
SDHs commonly occur in the posterior fossa
of neonates as the bridging veins of the cerebel-
lum might be damaged during suction and
forceps-­assisted birth. Observation of posterior
fossa SDH is the treatment of choice if there are
no signs of brain stem compression or obstructive
hydrocephalus.
The clinical presentation of an acute SDH in
young infants is generally more diffuse. They
appear pale with a full fontanelle and may or may
not have a focal neurological deficit. In children
with an open fontanelle, subdural collections are
tapped only when intracranial pressure is ele-
vated. Perinatal SDHs may rather be managed
conservatively, especially small ones because
they may resolve within a few days. Acute SDHs
are more commonly seen in older children. They
are often associated with multiple trauma, cere-
bral contusions, and skull fractures. Patients pre-
senting with impaired consciousness, sign of
acute rise in ICP, and focal deficits need emer-
gency surgical intervention. Large craniotomy
extending close to the midline allows easier visu-
alization of the bleeding site along the sagittal or
sigmoid sinus. The goal of hemostasis is to find
H. G. Eder
the cortical or bridging vein that has been torn by
the trauma. Sometimes, acute SDHs may lead to
diffuse edema and extensive swelling of the brain
which make it necessary to leave the dura open
for decompression of the brain. Intracranial pres-
sure monitors are recommended for the treatment
of subsequent cerebral swelling and in patients
with a GCS score of 8 or less before operation.
The outcome of patients with large acute SDH
and GCS score less than 8 generally is unsatisfac-
tory. Prompt intervention in this setting may sig-
nificantly improve patients’ outcome. Patients
with minimal neurological deficits and a small
SDH can be observed. Small SDHs in association
with diffuse brain injury or brain swelling are
also rarely operated on since these hematomas
play little or no role in the etiology of the coma.
Chronic SDHs are more common in infants
than neonates. Unrecognized trauma may be the
cause of these SDHs. These hemorrhages may
often be managed nonsurgically because the
infant cranium expands which reduces the intra-
cranial pressure and small SDHs may resolve
spontaneously. Larger chronic SDHs may need
surgical intervention because of repeated bleed-
ing of the fragile vessels of neo-membranes. The
infants usually present with signs of a slow
increase in ICP including nausea, vomiting, poor
appetite, and irritability. Seizures may be seen in
about 40% of these children (Albright et al. 1999;
Choux et al. 1999).
21.12 Intracerebral Hematomas
Intracerebral hematomas (ICHs) in neonates are
believed to be related to the movement of the
bony edges of the cranium against each other
during delivery. Bleeding can also occur due to
arteriovenous malformations, hypoxia at birth,
arterial emboli, or infarction. In older children,
ICHs occur more likely from large angular accel-
eration–deceleration forces during injury.
Characteristic findings of this trauma are small
hemorrhages in the deep white matter, corpus
callosum, and brain stem (diffuse axonal injury).
The hallmark symptom of diffuse axonal injuries
is immediate unconsciousness. Surgical interven-
21  Surgical Treatment of Severe Head Trauma
tion in this setting is usually not indicated, and
the patient will be treated conservatively
(Albright et al. 1999; Choux et al. 1999).
Focal contusions after head traumas are usu-
ally limited to the cortex, but they may expand
into the white matter (White et  al. 2009). It is
rare that contusion is considered a surgical lesion
in children as the tissue can often recover. If the
child is neurologically intact or has only mini-
mal deficits, ICHs can be carefully observed. A
large hematoma associated with significant mid-
line shift and deteriorating patient requires cra-
niotomy. A safe route to the hematoma should be
taken to remove hemorrhage as much as possi-
ble. Areas that may be safely operated on with-
out a clear neurological sequel are the frontal
and anterior temporal lobes in the nondominant
hemisphere.
21.13 Conclusion and Future
Direction
Although the majority of pediatric head injuries
are mild and have no serious long-term sequel, it
is important to identify children with severe head
injuries. The primary goal in treating any pediat-
ric patient with severe traumatic brain injury is
the prevention of secondary insults such as
hypotension, hypoxia, and cerebral edema.
Despite the publication of guidelines, significant
variations in the treatment of severe traumatic
brain injury continue to exist, especially in
regard to intracranial pressure-guided therapy.
After prompt resuscitation and clinical evalua-
tion, optimal management of intracranial hyper-
tension under multimodal monitoring has to be
done. The approach to the management of severe
traumatic brain injury based on the published
guidelines should be focused on ICP control,
which should ultimately improve cerebral perfu-
sion pressure. ICP monitoring by ventricular
catheter is considered the most accurate method
with the advantage of lowering the ICP by CSF
drainage. Cerebral perfusion pressure of at least
40–50 mm Hg in infants and 50–60 mm Hg in
children is probably reasonable goal. Depending
on clinical assessment and morphology, surgical
271
interventions may be necessary. Predictors of
poor outcome for accidental injuries include
multiple trauma and early hypoxia or shock.
Other factors include low GCS scores. Mortality
rates in children remain high, with 40–60% for
children with a GCS score of 3 or 4. Children
with a GCS score of 5 or above have a mortality
rate of less than 10%.
The developing brain in children absorbs the
forces of traumatic impact leading to more dif-
fuse injury and fewer mass lesions. Most mass
lesions in children can be managed conserva-
tively. However, mass lesions with neurological
deficits, deterioration of the neurological status,
and radiographic evidence of brain compression
require emergent surgical intervention. Linear
nondepressed skull fractures do not require any
treatment. Most CSF fistulas in children due to
skull base fractures cease spontaneously. The
usefulness of prophylactic antibiotics remains
unclear. A substantial number of neonatal
depressed skull fractures reform spontaneously
during the first few weeks after birth. Depressed
fractures should be elevated in case of dural lac-
eration or obvious compression of the brain.
Growing skull fractures as a complication of
large linear skull fractures usually associated
with dural laceration always require surgical
correction and can be detected on physical
examination.
Optimal care of the infant or child with a
severe traumatic brain injury requires a multidis-
ciplinary team in each phase of management.
Treatment starts with the resuscitation at the
scene of accident, extends over the neurointen-
sive care management including surgery, and
ends with sufficient rehabilitation.
References
Albright AL, Pollack IF, Adelson PD (1999) Principles
and practice of pediatric neurosurgery. Thieme,
New York
Bauer BL, Kuhn TJ (1997) Severe head injuries: pathol-
ogy, diagnosis and treatment. Springer, Berlin/
Heidelberg
Beier AD, Dirks PB (2014) Pediatric brainstem hem-
orrhages after traumatic brain injury. J Neurosurg
Pediatr 14(4):421–424
272
Belisle S, Lim R, Hochstadter E et  al (2018) Approach
to pediatric traumatic brain injury in the emergency
department. Curr Pediatr Rev 14(1):4–8
Bizhan A, Mossop C, Aarabi JA (2015) Surgical manage-
ment of civilian gunshot wounds to the head. Handb
Clin Neurol 127:181–193
Buttram SD, Garcia-Filion P, Miller J et  al (2015)
Computed tomography vs magnetic resonance imag-
ing for identifying acute lesions in pediatric traumatic
brain injury. Hosp Pediatr 5(2):79–84
Choux M, Di Rocco C, Hockey A, Walker M (1999)
Pediatric neurosurgery. Churchill Livingstone, London
Greenberg JK, Stoev IT, Park TS et al (2014) Management
of children with mild traumatic brain injury and
intracranial hemorrhage. J Trauma Acute Care Surg
76(4):1089–1095
He J, Chen J, Wu T et al (2019) The value of managing
severe traumatic brain injury during the periopera-
tive period using intracranial pressure monitoring. J
Craniofac Surg 30(7):2217–2223
Holmes JF, Borgialli DA, Nadel FM et  al (2011) Do
children with blunt head trauma and normal cranial
computed tomography scan results require hospital-
ization for neurologic observation? Ann Emerg Med
58(4):315–322
Irie F, Le Brocque R, Kenardy J et al (2011) Epidemiology
of traumatic epidural hematoma in young age. J
Trauma 71(4):847–853
Joyce T, Huecker MR (2019) Pediatric abusive head
trauma (shaken baby syndrome). In: StatPearls
[Internet] Treasure Island: StatPearls Publishing
Kannan N, Wang J, Mink RB et al (2018) Timely hemody-
namic resuscitation and outcomes in severe pediatric
traumatic brain injury: preliminary findings. Pediatr
Emerg Care 34(5):325–329
Khan MB, Riaz M, Javed G (2014) Conservative man-
agement of significant supratentorial epidural
hematomas in pediatric patients. Childs Nerv Syst
30(7):1249–1253
Levy DI, Rekate HL, Cherny WB et al (1995) Controlled
lumbar drainage in pediatric head injury. J Neurosurg
83(3):453–460
Lindholm EB, D’Cruz R, Fajardo R et al (2019) Admission
of pediatric concussion injury patients: is it necessary?
J Surg Res 244:107–110
McCarthy PJ, Patil S, Conrad SA, Scott LK (2010)
International and specialty trends in the use of prophy-
lactic antibiotics to prevent infectious complications
after insertion of external ventricular drainage devices.
Neurocrit Care 12(2):220–224
Mian M, Shah J, Dalpiaz A, et  al (2015) Shaken baby
syndrome: a review. Fetal Pediatr Pathol 34(3):169–
H. G. Eder
75. https://doi.org/10.3109/15513815.2014.999394.
[Epub ahead of print]
Morrissey K, Fairbrother H (2016) Severe traumatic brain
injury in children: an evidence-based review of emer-
gency department management. Pediatr Emerg Med
Pract 13(10):1–28. Epub 2016 Oct 2
Murad A, Ghostine S, Colohan AR (2012) A case for fur-
ther investigating the use of controlled lumbar cere-
brospinal fluid drainage for the control of intracranial
pressure. World Neurosurg 77(1):160–165
National Institute for Health and Care Excellence (NICE).
Head injury: triage, assessment, investigation and early
management of head injury in children, young people
and adults. NICE clinical guideline 176, January 2014
Nnadi MO, Bankole OB, Fente BG (2014) Epidemiology
and treatment outcome of head injury in children: a
prospective study. J Pediatr Neurosci 9(3):237–241
Prasad GL, Gupta DK, Mahapatra AK et  al (2015)
Surgical results of growing skull fractures in children:
a single centre study of 43 cases. Childs Nerv Syst
31(2):269–277
Quayle KS, Powell EC, Mahajan P et  al (2014)
Epidemiology of blunt head trauma in children
in U.S. emergency departments. N Engl J Med
371(20):1945–1947
Rosenfeld JV, Bell RS, Armonda R (2014) Current con-
cepts in penetrating and blast injury to the central ner-
vous system. World J Surg 39(6):1352–1362. https://
doi.org/10.1007/s00268-­014-­2874-­7. [Epub ahead of
print]
Soleman J, Marbacher S, Fandino J, Fathi AR (2012) Is
the use of antibiotic-impregnated external ventricular
drainage beneficial in the management of iatrogenic
ventriculitis? Acta Neurochir 154(1):161–164
Spaite DW, Bobrow BJ, Stolz U et al (2014) Evaluation
of the impact of implementing the emergency medical
services traumatic brain injury guidelines in Arizona:
the Excellence in Prehospital Injury Care (EPIC)
study methodology. Acad Emerg Med 21(7):818–830
Teichert JH, Rosales PR Jr, Lopes PB et  al (2012)
Extradural hematoma in children: case series of 33
patients. Pediatr Neurosurg 48(4):216–220
Vázquez E, Delgado I, Sánchez-Montanez A et al (2014)
Imaging abusive head trauma: why use both computed
tomography and magnetic resonance imaging? Pediatr
Radiol 44(Suppl 4):S589–S603
White CL, Griffith S, Caron JL (2009) Early progression
of traumatic cerebral contusions: characterization and
risk factors. J Trauma 67(3):508–14; discussion 514–5
Yang C, Li Q, Wu C et al (2014) Surgical treatment of trau-
matic multiple intracranial hematomas. Neurosciences
(Riyadh) 19(4):306–311
 Pediatric Orthopedic Trauma
22
Zacharias Zachariou, Eva E. Fischerauer,
and Annelie M. Weinberg

22.1 Introduction in the slightest represent the above-mentioned

aims of a modern fracture regime.
In contrast to adults, where the aim of the treat- Epidemiologically, the number of child inju-
ment is a quick rehabilitation in order to return to ries has increased most probably due to the
work, in children, the aim has never been defined. increased participation of youths in organized
Due to the inexorable growth of a child, fractures sports. The development of new sports equipment
heal very quickly so that the aim should actually and sport arts make these activities available to
be to not disturb this healing process in a child-­ even younger children leading to injuries that
appropriate environment. This should exclude older children were incurring before. This equip-
long hospitalization, above all short immobiliza- ment enables children to play at higher speeds
tion in unworthy positions, such as extensions, resulting in more complex incurred fractures.
and in no case with constant therapy changes, In addition, the diagnostic apparatus has
such as re-repositions under anesthesia, etc. improved enormously providing better preopera-
The ultimate aim is to provide a primary tive imaging that enables a much better decision
definitive therapy, which would allow a short regarding the necessity for surgical treatment and if
hospital stay as well as a quick recovery enabling so, to choose the appropriate method of treatment.
the child to return to a child suitable life. The new classification and “code of proce-
For many years, the so-called “conservative dure” contributes to the differentiation of the cor-
therapy” in the treatment of fractures in children rect indication for fracture treatment applying the
represented the standard that, however, does not age-appropriate therapy and not simply consider-
ing the child as a “small adult.” The consideration
of the healing behavior of fractures in children is
essential, especially in specific injuries of the
Z. Zachariou (*)
growth plate, the joints, as well as special frac-
Department of Pediatric Surgery, Makarios Children’s
Hospital, Nicosia, Cyprus Medical School, University tures like the greenstick and bowing fracture.
of Cyprus, Nicosia, Cyprus
e-mail: zzach@ucy.ac.cy
E. E. Fischerauer 22.2 Brief Historical Overview
University Clinic of Paediatric and Adolesecent
Surgery, Medical University, Graz, Austria
Findings from the stone age revealed that early
A. M. Weinberg humans splinted fractures to relieve pain and to
Department of Pediatric Traumatology, Johannes
allow a better function once the fracture had
Gutenberg University, Mainz, Germany

© Springer Nature Switzerland AG 2023 273

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_22
274
Fig. 22.1  Return of
Hephaestus to Olympus
on the horse with clump
feet. Ionic hydria, ca.
525 BC. Vienna,
Museum of Art History,
Collection of Classical
Antiquities
healed. The first evidence of manuscripts on the
treatment of fractures goes back to 1600  BC in
Egypt on a papyrus (Woods and Woods 2000).
The “Corpus Hippocraticum,” written around
430 BC in ancient Greece, devotes many sections
not only to the diagnosis and treatment of bone
fractures and dislocations but also to congenital
anomalies, such as clubfoot (Fig. 22.1), applying
a principle valid till today in orthopedics—trac-
tion and countertraction. The word orthopedics
consists of the word “ortho,” meaning correct,
and “paedi,” meaning child, referring to the cor-
rection of child deformities. The teachings of
Hippocrates were furthered by Galen, a Greek
physician, during the Roman era (129–199 AC),
who is referred to as the father of sports medicine
as he devoted his professional life to the treat-
ment of gladiators and athletes. In the Dark Ages,
orthopedics suffered, as did all medical practice,
due to the superstitions and religious bans on
people dealing with the human body.
It was not until the fifteenth and sixteenth cen-
turies when orthopedics advanced beyond the
description of diseases and an understanding of
the pathophysiology of the disease began.
Physicians like Ambroise Parė, Clopton Haves,
Albrecht von Haller, and Nicolas Andry contrib-
uted to the development of orthopedics (Beckett
1999).
Z. Zachariou et al.
However, looking at the above history, exclud-
ing the ancient Greek era, children were not of
interest to doctors. In 1765, Jean-André Venel
created a hospital for the treatment of crippled
children in a systematic fashion in Switzerland.
With the beginning of industrialization, at the end
of the nineteenth century, injuries among factory
workers increased creating a need for orthopedic
treatment. This development was intensified dur-
ing World Wars I and II, with the introduction of
osteosynthesis including plates, screws nails, and
intramedullary pinning (Rang 2000).
It was not until the 1980s that the new meth-
ods of operative correction of fractures in chil-
dren were introduced, overthrowing the generally
accepted conception until then that child frac-
tures should be treated conservatively.
22.3 Incidence
The incidence of trauma in children cannot be
defined, in general, as the conditions in which
children live are variable in different regions. In
this chapter, we will refer to Europe only. While
death and disability due to trauma in the overall
population are about 15%, in children, this num-
ber rises up to 50%, making trauma the leading
cause (Chan et al. 1989).
22  Pediatric Orthopedic Trauma
The majority of injuries in children are fortu-
nately minor, usually due to falls affecting a sin-
gle extremity, usually upper. Only about 12% are
admitted with serious injuries including the poly-
traumatized (Acierno et al. 2004). Although the
real incidence of traumatic injuries in childhood
is not exactly evaluated, it is a fact that injuries in
children increase as children increasingly interact
with the adult world. The distribution is age-­
dependent, with nonaccidental trauma in children
under the age of 12 months, traffic (as pedestri-
ans) and home injuries in the 2–6-year-old age
group as this group spends their time in and
around the home, play injuries in the group of
7–14 years old, and traffic injuries in the group of
adolescents as they are active cyclists and
motorists.
22.4 Pathophysiology
The presence of growth plates and the thick peri-
osteum represent the most obvious anatomic dif-
ference in the pediatric skeleton compared to
adults. The growth plates are located between the
Fig. 22.2  Anatomy of the growth plate in children
275
epiphysis and metaphysis in a child’s long bones
and appear as dark lines on X-rays (Fig. 22.2).
Injuries of the growth plate and epiphysis can
result in severe growth disturbances with the mal-
function of the extremities. These injuries require
anatomic reduction just as intraarticular injuries
in adults. The main reason for a rapid fracture
healing in children is the thicker periosteum,
which is not only more active but also tears less
readily, and not around the entire circumference
of the bone, possessing a much greater osteo-
genic potential. It also helps to reduce the amount
of displacement and contributes to the reduction
as well as the maintenance of the reduction by
quicker callus formation. Each part of the devel-
oping bone has its own characteristic injuries,
growth, and remodeling patterns.
As long as the epiphysis is completely carti-
laginous, it is protected from injury, and only
when bone is formed within the epiphysis, it is
more likely to break.
The metaphysis possesses a thinner cortical
area and increased trabecular bone, whereas the
porosity is greater than in the diaphysis. The
metaphyseal region is subjected to more bone
276
remodeling after fracture and transforms back
into a trumpet-shaped metaphyseal cortex with
longitudinal growth.
The diaphysis is extremely vascular in the
newborn, becoming less vascular as the child
grows, whereas cortical bone thickens, resulting
in a periosteum-mediated membranous bone for-
mation and growth.
The fracture healing process is divided into
three phases:
1. Inflammation.
2. Reparative phase.
3. Remodeling.
Inflammation describes the initial phase after
the loss of the osseous structure integrity. The
hematoma, which contains a lot of fibrin and
inflammatory mediators, fills the fracture gap,
attracting mesenchymal cells that differentiate
into chondroblasts, osteoblasts, and angioblasts.
This process closely resembles the stages of
physiological bone development.
In the reparative phase, the hard callus is
formed by periosteal cells that directly differenti-
ate into osteoblasts, inducing bone formation by
intramembranous ossification. In the center of the
fracture gap, chondrocytes form the soft callus,
which serves as a scaffold for woven bone formed
by endochondral ossification. Together, the hard
and the soft callus represent the so-called ‘provi-
sional callus’ that provides temporary stabiliza-
tion. Rigid stability, which is necessary for full
physical activity, occurs when lamellar bone is
laid down along the lines of stress and replaces
Table 22.1  Bone fracture classification according to location, displacement, and stability
Localization
Stability Diaphyseal/metaphyseal
Sufficiently stable Transverse fractures, with
for initial retention tolerance limits depending
on age
or
Oblique/spiral fractures of
one bone of the lower leg
or forearm
Unstable fractures All fully displaced
fractures
Z. Zachariou et al.
the provisional callus. This remodeling phase can
last months, even years in certain osseous
structures.
22.5 Classification
22.5.1 General Classification
In addition to the severity of the injury and the
fracture pattern, which are the parameters to be
considered in adult fractures, the component of
growth is added in pediatric fractures. Different
classifications were introduced over the years;
however, these were very specific, lacking a uni-
versal application, and were not scientifically
validated.
These numerous systems for classifying bone
fractures in children were defined according to:
• Localization, displacement, and stability
(Table 22.1).
• Localization and involvement of the growth
plate (Table 22.2).
• The Salter–Harris system (Salter and Harris
1963), based on the radiographic appearance
of the fracture in relation to the growth plate
and the Aitken system (Fig. 22.3, Table 22.3).
The detailed definition of the Salter–Harris
classification is as follows:
• Type I: This fracture is presented by a com-
plete separation of the epiphysis and physis
from the metaphysis. The fracture line
Articular Therapy
Nondisplaced or minimally Immobilization with plaster in
(<2 mm) displaced articular combination with cast wedging if
fractures necessary
Articular fractures with a Reduction under anesthesia with
gap >2 mm! either conservative (plaster) or
operative stabilization
22  Pediatric Orthopedic Trauma 277

Table 22.2  Classification according to location and growth plate involvement

Shaft Diaphyseal Stable Nondisplaced fractures without shortening
fracture Unstable Displaced fractures with shortening or having the tendency for
shortening
Greenstick Bowing fractures with complete fracture of one cortex and
incomplete fracture of the cortex of the contralateral side
Metaphyseal Buckle Compression of the metaphyseal cortex of one side
Incomplete Greenstick fracture in the metaphysis
Complete Complete separation including the periost
Lig. avulsion Ligament avulsion
Articular Epiphyseal Aitken I Salter–Harris See Table 22.3
fracture I + II
Aitken II + III Salter– See Table 22.3
Harris III + IV
Tillaux or bi-plane In puberty by partially closed growth plate
fracture
Flake fracture “Normally” in combination with joint dislocation
Lig. avulsion Bony or cartilage avulsion

1 2 3 4
Humerus Radius/ulna Femur Tibia/fibula

E = Epiphysis

1 = Proximal M = Metaphysis

Subtrochanteric line

2 = Diaphyseal D = Diaphysis

M = Metaphysis
3 = Distal

E = Epiphysis

Fig. 22.3  Fracture location related to bone segments and subsegments. For children, the defined square must be placed
over the larger parts of the growth plate as indicated

goes  through the hypertrophic zone of the
growth plate.
• Type II: This type describes a fracture which,
in addition to type I, includes a metaphyseal
fragment on the compression side of the frac-
ture, called the “Thurston-Holland sign.”
• Type III: This epiphyseal injury is composed
of a physeal separation and a fracture through
the epiphysis to the joint.
• Type IV: describes a fracture through the
metaphysis, physis, and epiphysis into the
joint.
278
Table 22.3  Salter–Harris/Aitken classification
I Epiphyseal separation
II Epiphysiolysis with metaphyseal wedge
III Epiphyseal fracture
Epiphyseal fracture with metaphyseal
IV
wedge
• Type V: is classified as a compression or
crushing injury to the growth plate and occurs
infrequently. On the initial radiograph, it can
be mistaken as a type I fracture, as its radio-
logical appearance is similar. However, this
fracture is associated with severe soft-tissue
injuries including vascular lesions and is due
to a considerable axial load.
22.5.2 Specific Pediatric Fractures
22.5.2.1 Diaphyseal Fractures
Greenstick Fracture
This fracture occurs when bones are bent
beyond their limits. It is due to a failure of the
tension side as the compression side bends. As
the fracture is incomplete, the remaining bone
undergoes plastic deformation. These fractures
are characterized by the possibility of a refrac-
ture, which is related to a consolidation failure
of the convex site. The risk of a refracture can
last up to 8  months post injury. This time
depends on the localization, as the more distal,
the shorter the time of possible refracture, and
on the patient’s age—the younger, the shorter
the time. Nevertheless, complete closure and
realignment of the fracture can only be
Z. Zachariou et al.
I
II
III
achieved by completing the fracture or over-
correcting the angulation (Atanelov and
Bentley 2020).
Toddler’s Fracture
The isolated tibial fracture during childhood
(until the age of 4) is named the “toddler’s frac-
ture.” The typical feature of this fracture is that
the child is unable to step on the foot and conse-
quently walk properly, thus limping. A radiologi-
cal diagnosis is often impossible. During
check-ups, the fracture can be secondarily diag-
nosed by a radiologically detectable callus for-
mation and decalcification of the fracture gap
around the tenth day post trauma. As the fracture
cannot be diagnosed at once, the child’s leg is
normally immobilized by an upper leg cast. The
treatment should be prolonged for another
10 days after verification of the toddler’s fracture
10–14 days later (Hogue et al. 2019).
22.5.2.2 Metaphyseal Fractures
Buckle Fracture
This is the most frequent fracture in children and
is due to a compression failure of the bones. It is
located in the metaphysis, as this part is of great-
est porosity, particularly in younger children
(Fig. 22.4).
22  Pediatric Orthopedic Trauma 279

The fracture location is related to the four

long bones and the specific pediatric subseg-
ments. The identification of the segments differs
from the adult classification, as for pediatric
long-bone fractures, the end segments have two
subsegments:

• Segment 1: Proximal—including epiphysis

(E) and metaphysic (M) subsegments.
• Segment 2: Diaphysis (D).
• Segment 3: Distal—including metaphysic (M)
and epiphysis (E) subsegments.

The overall structure of pediatric fracture

Fig. 22.4  Buckle fracture
classification is listed in Table 22.4 and the rele-
vant pediatric fracture patterns, expressed into a
Incomplete Fracture
“child code,” are grouped according to each frac-
This fracture is characterized by the plastic defor-
ture location category E, M, or D (Table  22.5,
mation of the bone as it bends, upon applying
Fig. 22.3).
increasing longitudinal force along the length of
the bone. A child’s bone can tolerate a greater
Guidelines for Correct Classification
degree of deformation than that of an adult before
The classification system for children is defined
it breaks. This fracture can occur without any
according to diaphysis (D), metaphysis (M), and
classical signs of a fracture, e.g., fracture gap. It
epiphysis (E). For easier and more accurate clas-
can only be diagnosed by documenting a radio-
sification, defined squares are copied to a trans-
logically visible bending/curve of the bone. At
parent application of the squares that are overlaid
the point where the applied force exceeds the
on the a.p. radiographic view. The proximal
structural bone integrity, a complete fracture
femur is an exception as metaphyseal fractures
occurs, something that is rarely comminuted in
are located between the growth plate of the head
children due to the greater flexibility of a child’s
and the intertrochanteric line and the defined
bone.
square does not apply.
Epiphyseal fractures include the known epiph-
22.5.2.3 AO Pediatric Comprehensive
yseal injuries Salter–Harris I–IV classified as
Classification of Long-Bone
child codes E/1 to E/4. Other child codes (E/5 to
Fractures (PCCF)
E/9) are used to identify Tillaux (two-plane) frac-
In 2007, the AO Pediatric Comprehensive
tures (E/5), tri-plane fractures (E/6), ligament
Classification of Long-Bone Fractures (PCCF)
avulsions (E/7), and flake fractures (E/8).
was introduced based on the principle that any
Metaphyseal fractures are categorized in three
classification of pediatric fractures must be appli-
patterns, e.g., buckle/torus or greenstick fractures
cable for all fractures and recognize the impor-
(M/2), complete fractures (M/3), and osteo-­
tance of growth through the epiphyseal plate (Agel
ligamentous, musculo-ligamentous avulsion, or
2018). This classification was validated in differ-
just avulsion injuries (M/7).
ent studies and should be applied for all pediatric
Diaphyseal fractures include bowing fractures
fractures (Slongo et al. 2006; Slongo et al. 2007a;
(D/1), greenstick fractures (D/2), complete trans-
Slongo et al. 2007b). It is available to all surgeons
verse fractures (angle ≤30°; D/4), complete
as a software package at www.aofoundation.org,
oblique/spiral fractures (angle >30°; D/5),
as well as on the app AO-Trauma.
Monteggia (D/6), and Galeazzi lesions (D/7).
280 Z. Zachariou et al.

Table 22.4  Overall classification system of pediatric fracture classification

Diagnosis
Localization Morphology
Bone Segment – Subsegment / Child Severity Displacement
1234 123 EMD 1–9 0.1 0.2 I–IV

Table 22.5  Classification of pediatric fractures (Agel 2018)

E = Epiphysis
E/7
E/1 E/4
S–H I S–H IV Ligament
avulsions
E/5
E/2 Tillaux (two E/8
S–H II plane) fractures Flake fractures

E/6
E/3 E/9
S–H III Tri- plane
fractures Other fractures

M = Metaphysis
M/2
Incomplete fracture M/3 M/7
(Torus / Buckle Complete Ligament
or greenstick)
fracture avulsion

M/9 Other fractures

D = Diaphysis
D/4
D/1 Complete D/6
Bowing transverse Monteggia
fractures fractures lesions
<30°
D/5
D/2 Complete D/7
Greenstick oblique/ Galeazzi
fractures spiral fracture lesions
>30°
D/9 Other fractures

The grade of fracture severity distinguishes
between simple (noted as 0.1) and wedge (par-
tially unstable fractures with three fragments
including a fully separated fragment) or complex
fractures, e.g., totally unstable fractures with
more than three fragments (noted as 0.2)
(Fig. 22.5). Severity implies anticipated difficul-
ties and methods of treatment, not the prognosis.
For the following specific fractures, the clas-
sification is as follows:
• Supracondylar humeral fractures (code
13-M/3) are given an additional code regarding
the grade of displacement at four levels (I–IV).
• Radial head fractures (code 21-M/2 or /3, or
21-E/1 or /2) are given an additional code (I–
22  Pediatric Orthopedic Trauma
a b
Fig. 22.5  Fracture severity: (a) simple (noted as 0.1); (b) partially unstable or totally unstable fractures with more than
three fragments (noted as 0.2)
III) regarding the axial deviation and level of
displacement (I  =  no angulation and no dis-
placement, II = angulation with displacement
that is less than half the bone diameter, and
III  =  angulation with displacement that is
more than half the bone diameter).
For the paired bones radius/ulna and tibia/fib-
ula, both bones must be included in the defined
square (Fig.  22.3). Except for the known
Monteggia and Galeazzi lesions, the paired bone
fractures are classified with a single code, with
the severity code describing the worst of the two
fractures.
When a single bone is fractured, a small letter
describing that bone (e.g., “r,” “u,” “t” or “f”)
should be added after the segment code (e.g., the
code “22u” identifies an isolated diaphyseal frac-
ture of the ulna).
When paired bones are fractured, each with a
different child pattern (e.g., complete fracture of
the radius and a bowing fracture of the ulna),
each bone must be coded separately including
the appropriate small letter (22r-D/5.1 and
22u-D/1.1).
The malleolar fractures, which are uncommon
in children, are simply coded as distal tibia frac-
tures (e.g., the fracture of the medial malleolus is
a typical Salter–Harris III or IV fracture of the
distal tibia, coded as 43).
Ligament avulsions (intraarticular and extraar-
ticular) are epiphyseal and metaphyseal injuries,
respectively. The side of ligament-avulsion frac-
tures of the distal humerus and of the distal femur
is indicated by the small letter “u” (ulnar/medial)
281
or “r” (radial/lateral) for the humerus and by “t”
(tibial/medial) or “f” (fibula/lateral) for the femur.
Femoral neck fractures can be differentiated
as epiphysiolysis or epiphysiolysis with a
metaphyseal wedge and are coded as normal
type-E epiphyseal Salter–Harris I and II frac-
tures, E/1 and E/2. Fractures of the femoral neck
are coded as normal type-M metaphyseal frac-
tures coded from I to III.
22.6 Diagnosis
The diagnosis of a fracture in a child can be chal-
lenging as, in most cases, the trauma mechanism
is not clear and depends on the age of the child
and/or the presence of the caretaker at the time of
injury. Usually, the injury can be diagnosed by
observing the posttraumatic swelling and the
pain at the site of the lesion. Approximately two-­
thirds of injuries concern the upper limb and
about one-third concerns the lower limb.
In contrast to the principle of clinical exami-
nation, manual palpation is not necessary, espe-
cially in children as pain is induced causing the
child to cry, without the possibility to diagnose a
fracture or observe its progressive course.
An initial X-ray in two planes, an ap and a lat-
eral projection, represents the golden standard. In
cases of an obvious defective position, a one
plane-X-ray is initially justifiable. Oblique X-ray
imaging is sometimes indicated, especially when
the upper ankle or the knee are involved. Computer
tomography (CT) or magnetic resonance imaging
(MRI) is rarely indicated as progressive primary
282
examination for the diagnosis of a fracture in chil-
dren, excluding facial and head injuries.
Ultrasonography applying “point-of-care
ultrasound” has been promoted in recent years;
however, studies showed that although the speci-
ficity rate was high, the clinician has to be focused
on training. The accuracy was highest at the
diaphysis of long bones, while it was less depend-
able for compound injuries and fractures adjacent
to joints, lesions of the small bones of the hand
and foot, nondisplaced epiphyseal fractures, and
those with a fracture line of less than 1  mm.
Point-of-care ultrasound may serve as a rapid
alternative means to diagnose midshaft fractures
(Weinberg et al. 2010; Hübner et al. 2000).
Table 22.6  Suggested radiological investigations
Shoulder
Shoulder a.p. (under traction on both arms, 15 kg
comparison of both sides)
Shoulder in glenoid tangential projection
Y-view (acromion, coracoid, and the tangential view of the
scapula together form the leg of the “Y”)
Shoulder transthoracal
Target X-ray
Elbow joint
Elbow a.p. and lateral
Elbow under imaging intensifier
Hand
Hand dorso-palmar and lateral
Special view of the scaphoid (scaphoid quartet)
Pelvis
Pelvis normal a.p.
Hip joint axial
Lauenstein X-ray of the ala of the ileum
Obturator X-ray
Ankle joint
A stress X-ray should not be taken. MRI shows ligament avulsions much better; however, the diagnosis is of
academic value since ligament avulsion is no longer treated surgically
a.p. and lateral view
Foot
Foot a.p.
Foot in dorso-plantar oblique projection
Foot in plantar–dorsal oblique projection (prone position of Medial tarsal bones (cuneiform medial and
the child)
Z. Zachariou et al.
It is obsolete to take X-rays of the opposite
side in order to detect the plane of the growth
plate.
Fractures that are treated conservatively with
cast treatment and risk dislocation should be con-
trolled by X-ray examination on days 5–8 after
trauma.
In elderly children, when no deformity can be
accepted and the potential of spontaneous correc-
tion is diminished, an X-ray examination after
14 days is advisable.
A guide to radiological investigation of the
extremities is given in Table 22.6.
Dislocation of the acromioclavicular joint
Overlapping free view of the humeral head and the
glenoid joint (“true” a.p. view)
Side difference indicates an injury of the acromion
Dislocation of the shoulder (needs a second view)
Subcapital fractures of the humerus (second view
needed)
For special indications (suspicion of a Bankart lesion
or Hill–Sachs lesion)
Fractures
For special indications (radial head, coronoid process)
Fractures and dislocations, swellings
Fractures
Fractures
Fractures
Fractures of the ileum, the anterior wall
Fractures of the anterior acetabular column, pubis,
and the posterior acetabular wall
Fractures, joint incongruence
Fractures
Fractures
intermedium)
22  Pediatric Orthopedic Trauma
22.7 Management
It is impossible to cover the whole spectrum of
the management of pediatric fractures in a single
chapter; thus, the author provides an overview of
the principles of pediatric fracture treatments as
well as the principles of the treatment methods.
In addition, the treatment of specific pediatric
fractures is reported.
The first step in the therapy of fractures is the
decision if a reduction is indicated or not.
Table 22.7 lists these criteria.
Fractures in children can be treated either con-
servatively or surgically. Equally which method
has used the principles are the same: restoration
of bony anatomy, bone union, and recovery of
optimal physical and psychologic function.
22.7.1 Therapy Principles
Nowadays, in children, most fractures are closed
reduced even when osteosynthesis has to be per-
formed. Joint fractures should be demonstrated
(open or by arthroscopy), anatomically reduced,
and stabilized. However, open reductions are
rarely indicated during childhood.
22.7.1.1 Conservative Methods
Closed reduction is mainly used for the forearm
of upper extremities and should be carried out
under anesthesia (plexus block or general anes-
Table 22.7  Criteria for reduction
No reduction
• All stable, un-dislocated/un-displaced fractures of the
upper and lower extremity independent of age.
• Stable buckle fractures or metaphyseal fractures
within the tolerated limits.
• Unstable humeral fractures of initial tolerated angular
deformity.
• All clavicular fractures <12 years.
Reduction
• All stable and unstable fractures that neither meet the
age-dependent criteria nor the localization-­dependent
criteria for conservative treatment without reduction.
• Fractures of the lower arm, as angular deformities
can lead to functional impairment.
• In fractures of the lower extremity, it is important to
obtain normal length, axis, and rotation.
283
thesia). The reduction is performed applying a
maneuver that mimics the movement that led to
the fracture. Thereafter, immobilization with a
dorso-volar plaster splint is applied.
Semi-open reduction is indicated in partial or
total unstable displaced fractures that can be
reduced definitively. After closed reduction under
sterile conditions, percutaneous K-wire fixation
is performed with additional immobilization with
a plaster splint.
Traction (overhead traction) used to be a stan-
dard method years ago; however, nowadays, the
indication is limited to fractures of the femur in
children between birth and 3, maximum 4 years
of age, depending on the child’s weight. The
plaster is first stuck on the full length of the
healthy leg. The plaster on the injured leg is
placed only distal to the fracture. Both legs are
fixed on the overhead arch so that the child’s but-
tocks are raised; so that a flat hand can move
freely under the buttocks without touching them.
22.7.1.2 Surgical Methods
During growth, a successful fracture osteosyn-
thesis is based on comprehensive knowledge
about different kinds of osteosynthesis tech-
niques and their appropriate use in children. For
many years, the screw, plate, and tension band
wiring osteosynthesis, which is performed ana-
logically to adults, has been displaced in opera-
tive pediatric orthopedic trauma treatment and is
now mainly performed in children over 12. In
younger children, in most cases, K-wire osteo-
synthesis, elastic-stable intramedullary nailing,
cannulated screw osteosynthesis, or external fix-
ator is performed.
Osteosynthesis with K-wires is indicated in
cases of:
• closed or open reduced metaphyseal
fractures,
• fractures of the hand and foot.
K-wires enable a mere adaptation providing
maximal stability to movement but not weight-­
bearing. Percutaneous placement is preferable
as it enables removal without anesthesia. The
crossing points of the K-wires should be prox-
284 Z. Zachariou et al.

imal to the fracture line and penetrate the External fixator osteosynthesis is indicated
opposite cortex. In cases where the epiphysis for:
must be crossed, thin K-wires reduce epiphy-
sis damage and repeated attempts should be • comminuted fractures mainly in older chil-
avoided. This osteosynthesis requires addi- dren (femur, tibia, forearm),
tional cast immobilization. Diaphyseal frac- • polytrauma,
tures are not suitable for K-wire application • long spiral fractures, e.g., spiral wedge of the
(Fig. 22.6). femur in older children.

Fig. 22.6 (a) Dislocated distal forearm fracture (ap and lateral view); (b) typical K-wire fixation for stabilization of the
reduced fracture
22  Pediatric Orthopedic Trauma
a b
Fig. 22.7 (a) Open distal (supracondylar) humerus fracture, before and after debridement and jet lavage, (b) osteosyn-
thesis with and external fixator, and (c) clinical situation after surgery
In most cases, a closed reduction with or with-
out traction table is performed followed by
­application of any external fixator system avail-
able under imaging intensifier. After the final
reduction, all clamps have to be closed and
secured.
The external fixator is a supplementary alter-
native method to the ESIN and is used especially
in very unstable fractures of an open angle of 3°
and in older children (Fig. 22.7).
Elastic stable intramedullary nailing (ESIN) is
indicated in transverse, oblique and short spiral,
diaphyseal fractures in children between the 3
and 15  years of age. It is minimally invasive
while providing sufficient stability for move-
ment, as well as partial weight-bearing. As this
method represents a closed treatment procedure,
it is only possible to achieve an axis-compatible
position and no anatomical reduction.
The biomechanical principle of elastic-stable
intramedullary nailing is based on the symmetric
tension of two metaphyseal-inserted nails which
have three intramedullary supporting points.
Optimal results are warranted by correctly
symmetric-­positioned nails (Fig. 22.8).
285
c
Lag screw osteosynthesis is indicated in cases
of articular and peri-articular fractures where,
additionally, the bone fragments are under com-
pression. The bone can be stabilized better,
although a cast is also required in most cases.
Furthermore, these screws are excellent for closed
fixations, especially at the lower limb (Fig. 22.9).
Plate osteosynthesis should be reserved to
exceptions and in special cases. The indication is
in cases of comminuted fractures, mainly in older
children (femur, tibia, and forearm), and long spi-
ral fractures with/without a spiral wedge of the
femur in older children (Fig.  22.10). However,
ESIN can be an alternative.
22.7.2 Conservative Treatment
The main aim of conservative treatment is to
diminish the acute pain and minimize the risk of
a secondary dislocation until the full restoration.
The most common conservative treatment of
fractures is plaster immobilization; something
that, in my opinion, every physician who deals
with fractures should know how to apply. A cast
286 Z. Zachariou et al.

a b

Fig. 22.8 (a) Subcapital proximal humerus fracture and (b) osteosynthesis with ascending elastic-stable intramedul-
lary nails

a b

Fig. 22.9 (a) Dislocated radial condyle fracture of the elbow and (b) intraoperative cannulated screw fixation (ap and
lateral view)
22  Pediatric Orthopedic Trauma
Fig. 22.10  Plate osteosynthesis
reduces the possibility of secondary displacement
and ensures healing by stabilizing the fracture in a
reduced position. Stabilization can be provided by
individual, appropriate kinds of casts.
Plaster splint is indicated in:
• stable fractures,
• distortions,
• pain after bone bruises.
Nowadays, prefabricated fiberglass splints are
available, which are very user-friendly and enable
circular closure after edema regression.
Circular plaster cast is indicated in:
• unstable fractures,
• functional treatment,
• secondary postplaster splint (after edema
regression).
287
A primarily applied circular plaster should
be cut longitudinally, especially when the tis-
sue is still swollen directly after trauma. After
the edema subsides, the plaster can be closed
again.
Plaster cast windows are indicated in cases of:
• open wounds,
• pins/K-wires.
Any window in the plaster must not impair the
stability and is created after the plaster is dry. The
window must always be covered with the piece
that was taken out in order to prevent edema in
this region.
Plaster cast wedging is indicated in cases of
remaining angulation of an undisplaced, stable
fracture, or after fracture stabilization without
reduction. This can be performed after the edema
and pain have regressed.
The more peripheral the fracture, the more
proximal the wedge should be, placed at the
deepest point of the concavity of the deformity.
22.7.3 Specific Pediatric Fracture
treatments
22.7.3.1 Monteggia Fracture
A Monteggia lesion combines a fracture of the
proximal or middle third of the ulnar diaphysis
with a dislocation of the radial head. Although
this injury can be detected easily by X-ray, it is
still frequently missed as the ulna fracture
impresses. Therefore, every ulnar fracture
requires radiographs of the elbow joint. In all
views, a line through the long axis of the radius
should pass through the humerus capitellum.
These lesions are treated by anatomical reduc-
tion and stabilization of the ulnar fracture and
correction of the radial head dislocation
(Fig.  22.11). In case of a greenstick or buckle
fracture of the ulna, conservative treatment
(upper arm-cast-longuette in supination position
of the lower arm and 90 degrees flexion in the
elbow joint) can be performed.
288
a b b1
Fig. 22.11  Monteggia lesion. (a) Diaphyseal ulnar fracture and radial head dislocation. (b) Intramedullary nailing (b1
ap, b2 lateral view)
22.7.3.2 Supracondylar Humeral
Fracture
This special fracture is the most frequent one dur-
ing childhood and especially concerns children
between 3 and 10 years old. In 98% of cases, it
represents an extension type fracture; in 2%, a
flexion type fracture occurs.
According to Gartland, extension type frac-
tures are classified as:
• Type I: Nondisplaced fracture.
• Type II: Minimally displaced fracture with
intact posterior cortex.
• Type III: Completely displaced fracture with
no cortical contact.
Type I, undisplaced stable fractures, with at
least partially intact periosteum and without sec-
ondary injuries, can be treated conservatively by
a dorsal upper arm-cast-longuette in neutral
Z. Zachariou et al.
b2
p­ osition of the lower arm and 90 degrees flexion
in the elbow joint or a cuff and collar sling.
Type II, displaced supracondylar fractures can
be clinically distinguished from the undisplaced
fractures by the elbow deformation and the func-
tio laesa (loss of function) and should be consid-
ered as an emergency. The neurovascular status
has to be examined because side injuries, such as
neurological defects, may occur. Radiologically,
extensional fractures display a dorsal displace-
ment of the distal fragment, whereas flexional
fractures show a ventral dislocation of the bone
fragment.
Type III, displaced unstable supracondylar
fractures with torn periosteum require closed
anatomical reduction and are mostly stabilized
with K-wires or intramedullary nailing osteosyn-
theses (Fig. 22.12). In case the closed reduction
is impossible, an open reduction and stabilization
are indicated.
22  Pediatric Orthopedic Trauma 289

a b

c d

Fig. 22.12  Supracondylar humerus fracture. (a) Supracondylar fracture, (b) K-wire osteosynthesis, (c) intramedullary
nailing, and (d) cubitus valgus because of displaced supracondylar humerus fracture

22.7.3.3 Fracture of the Lateral
Condyle
Milch’s classification is most frequently used for
lateral condyle fractures:
• Type I: the lateral trochlear ridge is attached to
the main portion of the ridge.
• Type II: the lateral trochlear ridge is attached
to the displaced fragment.
Lateral condylar fractures represent Salter
type IV epiphyseal injuries of the distal humerus.
Undisplaced fractures can be overlooked, espe-
cially in the case of hairline fractures, which can
be missed on radiographs, or in very young chil-
dren, whose ossification center is very small
leading to a misunderstanding of the nature of the
injury. Clinically, the elbow displays signs of
traumatization including pain and swelling, and a
cubitus valgus which inevitably leads to ulnar
nerve palsy.
In contrast, displaced fractures of the lateral
condyle are often accompanied by lateral sublux-
ation of the ulna. Undisplaced fractures of the
lateral condyle appear horizontal and are moder-
ately stable due to an intact cartilage hinge.
290
a b
Fig. 22.13  Fracture of the lateral condyle. (a) Lateral condyle fracture; (b) K-wire osteosynthesis
Normally, undisplaced fractures of the lateral
condyle are treated by a long arm cast with the
elbow at 90° for 3–4 weeks or by K-wire osteo-
synthesis. In the case of cast treatment, X-ray
examinations are indicated every 3–4 days for the
first 2 weeks in order to recognize secondary dis-
placement. Most undisplaced fractures heal well
with cast immobilization, whereas displaced
fractures require immediate open reduction and
internal fixation to achieve excellent results
(Fig. 22.13).
The complications of this fracture are mal-
union, resulting in stiffness and cubitus valgus,
nonunion due to immobilization, synovial fluid
bathing the fracture, or soft-tissue interposition,
and growth arrest due to an accompanying growth
plate lesion.
Z. Zachariou et al.
22.7.3.4 Metaphyseal Fracture
of the Proximal Tibia
All metaphyseal tibial greenstick fractures are
related to a temporary partial stimulation of the
medial part of the proximal tibial growth plate,
consequently leading to a valgus deformation. In
most cases, a slight valgus position is already
preexisting. An untreated gapping fracture is
associated with a delayed consolidation, conse-
quently leading to increased stimulation of the
medial physis. An enhanced medial growth inten-
sifies the valgus deformation, resulting in a uni-
lateral genu valgum. In order to avoid an
exorbitant valgus deformation, it is important to
correct the primary valgus position by compress-
ing the medial (convex) fracture gap, either by
wedging the cast or closed reduction and medial
22  Pediatric Orthopedic Trauma
a b
Fig. 22.14  Typical medial metaphyseal fracture of the tibia. (a) Metaphyseal greenstick fracture; (b) typical valgus
deformity
external fixator application. However, in young
patients, the valgus deformation normally cor-
rects spontaneously during following growth
(Fig. 22.14).
22.7.3.5 Fracture of the Medial
Malleolus
This fracture represents an intraarticular epiphy-
seal fracture of the distal tibia. It can occur with
(Salter IV) or without (Salter III) an additional
metaphyseal fragment. In the case of a wide-open
291
growth plate, a premature partial growth arrest
occurs in 20%, consequently resulting in a varus
deformation. This growth disturbance is espe-
cially associated with a displaced fracture.
However, in most cases, the growth disturbance
is not very distinctive due to the fact that frac-
tures of the medial malleolus occur beyond the
age of 10. Nowadays, undisplaced fractures are
treated conservatively, while displaced fractures
are treated by K-wire or screw osteosynthesis
(Fig. 22.15).
292 Z. Zachariou et al.

a b c

Fig. 22.15  Fracture of the medial malleolus. (a) Fracture of the medial malleolus (ap and lateral view), (b) screw
osteosynthesis (ap and lateral view), and (c) K-wire osteosynthesis (ap and lateral view)

22.8 Complications

Although complications after fractures are simi-

lar in children as in adults, there are some specific
complications that only occur in children, caused
by the fast fracture healing and the growth poten-
tial that adults do not possess. These include
malalignment due to the natural corrections by
growth (remodeling), physeal arrest, overgrowth,
and refracture.
Malalignment with late angulation is a com-
mon problem, especially in fractures of the proxi-
mal tibial metaphysis. Remodeling of
posttraumatic uncorrected angular deviations,
also referred to as spontaneous correction, leads
to a posttraumatic bone length discrepancy, espe-
cially important in upper and lower leg fractures.
Conversely, length discrepancies caused by
remodeling of angular deviations of the upper
limb are not of the same importance. Spontaneous Fig. 22.16  Partial physeal arrest with bony bar formation
improvement is possible so that a “wait and see”
attitude should be adopted for at least 18 months influenced by any treatment, and consequently
to 2  years, to be confident that the maximal cannot be avoided. Supposedly, traumatic inju-
improvement has occurred (Wilkins 2005). ries of the soft tissues, severe contusions, or
Physeal arrest is a unique complication of metaphyseal fractures and operations, which
physeal injuries due to the interruption of physeal interrupt the blood flow to the growth plate, are
growth, leading to a range of growth distur- the etiological causes and may lead to the forma-
bances, from a complete arrest with no further tion of transient physeal bone bridges (Fig. 22.16),
growth to a partial arrest with progressive angula- possibly resulting in premature ossification of the
tion. Growth arrest represents the worst compli- growth plate or even cartilage necrosis.
cation during fracture repair. This growth Furthermore, premature closure of the growth
disturbance occurs spontaneously and cannot be plate can be a result of radiation, burning, infec-
22  Pediatric Orthopedic Trauma
tion, and long immobilization, due to cast treat-
ment that may even force the child to stay in bed
for a long period.
Clinically, it is important to know that every
trans-epiphyseal lesion that occurs either post-
traumatically or due to the use of physeal-­
crossing implants is associated with the formation
of physeal bone bridges, which could disrupt the
physiological growth. However, the final length
alteration and the corresponding growth progno-
sis are dependent on the child’s age at the time of
bone trauma.
The overgrowth phenomenon, according to
von Laer, represents a posttraumatic stimulation
of one or more growth plates in its overall size
and must be expected in every fractured growing
bone. Maximum stimulation occurs during the
child’s growth years and the least in the prema-
ture age, as the stimulation leads to premature
physis closure compensating overgrowth. If the
trauma takes place in the time of growth plate
closure, the process is accelerated and results in
bone shortening. This commonly means that
under the age of 10, a growth stimulation with
increased bone length is likely, whereas children
over 10 probably have to deal with bone shorten-
ing. Clinically, it is important to avoid difficult
reduction maneuvers but to look for the best pos-
sible alignment of the fracture ends.
Refracture is a complication more often in
young boys and is the result of therapy misman-
agement. Bone healing in children is rapid and
immobilization often is discontinued earlier than
requested. The conservative reduction of a
refracted bone is difficult so that surgical correc-
tion may be necessary as an increase in deformity
is more probable.
22.9 Conclusions
The main difference that influences therapy of
pediatric musculoskeletal trauma is the fact that
the pediatric skeleton matures rapidly by growing
in length and width, changing shape, alignment,
and rotation.
Although most injuries are simple fractures
and conservative therapy is in most cases suffi-
293
cient, the diagnosis can be missed leading to a
catastrophic result, especially if the growth plate
is involved.
The advantage of rapid healing in childhood
due to the extremely osteogenic periosteum could
lure the treating physician to reduce immobiliza-
tion with the consequence of refracture and other
complications.
Anatomic alignment should be the primary
concern, even if a certain degree of angulation is
acceptable when treating fractures in children, it
is best to keep the amount of angulation as small
as possible whatever the patient’s age.
As physeal involvement seems to be the weak-
est and most vulnerable point regarding the ther-
apy of fractures in children, it is mandatory to
treat the growth plate as gently as possible, which
when destroyed inevitably will end in disruption
of longitudinal growth of the bone with func-
tional impairment. The treatment of physeal inju-
ries is based on the severity of the injury, its
anatomic location, and the age of the patient.
Equally of the fracture type, a gentle reduction is
mandatory in children avoiding physeal damage
applying preferably the definitive treatment from
the beginning.
Orthopedics is a field influenced by evolving
technologies in radiology, metallurgy, computer
technology, and medical research. Despite this
evolution, the reputation of orthopedic surgeons
as bonesetters will remain.
Acknowledgments  I would like to thank the authors of
this chapter in the previous edition of this book Eva
E. Fischerauer and Annelie M. Weinberg for enabling the
usage of the following figures: 17.2, 17.3a–d, 17.4, 17.5,
17.6, 17.7, 17.8, as well as Table 17.1.
References
Acierno SP, Jurkovich GJ, Nathens AB (2004) Patterns of
emergent operative intervention in the injured child. J
Trauma 56:969–964
AOTrauma International Board representatives: James
F. Kellam MD, and Eric G. Meinberg MD Orthopaedic
Trauma Association representatives: Julie Agel
MA, ATC, Matthew D.  Karam, MD, and Craig
S.  Roberts MD, MBA AO.  Pediatric Comprehensive
Classification of Long Bone Fractures (PCCF). J
294
Orthop Trauma 2018; 32:S117–S140. https://doi.
org/10.1097/BOT.0000000000001065
Atanelov Z, Bentley TP (2020) Greenstick fracture.
StatPearls, Treasure Island, Florida
Beckett D (1999) From bonesetters to orthopaedic sur-
geons: a history of the specialty of orthopaedics. Surg
Technol:7–10
Chan BSH, Waker PJ, Cass DT (1989) Urban trauma:
an analysis of 1116 pediatric cases. J Trauma
29:1540–1547
Hogue GD, Wilkins KE, Kim IS (2019) Management
of Pediatric Tibial Shaft Fractures. Am Acad Orthop
Surg 27(20):769–778
Hübner U, Schlicht W, Outzen S, Barthel M, Halsband
H (2000) Ultrasound in the diagnosis of fractures in
children. JBJS 82-B(8):1170–1173
Rang M (2000) The story of orthopaedics. Saunders,
Philadelphia
Salter RB, Harris WR (1963) Injuries involving the epiph-
yseal plate. JBJS 45(3):587–622
Slongo T, Audigé L, Schlickewei W et  al (2006)
Development and validation of the AO pediatric
Z. Zachariou et al.
comprehensive classification of long bone fractures
by the Pediatric Expert Group of the AO Foundation
in collaboration with AO Clinical Investigation and
Documentation and the International Association for
Pediatric Traumatology. J Pediatr Orthop 26(1):43–49
Slongo T, Audigé L, Clavert JM et  al (2007a) The AO
comprehensive classification of pediatric long-bone
fractures: a web-based multicenter agreement study. J
Pediatr Orthop 27(2):171–180
Slongo T, Audigé L, Lutz N et al (2007b) Documentation
of fracture severity with the AO classification of pedi-
atric long-bone fractures. Acta Orthop 78(2):247–253
Weinberg ER, Tunik MG, Tsung JW (2010) Accuracy of
clinician-performed point-of-care ultrasound for the
diagnosis of fractures in children and young adults.
Injury 41(8):862–868
Woods M, Woods MB (2000) Ancient medicine from sor-
cery to surgery. MN Runestone Press, Minneapolis
Wilkins KE (2005) Principles of fracture remodeling in
children. Injury 36(Suppl 1):A3–A11
 Injuries to the Tendons
of the Hand
Georg Singer and Heidi Friedrich
23.1 Introduction
The hand represents the most frequently injured
part of the body in children. The annual rate of
hand injuries in children has been estimated to be
11.6 per 1000 population. Within these injuries, a
small amount may involve flexor or extensor ten-
dons. If misdiagnosed or not treated correctly,
these tendon lacerations may lead to lifelong dis-
abilities. However, children represent a demand-
ing group of patients with unique challenges
regarding diagnosis, treatment, and postoperative
rehabilitation (Lee and Schreiber 2015).
23.2 History
In the second century, Galen described the repair
of flexor tendons and nerves in the forearm, and
reported that repair of these similar-appearing
structures led to uncontrollable convulsions
(Kleinert et al. 1995). Further development took
place in the early sixteenth century when one dif-
ferentiated between nerves and tendons as, e.g.,
depicted by Leonardo da Vinci, Andreas Vesalius,
G. Singer (*) · H. Friedrich
University Clinic of Pediatric and Adolescent Surgery,
Medical University of Graz, Graz, Austria
e-mail: georg.singer@medunigraz.at
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_23
23
and Rembrandt. The breakthrough, however, was
achieved in the twentieth century with the advent
of improved suturing material. Horsehair, linen,
cotton, and silk threads used until then only
allowed poor gliding of the repaired tendons in
the tendon sheath, such that as late as 1916, it was
recommended to amputate a finger with a flexor
tendon injury so as not to subsequently impair the
fingers or the whole hand.
In 1948, Bunnell termed the zone between the
distal palmar crease and the insertion of the flexor
digitorum superficialis (FDS) tendon “no man’s
land” (Newmeyer 3rd and Manske 2004). Both
Verdan and Bunnell recommended tendon graft-
ing for repair in this zone. In 1961, however,
Claude Verdan came to the conclusion that pri-
mary repair of injured flexor tendons was not
only possible but also advisable (Verdan and
Michon 1961). Six years later—at the 22nd
annual meeting of the American Society of the
Surgery of the Hand in San Francisco—Harold
Kleinert reported a 10 year series of primary ten-
don repairs, coupled with an aggressive program
of postoperative rehabilitation using dynamic
rubber band splinting. This regimen, which was
published in 1973, produced results that were
previously unobtainable (Kleinert et al. 1973).
Regarding suturing techniques, Kirchmayr—a
surgeon from Vienna—first described a “grasp-
ing” method of tendon suturing (Kirchmayr 1917).
His technique is, in different variations, still used
today. Other suturing techniques published by, for
295
296 G. Singer and H. Friedrich

instance, Kessler, Tajima, and Tsuge in the 1970s,

also still remain popular (Kleinert et al. 1995).
Nevertheless, research focusing on optimiza-
tion material as well as suturing techniques and
the ideal postoperative rehabilitation programs is
ongoing, and there remain controversies and
open questions that have to be addressed.

23.3 Flexor Tendon Injuries

23.3.1 Anatomy

Exact knowledge of the anatomy of flexor ten-

dons is of pivotal importance to correctly diag-
nose and treat flexor tendon injuries. Originating
from muscle bellies of the palmar side of the fore-
arm four tendons of the M. flexor digitorum
superficialis (FDS) run superficially from four
tendons of the M. flexor digitorum profundus
(FDP). At the level of the proximal phalanges, the
FDS tendon splits into two separate cords forming
the chiasma tendineum to let the tendon of the
FDP pass. The FDS tendon inserts at the middle
phalanx and therefore bends the metacarpopha-
langeal (MCP) and the proximal interphalangeal
(PIP) joint. The tendon of the FDP inserts at the
distal phalanx and therefore additionally bends
the distal interphalangeal (DIP) joint. The thumb,
however, is flexed by the M. flexor pollicis longus
inserted at the distal phalanx and the M. extensor
pollicis brevis inserted at the base of the proximal
phalanx, and does not have a chiasma tendineum.
On the fingers, the tendons are held in place
by a pulley system consisting of four annular
ligaments (A1–A4) and three cruciate ligaments
(C1–C3). Damage to annular ligaments, notably
A2 (over the proximal phalanx) and A4 (over the
middle phalanx), leads to the “bow-string phe-
nomenon,” i.e., a protrusion of the tendon result-
ing in a reduced glide range and a reduced
bending at the interphalangeal (IP) joints.
To a large extent, vascularization of the ten-
dons is achieved by the vincula tendinea, soft-­
tissue sail-like structures containing arteries and
veins (Fig. 23.1). If they are severed by the injury,
the tendons may retract back into the palm. If the
vincula are preserved, the proximal tendon stumps
can be found within the tendon sheath canal. Fig. 23.1  Position of the vincula tendinea
23  Injuries to the Tendons of the Hand
23.3.2 Epidemiology
Flexor tendon injuries are rare injuries in children
and adolescents. Nietosvaara and coworkers have
reported an incidence of 3.6 per 100.000 children
(Nietosvaara et  al. 2007). Generally, there is a
male preponderance with peaks in children aged
2-3  years and adolescents between 13 and
15 years old (Singer et al. 2017). In younger chil-
dren, the predominant mechanisms of injury are
falls while carrying a glass or falls into glass
panes. Older children, however, hurt themselves
when playing with knives. Annual outbreaks of
pediatric flexor tendon injuries have been
described. In North America and in several coun-
tries in Europe, children carve pumpkins at
Halloween. In Saudi Arabia and other Muslim
countries, it is tradition to slaughter sheep at
home during the Hajj ceremony (Al-Qattan
2014). During these periods, a rise in flexor ten-
don injuries can be observed.
23.3.3 Zones
The most commonly used classification system
categorizing pediatric flexor tendon injuries is
the one published by Verdan (Fig. 23.2) (Verdan
and Michon 1961). Zone 1 ranges between the
insertion of the FDP tendon and distal of the
insertion of the FDS tendon. Zone 2 injuries
occur between the insertion of the FDS tendon
and the proximal border of the A1 pulley. Zone 3
extends from the proximal aspect of the pulley
system to the distal end of the transverse carpal
ligament. Zone 4 injuries concern the rare lacer-
ations within the carpal tunnel. Finally, Zone 5
ranges from the proximal border of the trans-
verse carpal ligament to the myotendinous junc-
tion of the forearm. The zones of the thumb are
named Th1–Th3.
23.3.4 Diagnosis
The correct diagnosis of pediatric flexor tendon
injuries still remains a challenge for the treating
surgeons. Their diagnosis is more challenging in
younger children compared to adults, and there-
297
1
1
1
2 2
1
2
Th1
2
Th2
Th3
3
4
5
Fig. 23.2  Verdan zones of flexor tendon injuries (Verdan
and Michon 1961)
fore, late presentation seems to be more common
(Kato et  al. 2002). Knowledge of the correct
examination of the FDP and FDS tendons is of
pivotal importance.
In some cases, the mechanism of injury and
the location of the laceration already allow the
suspicion of a flexor tendon injury. The examina-
tion of the child begins with an inspection of the
hand with close observation of the cascade of the
digits. While the joints of the fingers are slightly
flexed when the flexor tendons are not injured
(Fig.  23.3), specific flexion failures occur when
flexor tendons are severed. If the distal phalanx is
extended and there is still a certain flexion of the
PIP joint, the tendon of the FDP is injured
(Fig. 23.4). In case of transection of the FDS ten-
don, the finger is slightly bent in both the PIP and
DIP joint (Fig. 23.5). Extension of the DIP and
PIP joint indicates injury of the FDP and FDS
tendon (Fig. 23.6). In these cases, however, a cer-
tain amount of flexion in the MCP joint is still
possible due to the function of the Mm. lumbrica-
les and interossei.
298
Fig. 23.3  Normal finger cascade, all finger joints are
slightly flexed
Fig. 23.4  Injury of the FDP tendon, the DIP joint is
extended and there is still a certain amount of flexion of
the PIP joint
Following the observation, a structured hand
exam should be performed provided the child
cooperates. With the PIP joint fixed by the
examiner, the active flexion of DIP joint and
­ of the tendons, potential sensitive deficits due to
thereby the function of the FDP tendon is tested
(Fig. 23.7a). To test the function of the tendon of
the M. flexor pollicis longus (FPL), the MCP
joint is fixed and the patient is asked to flex the IP
G. Singer and H. Friedrich
Fig. 23.5  Transection of the tendon of the M. flexor digi-
torum superficialis; the finger is slightly flexed in the DIP
and PIP joint
joint of the thumb. The function of the FDS ten-
don is tested with the neighboring fingers fixed in
extended position (Fig. 23.7b).
In younger patients, when a structured hand
exam is not possible, the tenodesis effect can be
indicative of a flexor tendon injury. Wrist exten-
sion causes flexion of the fingers in a normal fin-
ger cascade. If a finger does not bend according
to the anatomical cascade, flexor tendon injury
can be suspected. Additionally, applying pressure
on the palmar side of the forearm causes flexion
of the fingers.
However, it has to be kept in mind that all
these examinations may be without pathological
findings in case of partial transections of the
flexor tendons. Also, these injuries have to be
repaired in order to avoid secondary ruptures. In
younger children, a structured hand exam may
not be possible due to anxiety. Moreover, the size
of the wound does not allow the prediction of a
flexor tendon injury; even very small wounds
resulting, i.e., from stab injuries, can be associ-
ated with flexor tendon injuries. As a conse-
quence, every suspicious wound has to be
inspected in general anesthesia in order to mini-
mize the possibility of missing a flexor tendon
laceration.
In addition to the examination of the function
concomitant laceration of the neurovascular bun-
dle have to be assessed. Hypesthesia of the region
distally to the transection of the digital nerve may
be difficult to determine in younger children.
23  Injuries to the Tendons of the Hand 299

Fig. 23.6  Injuries of the FDS and FDP of the 3rd–5th finger of the right hand; the fingers are extended at the DIP and
PIP joint (left panel); following repair, the normal finger cascade is restored (right panel)

a b

Fig. 23.7 (a) Clinical test of the FDP tendon, the PIP fingers are fixed in extension and the patient is asked to
joint is fixed and the patient is asked to flex the distal pha- flex the finger
lanx; (b) clinical test of the FDS tendon, the neighboring

Therefore, examination under general anesthesia
should again be performed in suspicious cases
and both neurovascular bundles have to be exam-
ined during tendon repair. Transection of both
neurovascular bundles rarely occurs, but signs of
completely devascularized fingers represent a
surgical emergency (Lee and Schreiber 2015).
Conventional radiographs of the affected fin-
ger in two planes should be performed to exclude
osseous injuries as well as the presence of foreign
bodies.
23.3.5 Therapy
In case of suspected flexor tendon injuries, the
wound is to be cleaned, provisionally closed and
a splint applied. Additionally, antibiotic treat-
ment should be started. Primary repair or delayed
primary repair in the following days should be
performed by surgeons experienced in pediatric
hand surgery. After the injury has healed, flexor
tendon repairs are termed secondary. There is no
rigid space of time for performing secondary
sutures, but the chances for a successful recon-
nection of the tendon stumps without the neces-
sity of tendon transplantation diminish with time.
Surgical exploration should always be per-
formed under general anesthesia with magnifica-
tion and a tourniquet on the upper arm. Duration
of tourniquet inflation must not exceed 2 hours.
The wound is extended using Bruner zig-zag
incisions. Longitudinal cuts on the fingers should
be avoided in children, because they can lead to
stringent scar formation. When opening the ten-
don sheaths care has to be taken to preserve the
300
A2 and A4 pulley. Injured A2 and A4 pulleys
have to be reconstructed in order to avoid subse-
quent bow-string phenomena. In some cases, a
Z-shaped extension of the pulleys might be nec-
essary to facilitate unrestricted tendon gliding.
Following the opening of the tendon
sheaths, the stumps of the tendons are localized.
However, the ends of the tendons cannot always
be found at the level of the wound and their local-
ization depends on the position of the finger at
the time of injury and the status of the vincula
tendinea. If the vincula are injured and/or the fin-
ger was extended at the time the injury happened,
the proximal stump of the tendon may be far
retracted proximally. Flexion of the wrist, milk-
ing movements, or the use of rigid or flexible ten-
don retrievers may assist to deliver the proximal
stump to the repair site (Ozturk et al. 2013).
In 1917, Kirchmayr—a surgeon from Vienna—
first described a “grasping” method of tendon
suturing (Kirchmayr 1917). While in the original
description, the knot was placed between the two
tendon stumps, the variation according to Kessler
consists of placement of the knot distally or proxi-
mally on the tendon surface (Kessler 1973)
(Fig. 23.8). Zechner reported a method to coun-
tersink the knot in a small incision in the tendon
(Zechner et  al. 1985); however, this may not
always be possible depending on the size of the
tendon. Kirchmayr’s technique is, in different
variations, still commonly used today. Since this
first description, a plethora of different suturing
techniques have been published including two-,
Fig. 23.8  Kirchmayr–Kessler suture of a flexor tendon,
the first stitch is placed across the proximal or distal stump
of the tendon
G. Singer and H. Friedrich
four-, and six-strand techniques (Al-Qattan 2014,
Elhassan et  al. 2006, Fitoussi et  al. 1999, Kato
et al. 2002, Nietosvaara et al. 2007, Singer et al.
2017). Biomedical tests have shown that with an
increasing number of core sutures, the strength of
the suture also increases (Thurman et  al. 1998).
However, increased amounts of suture material on
the tendon surface also cause more gliding resis-
tance and promote postoperative adhesions
(Moriya et al. 2010, 2012). Additionally, pediatric
flexor tendons are significantly smaller than adult
tendons (Al-Qattan 2014). Therefore, two-­strand
techniques are still successfully used in pediatric
patients (Moehrlen et al. 2009; Singer et al. 2017).
We use a two-strand core suture according to
Kirchmayr–Kessler with a 3/0 or 4/0 monofilament
or multifilament non-absorbable suture double
armed with straight needles. Thereafter, a circum-
ferential epitendinous suture is applied using 5/0 or
6/0  in a continuous locking technique which has
been shown to increase suture strength by 50%
(Wade et al. 1989; Wieskotter et al. 2018). In con-
trary to other authors, we recommend to suture
both the FDS and FDP tendon in Zone 2, whereas
the relatively flat FDS slips near the insertion of the
tendon can be attended to with U-shaped sutures.
Sutures in cases of FPL tendon lacerations are
performed accordingly. In adults, the outcome
following FPL tendon repairs is worse compared
to the other fingers (Al-Qattan 2014). In children,
however, comparable data are missing.
In addition to the tendon repair, both neuro-
vascular bundles have to be inspected. In case of
transection of the digital nerve, coaptation of the
nerve should be performed with non-absorbable
interrupted 9/0 or 10/0 nylon sutures.
At the end of the operation, we apply a dorsal
forearm cast with 5° flexion of the wrist joint.
Rubber bands glued to the fingernails of the
affected fingers keep the MCP and IP joints in
flexion at rest.
23.3.6 Postoperative Rehabilitation
In 1967, the introduction of the active rehabilita-
tion program by Harold Kleinert consisting of
active extension and passive flexion made the
successful reconstruction of flexor tendons in
23  Injuries to the Tendons of the Hand
Zone 2 possible. Since then, this postoperative
treatment regimen has been applied to both adults
and adolescents. While adults, adolescents, and
older children are mostly treated applying this
technique, there is still no consensus concerning
the ideal rehabilitation regimen for younger chil-
dren. Often, a strict postoperative immobilization
program with casts applied for 4 weeks in chil-
dren who are unable to perform postoperative
exercises is applied (Ebinger et al. 2003; Fitoussi
et al. 1999). Other pediatric hand surgeons use an
active rehabilitation program irrespective of age
(Friedrich and Bäumel 2003; Moehrlen et  al.
2009; Singer et al. 2017).
We start with mobilization of the affected fin-
gers on the first postoperative day, with controlled
active extension and passive flexion exercises of
the MCP and IP joints at a minimum of five times
a day. With patients who are unable to actively
extend the fingers due to young age or lacking
cooperation, careful passive extension of the fin-
gers can be performed. On the second postopera-
tive day, the first cast is replaced by a dorsal
thermoplastic splint to the MCP joints in 0° to 5°
flexion of the wrist joint with rubber bands on the
affected fingers (Fig. 23.9). Patients and/or their
parents are instructed to repeat tendon gliding
Fig. 23.9  Thermoplastic splint with rubber band glued to
the fingernail for an injury of the FDS and FDP tendon of
the fourth finger
301
exercises by themselves on an hourly basis dur-
ing the daytime.
Following discharge from inpatient care, the
patients are to be seen once a week by a multi-­
professional team consisting of an occupational
and/or physical therapist and a pediatric hand
surgeon. Splinting and exercises should be con-
tinued for 5 weeks postoperatively. For patients
developing an extension deficit of the affected
fingers, an extension splint can be applied during
night hours. Sports activities and carrying heavy
loads are prohibited for 3 months. Regardless of
the type of postoperative treatment regimen, the
main goal is the avoidance of flexion contractures
at the proximal interphalangeal joint.
Outpatient follow-up examinations in increas-
ingly larger intervals are strongly recommended
and should be maintained until the end of adoles-
cence. This is necessary, because growth distur-
bances with concomitant tendon shortening may
require initiation of occupational therapy.
23.3.7 Outcome and Complications
The outcome following flexor tendon repair in
children is generally good. Sikora and coworkers
demonstrated an excellent outcome in 96% of
their 47 patients with a median age of 8  years
with 100 flexor tendon injuries using immobili-
zation in younger children (Sikora et  al. 2013).
Likewise, Möhrlen and coworkers achieved an
excellent and good outcome in the vast majority
of their patients using an age-adapted active reha-
bilitation program (Moehrlen et  al. 2009). In a
large cohort published by the author of this chap-
ter, uniformly excellent and good results follow-
ing flexor tendon repair followed by the
abovementioned active rehabilitation program
irrespective of age were shown (Singer et  al.
2017).
Secondary ruptures and adhesions represent
the two main complications following flexor ten-
don repair. In adult patients, these two complica-
tions have been shown to occur in 4% of the
patients, respectively (Dy et  al. 2012).
Comparable rates have been described in chil-
dren (Fitoussi et al. 1999; Singer et al. 2017).
302
In cases with early diagnosis of the re-rupture
and fingers with free passive range of motion,
operative revision with suture of the two tendon
stumps may be possible. Otherwise, tendon
reconstruction must be performed. Whether one-
or two-staged reconstruction is preferable in
pediatric patients remains to be answered in
future studies.
Tenolysis for adhesions should be performed
3–6  months following the initial operation. An
important prerequisite, however, represents the
cooperation of the patient, since an intensive
mobilization program is required postoperatively
in order to avoid recurring adhesions.
23.4 Extensor Tendon Injuries
The extensor tendons are more prone to injuries
due to their location directly under the skin, and
therefore, these injuries are encountered more
often than flexor tendon injuries (Armstrong and
Adeogun 2009; Fitoussi et al. 2007). Nevertheless,
excellent outcome can be expected following
correct treatment of extensor tendon injuries in
children and adolescents.
Pediatric extensor tendon injuries have a rela-
tively even gender distribution and most com-
monly affect the index and middle finger (Fitoussi
et al. 2007). Open injuries are usually caused by
broken glass, knives, or scissors. Subcutaneous
ruptures occur most often on the distal phalanx
and in some cases on the intermediate phalanx,
and are mostly caused by ball sports injuries. In
other locations, closed injuries usually do not
occur in children and adolescents.
23.4.1 Anatomy
In contrast to the flexor tendons, the extensor ten-
dons change their shapes several times in their
course from the forearm to the distal phalanx. On
the forearm, the extensor tendons are round similar
to flexor tendons. At the level of the wrist, the ten-
dons pass the retinaculum in six compartments that
prevent bowstringing when the wrist is extended.
While the tendons are flat and oval over the meta-
G. Singer and H. Friedrich
carpal bones and connected by intertendinous con-
nections, they insert in the extensor hood at the
level of the MCP joints. The extensor hood of the
finger trifurcates over the proximal phalanx. The
major central slip inserts on the base of the middle
phalanx to extend the PIP joint. Lateral branches of
the extensor apparatus merge with the tendons of
the interossei and lumbricals to form the conjoined
lateral bands, which then fuse over the middle pha-
lanx to form the terminal extensor mechanism
inserting on the base of the distal phalanx.
23.4.2 Zones
Lacerations of the extensor tendons can be cate-
gorized into eight zones for the second-to-fifth
finger and five zones for the thumb according to
Verdan’s classification system (Kleinert and
Verdan 1983) (Fig. 23.10). In 1999, Doyle et al.
have introduced a ninth zone on the musculoten-
dinous junction of the forearm (Doyle 1999). This
system dictates the preferred treatment regimen.
1
2
3
4
T1
T2 5
T3
6
T4
7
T5
8
Fig. 23.10  Zones of extensor tendon injuries according
to Verdan (Kleinert and Verdan 1983)
23  Injuries to the Tendons of the Hand
23.4.3 Diagnosis and Operative
Treatment
Following the inspection of the wound passive
and active motion of all joints should be tested.
Due to the complex anatomy, the possibility of
partial lacerations and anxiety of the children
correct diagnosis of extensor tendon injuries can
be challenging. However, if there is any doubt, an
inspection of the wound should be performed in
order to rule out underlying tendon injuries. A
radiograph is mandatory to exclude osseous inju-
ries (Dwyer et al. 2015).
Repair of extensor tendon injuries should be
performed in general anesthesia, loop magnifica-
tion, and a tourniquet on the day of the injury or
in the following days. In these cases, temporary
wound closure, antibiotic therapy, and splinting
are recommended.
In some cases, the proximal stump of the
injured tendon may retract proximally. The
degree of retraction depends on the localization
of the injury. Due to the intertendinous connec-
tions over the metacarpals, extensor tendon inju-
ries of the fingers usually do not retract. On the
contrary, the tendons may retract several centi-
meters in cases of injuries on the back of the hand
or the distal forearm or lacerations of the tendon
of the M. extensor pollicis longus.
In Zones 1–5, the extensor tendons are flat
and oval, and simple U-sutures are adequate for
tendon repair. In these zones, we recommend to
use absorbable sutures to avoid postoperative
complications such as suture granulomas. In
Zones 6-8., the tendons become thicker and sim-
ilar to flexor tendons. Therefore, Kirchmayr
sutures and their modifications are appropriate
for repair.
23.4.4 Injuries of Zone 1 (DIP Joint)
Injuries of Zone 1 (over the DIP joint) are called
mallet finger injuries and are usually closed inju-
ries and resemble ruptures of the extensor tendon
with or without avulsion of a bony fragment
caused by a forced flexion of the extended DIP
joint. The most common mechanism of injury is
303
being hit by a ball on an extended DIP joint with
forced palmar flexion and consecutive rupture of
the terminal extensor mechanism. Clinically, the
patients are not able to actively extend their distal
phalanx.
According to Patel and Garberman, mallet fin-
gers can be classified as acute with diagnosis
within 4  weeks following trauma and chronic
with diagnosis after 4  weeks (Garberman et  al.
1994; Patel et al. 1986).
The aim of treatment is the restoration of
active extension of the DIP joint and avoidance of
posttraumatic swan-neck deformity, which can
develop when the injury is left untreated. Closed
injuries without bony components can be suc-
cessfully managed with a Stack splint for
5–6 weeks (Fig. 23.11). This regimen can also be
applied for chronic subcutaneous injuries.
Discussion is still ongoing concerning the
necessity of operative intervention for mallet fin-
gers with bony components. Reported indica-
tions included fractures involving more than a
third or half of the joint surface and subluxation
of the distal phalanx. Recent reports with excel-
lent results following conservative treatment,
however, confirm the success of conservative
treatment (Lin and Samora 2018a, 2018b). It is
recommended to use a forearm cast in intrinsic
plus position with extension of the DIP joint for
the first 3  weeks followed by treatment with a
Stack splint for additional 3 weeks. Extension of
the DIP joint leads to approximation of the frag-
ments which has to be radiologically confirmed
at the time of cast application and 1, 2, and
3 weeks post-injury and at the end of treatment
(Fig. 23.12).
Fig. 23.11  Stack splint for treatment of mallet finger
injuries
304
Fig. 23.12  Mallet finger in a 12-year-old female patient (left panel), radiograph 1 week following the injury (middle
panel) and 6 weeks post-injury (right panel)
Operative treatment should be reserved for
cases with involvement of more than 50% joint
surface and subluxation of the distal phalanx and
in children in whom the extension of the distal
phalanx does not lead to sufficient approximation
of the bony fragment. The extension block tech-
nique according to Ishiguro has been shown to
lead to excellent results (Fig. 23.13).
Open lacerations are usually caused by cutting
injuries which are treated with either reconstruc-
tion of the tendon or dermatotenodesis and sub-
sequent splinting for 6 weeks. Depending on the
age of the patients and the extent of the lacera-
G. Singer and H. Friedrich
tion, temporary arthrodesis of the DIP joint can
be performed for the protection of the suture.
23.4.5 Injuries of Zone 2
(Middle Phalanx)
Zone 2 injuries are located over the middle pha-
lanx and or open in the vast majority of the cases.
Lacerations of both lateral bands cause inability
to extend the distal phalanx and should be
attended operatively, and the lateral bands recon-
structed followed by immobilization for
23  Injuries to the Tendons of the Hand
2 recommended. Injuries of one lateral band are
3 associated with a decreased extension strength
1 disruptions of the central slip cause decreased
Fig. 23.13 Extension block technique according to
Ishiguro for the treatment of bony mallet finger injuries;
forced flexion of the DIP joint reduces the fragment which
is then locked in the reduced position with a Kirschner
wire (1); the DIP joint is now extended to reduce the frac-
ture (2); a trans-articular Kirschner wire holds the distal
interphalangeal joint in slight flexion (3)
305
5–6  weeks in intrinsic plus position. An addi-
tional temporary arthrodesis of the DIP joint is
and should be repaired with absorbable U-sutures.
23.4.6 Injuries of Zone 3 (PIP Joint)
The extensor tendon over the PIP joint consists of
the central slip and both lateral bands. Isolated
strength of extension; the finger can be held in
position following passive extension. In cases
with injuries of the central slip and both lateral
bands, the finger is flexed at the PIP joint and the
patient is unable to extend the finger.
If left untreated, a boutonnière or buttonhole
deformity gradually develops within 10–14 days
following the injury.
Initial treatment of closed injuries consists of
immobilization of the PIP joint for 4–6  weeks
with the other joints left free. Open injuries need
surgical repair and care has to be taken to exclude
a potential involvement of the joint space.
23.4.7 Injuries of Zone 4 (Proximal
Phalanx)
Total lacerations of the extensor apparatus in
Zone 4 (proximal phalanx) are rarely seen
because of its broad configuration. Therefore, the
function of the tendon is usually not affected.
Small lacerations can be treated conservatively
with splinting of the PIP joint in extension for
5  weeks. Larger lacerations should be repaired
using non-absorbable sutures and subsequent
immobilization.
23.4.8 Injuries of Zone 5 (MCP joint)
Injuries of Zone 5 involve the terminal part of the
extensor tendons or the extensor caps over the
metacarpophalangeal joint. Extension of the
affected finger in the metacarpophalangeal joint
is not possible. However, the extension of PIP
306 G. Singer and H. Friedrich

and DIP joint is still possible by the correspond-
ing palmar muscles leading to the so-called “drop
finger”. Open injuries need to be repaired fol-
lowed by postoperative immobilization for
5–6 weeks. Proper repair keeps the tendon cen-
tralized over the joint avoiding subluxation of the
extensor tendon with associated loss of
extension.
23.4.9 Injuries of Zone 6
(Metacarpals)
The loss of extension of the fingers in case of
extensor tendon injuries at the dorsum of the
hand depends on the location of the laceration to
the intertendinous connections of the tendons. In
injuries proximal to the intertendinous connec-
tions, extension of the finger is still possible by
the tendons of the neighboring fingers. Extension
of the PIP and DIP joint is still possible by the
intrinsic hand muscles via the lateral bands.
Tendon repair can be performed either with
U-type sutures or modified Kirchmayr sutures
with subsequent splinting (wrist and fingers in
extension) for 4–6 weeks.
23.4.11 Injuries of Zone 8 (Distal
Forearm)
Extensor tendon injuries in this zone are caused
by severe injuries usually associated with lacera-
tions of more than one tendon. Tendons are
reconnected either by U-type sutures or by meth-
ods known from flexor tendon reconstruction.
Postoperatively, immobilization with a cast on
the palmar side of the forearm in extension of the
wrist joint and moderate flexion of the MCP
joints for 3–4 weeks should be applied. In com-
plex cases with involvement of several extensor
tendons, early dynamic mobilization with active
flexion and passive extension can be successfully
performed in order to prevent the development of
adhesions (Fig. 23.14).

23.4.10 Injuries of Zone 7
(Wrist Joint)

Extensor tendon injuries of the wrist are associ-

ated with damage of the dorsal retinaculum, and
often, the proximal stumps retract proximally.
Surgical repair is performed using either standard
core type sutures or U-sutures for the flat exten-
sor digitorum tendons. In order to facilitate unre-
stricted gliding of the tendon postoperatively, it
may be necessary to incise the affected compart-
ment or even a Z-plasty of the retinaculum.
Postoperatively, the forearm is immobilized
with a palmar-sided cast for 4-5  weeks with
extension of the wrist joint and moderate flexion
of the MCP joints. For younger children, an Fig. 23.14  Dynamic mobilization with active flexion and
above elbow cast is appropriate in order to pre- passive extension for complex injuries of the extensor ten-
vent accidental removal. dons on the forearm
23  Injuries to the Tendons of the Hand 307

23.4.12 Injuries of the Thumb Dwyer CL, Ramirez RN, Lubahn JD (2015) A brief

review of extensor tendon injuries specific to the pedi-
Extensor Tendons atric patient. Hand (N Y) 10:23–27
Dy CJ, Hernandez-Soria A, Ma Y, Roberts TR, Daluiski
Injuries of the thumb extensor tendons are cate- A (2012) Complications after flexor tendon repair: a
gorized into the Zones T1–T5 (Fig.  23.10). systematic review and meta-analysis. J Hand Surg Am
37:543-51e1
Lacerations in Zone T1 (interphalangeal joint)
Ebinger T, Fischer A, Katzmaier P, Wachter NJ, Traub
and Zone T2 (proximal phalanx) are treated anal- SE, Gulke J, Mentzel M (2003) Treatment of flexor
ogously to corresponding injuries of the second-­ tendon injuries in children. Handchir Mikrochir Plast
to-­fifth finger. In cases of complete transections Chir 35:353–357
Elhassan B, Moran SL, Bravo C, Amadio P (2006)
of the tendon of the extensor pollicis longus in Factors that influence the outcome of zone I and zone
Zone T3 (MCP joint), T4 (first metacarpal), or T5 II flexor tendon repairs in children. J Hand Surg Am
(carpometacarpal joint), the patients cannot 31:1661–1666
extend the interphalangeal joint of the thumb. Fitoussi F, Lebellec Y, Frajman JM, Pennecot GF (1999)
Flexor tendon injuries in children: factors influencing
However, due to the function of the M. extensor prognosis. J Pediatr Orthop 19:818–821
pollicis brevis, there is still a (weaker) extension Fitoussi F, Badina A, Ilhareborde B, Morel E, Ear R,
of the first MCP joint. Reconstruction of the ten- Pennecot GF (2007) Extensor tendon injuries in chil-
don can be performed with sutures according to dren. J Pediatr Orthop 27:863–866
Friedrich H, Bäumel D (2003) The treatment of flexor
Kirchmayr. Postoperatively, a cast with extended tendon injuries in children. Handchir Mikrochir Plast
MCP joint and moderate extension of the wrist is Chir 35:347–352
applied for 4-5 weeks. Garberman SF, Diao E, Peimer CA (1994) Mallet finger:
results of early versus delayed closed treatment. J
Hand Surg Am 19:850–852
Kato H, Minami A, Suenaga N, Iwasaki N, Kimura T
23.5 Conclusion (2002) Long-term results after primary repairs of zone
2 flexor tendon lacerations in children younger than
The extensor tendons are more prone to injuries age 6 years. J Pediatr Orthop 22:732–735
Kessler I (1973) The “grasping” technique for tendon
due to their location directly under the skin; repair. Hand 5:253–255
therefore, these injuries are more often encoun- Kirchmayr L (1917) Zur Technik der Sehnennaht
tered than flexor tendon injuries. Due to anxiety, Zentralblatt Chirurgie 44:906–907
examination of children is often complicated and Kleinert HE, Verdan C (1983) Report of the Committee on
Tendon Injuries (International Federation of Societies
requires time and patience. Nevertheless, timely for Surgery of the Hand). J Hand Surg Am 8:794–798
diagnosis and correct treatment of these injuries Kleinert HE, Kutz JE, Atasoy E, Stormo A (1973) Primary
are required in order to avoid devastating long-­ repair of flexor tendons. Orthop Clin North Am
term sequelae. Empathy and broad knowledge 4:865–876
Kleinert HE, Spokevicius S, Papas NH (1995) History of
are therefore necessary in order to correctly diag- flexor tendon repair. J Hand Surg Am 20:S46–S52
nose and treat tendon injuries of the hand in Lee SK, Schreiber JJ (2015) Flexor tendon injuries. In:
children. Abzug JM, Kozin SH, Zlotolow DA (eds) The pedi-
atric upper extremity. New  York, NY, Springer,
pp 909–927
Lin JS, Samora JB (2018a) Outcomes of splinting in pedi-
References atric mallet finger. J Hand Surg Am 43(1041):e1–e9
Lin JS, Samora JB (2018b) Surgical and nonsurgical man-
Al-Qattan MM (2014) Flexor tendon injuries in the child. agement of mallet finger: a systematic review. J Hand
J Hand Surg Eur 39:46–53 Surg Am 43:146-63e2
Armstrong MB, Adeogun O (2009) Tendon injuries in the Moehrlen U, Mazzone L, Bieli C, Weber DM (2009)
pediatric hand. J Craniofac Surg 20:1005–1010 Early mobilization after flexor tendon repair in chil-
Doyle JR (1999) Extensor tendons: acute injuries. In: dren. Eur J Pediatr Surg 19:83–86
Green DP, Hotchkiss RN, Pederson WC (eds) Green’s Moriya T, Zhao C, Yamashita T, An KN, Amadio PC
operative hand surgery. Churchill Livingstone, (2010) Effect of core suture technique and type on
New York, NY, pp 1962–1987 the gliding resistance during cyclic motion following
308
flexor tendon repair: a cadaveric study. J Orthop Res
28:1475–1481
Moriya T, Larson MC, Zhao C, An KN, Amadio PC
(2012) The effect of core suture flexor tendon repair
techniques on gliding resistance during static cycle
motion and load to failure: a human cadaver study. J
Hand Surg Eur 37:316–322
Newmeyer WL 3rd, Manske PR (2004) No man’s land
revisited: the primary flexor tendon repair controversy.
J Hand Surg Am 29:1–5
Nietosvaara Y, Lindfors NC, Palmu S, Rautakorpi S,
Ristaniemi N (2007) Flexor tendon injuries in pediat-
ric patients. J Hand Surg Am 32:1549–1557
Ozturk MB, Basat SO, Kayadibi T, Karahangil M, Akan
IM (2013) Atraumatic Flexor tendon retrieval – a sim-
ple method. Ann Surg Innov Res 7:11
Patel MR, Desai SS, Bassini-Lipson L (1986) Conservative
management of chronic mallet finger. J Hand Surg Am
11:570–573
Sikora S, Lai M, Arneja JS (2013) Pediatric flexor ten-
don injuries: A 10-year outcome analysis. Canadian J
Plastic Surg 21:181–185
Singer G, Zwetti T, Amann R, Castellani C, Till H,
Schmidt B (2017) Long-term outcome of paediatric
G. Singer and H. Friedrich
flexor tendon injuries of the hand. J Plast Reconstr
Aesthet Surg 70:908–913
Thurman RT, Trumble TE, Hanel DP, Tencer AF, Kiser
PK (1998) Two-, four-, and six-strand zone II flexor
tendon repairs: an in situ biomechanical comparison
using a cadaver model. J Hand Surg Am 23:261–265
Verdan C, Michon J (1961) The treatment of injuries of
the flexor tendons of the fingers. Rev Chir Orthop
Reparatrice Appar Mot 47:285–425
Wade PJ, Wetherell RG, Amis AA (1989) Flexor ten-
don repair: significant gain in strength from the
Halsted peripheral suture technique. J Hand Surg Br
14:232–235
Wieskotter B, Herbort M, Langer M, Raschke MJ,
Wahnert D (2018) The impact of different periph-
eral suture techniques on the biomechanical stabil-
ity in flexor tendon repair. Arch Orthop Trauma Surg
138:139–145
Zechner W, Buck-Gramcko D, Lohmann H, Goth D,
Stock W (1985) Improvement of suture technic in
flexor tendon injuries. Clinical and experimental
study. Handchir Mikrochir Plast Chir 17:8–13
 Burns
Alan David Rogers and Heinz Rode
24.1 Introduction
Burn injuries remain one of the most severe
causes of traumatic morbidity and mortality
internationally. Over ten million burns are man-
aged in a hospital setting every year, and as many
as 200,000 people die every year from these inju-
ries. Although only 5% of all burns require hospi-
talization, the resources required to care for
admitted patients are considerable, and the over-
all economic costs increase in systems poorly
prepared to holistically address their needs. As
most burns occur in developing countries, the
barriers to provide satisfactory burn care are
often thwarted by administrative oversight in
addition to inadequate resources. For those who
survive severe burn injuries, recovery is often
A. D. Rogers (*)
Ross Tilley Burn Centre, Sunnybrook Health
Sciences Centre, Toronto, Canada
Division of Plastic and Reconstructive Surgery,
Department of Surgery, University of Toronto,
Toronto, ON, Canada
e-mail: alandavid.rogers@sunnybrook.ca
H. Rode
Division of Paediatric Surgery, Department of
Surgery, University of Cape Town, Cape Town, South
Africa and Red Cross War Memorial Children’s
Hospital, Cape Town, South Africa
e-mail: heinz.rode@uct.ac.za
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_24
24
characterized by significant functional, cosmetic,
and psychosocial challenges.
Although the norm in developed countries for
decades, there is increasing recognition that
patients with these injuries should be managed at
regional burn centres, in order to obtain the best
possible outcomes (Palmieri et al. 2015). Within
such facilities, burn care is a goal-directed pro-
cess incorporating fluid resuscitation, critical
care, infection control, surgery, wound care, pain
relief, nutrition, as well as psychosocial and
physical rehabilitation. Health care workers dedi-
cated to these facilities have special training and
experience and include nurses (skilled at both
critical care and wound management), occupa-
tional and physical therapists, social workers,
pharmacists, surgeons, intensivists, dieticians,
and child psychology experts, amongst others. A
full spectrum of medical and surgical consultants
is also frequently called upon to offer guidance
and management as required. The facility should
have a streamlined two-way system of referral
and repatriation throughout the region it is
responsible for, and is preferably located at a
trauma centre within a university academic medi-
cal system. In addition to this interdisciplinary
environment, a community and family support
structure is really important during sustained
periods in the hospital and rehabilitation setting.
The indications for referral to a burn centre are
listed in Table 24.1.
309
310
Table 24.1  American Burn Association criteria for refer-
ral to a burn unit
 1. Partial-thickness burns >10% of the total body
surface area.
 2. Deep burns.
 3. Burns involving the face, hands, feet, genitalia,
perineum,
 4. or major joints,
 5. Circumferential extremity burns.
 6. Electrical burns and lightning injury.
 7. Chemical burns.
 8. Inhalational injury.
 9. Potentially complicating comorbidities.
10. Significant associated trauma.
11. Lack of qualified personnel or equipment for the
care of children.
12. Patients who require special social, emotional,
rehabilitation, e.g. suspected child abuse.
24.2 Diagnosis
Burn injuries are common in the paediatric popu-
lation and there is usually a clear history of expo-
sure to flames, hot liquids, chemicals, or contact
with a warm surface, from the parent, caregiver,
or child themselves. Clinically, these injuries
may extend to involve any depth, and from small
to extensive areas of the total body surface area,
as described in later sections.
24.3 Differential Diagnosis
Although a burn injury is at the top of the differ-
ential diagnosis for traumatic loss of layers of the
skin, there are a number of alternatives that ought
to be considered. The history elicited does not
always clearly differentiate some of the rarer
causes, but the pattern of injury and the condition
of the child may offer clues as to the diagnosis.
The most common of these are abrasions associ-
ated with a fall or pedestrian injury, for which the
clinician should carefully investigate underlying
fractures or organ trauma. There are a number of
inflammatory and infectious causes of skin loss
that ought to be considered, including the spec-
trum of childhood infectious and allergic derma-
tological ailments and the continuum of blistering
drug reactions and exfoliative disorders, incorpo-
A. D. Rogers and H. Rode
rating Stevens Johnson and Toxic Epidermal
Necrolysis Syndromes. Soft tissue bacterial
infections may also result in tissue loss, the most
significant of these being necrotizing soft tissue
infections. Frost bite should be considered as a
possibility in cold climates.
24.4 Aetiology
The severity of thermal injuries is influenced by
the age and physiological condition of the patient,
the mechanism of injury, the duration of contact
with the agent, the depth and size of the burn, the
anatomic site involved, as well as any associated
injuries. In children, two-thirds of burns occur in
the domestic setting, usually by scalds from boil-
ing water or cooking oil. The extent and pattern
of these injuries is influenced by gravity and the
fluid’s viscosity. The resulting duration of expo-
sure may in part explain the higher likelihood
that an oil burn will be deep. Similarly, in immer-
sion scalds, the duration of contact between the
hot liquid and the skin is considerably longer
than that with spill or splash scalds.
Flames burns are often deeper than fluid
scalds, and while they comprise only 20% of hos-
pitalized paediatric burns, they account for over
half of the mortality. These are most commonly
sustained during house or informal dwelling
fires, or associated with recreational or cooking
accidents in younger children, and risk-taking
behaviour in older boys. Contact, electrical, and
chemical burns account for the balance of paedi-
atric burns and are more frequently deep injuries
to the palms of the hands or the mouth, often
resulting in significant morbidity relative to the
size of the burn.
Accidental burns, such as those caused by a
child spilling a cup of coffee, a pot or a kettle,
result in the involvement of the head, trunk, and
palmar surface of hands and feet. Spill scald
burns are more likely accidental in nature and
have uneven borders with the deepest area in the
centre. In contrast, nonaccidental injuries, com-
prising as many as 15% of all admitted cases,
tend to occur more frequently on the dorsum of
hands and feet, the buttocks, and the perineum.
24 Burns
These burns can be distinguished from accidental
burns by the pattern and site of the burn, histories
given by the caretaker and patient, and a medical
examination of old scars (e.g. cigarette burn
scars), bruises, or fractures of different ages.
Nonaccidental burns often have distinct demarca-
tions between burned and normal skin, e.g. when
a child’s limb has been held in the water.
24.5 Pathogenesis
Thermal injury is a unique form of traumatic
injury, characterized by direct tissue injury to the
skin with associated inflammation and plasma
loss, which in turn has the capacity to compro-
mise multiple organ systems. Three concentric
‘Jackson zones’ of tissue injury characterize a
full-thickness burn: coagulation, stasis, and
hyperaemia. The central zone of coagulation
results from coagulative and ischaemic necrosis
from direct contact with the heat source and may
appear white or charred. The intermediate zone
of stasis is usually red, and while it may initially
blanch with pressure, the area may evolve to be
deeper, depending on the success or otherwise of
local and systemic resuscitation efforts. The
outer zone of hyperaemia blanches with the
application of pressure, indicating that it has
intact circulation.
Major burn injury is characterized by a syn-
drome of sustained hypermetabolism, with an
exaggerated systemic release of cytokines and
other mediators of inflammation. Increased vas-
cular permeability leads to extravasation and
oedema, which becomes greater with the admin-
istration of the fluid necessary to maintain organ
perfusion. Hypovolaemia accounts for decreased
perfusion and oxygen delivery, and the release of
catecholamines, vasopressin, and angiotensin can
lead to peripheral and splanchnic vasoconstric-
tion, further compromising perfusion to the burn
and the gastrointestinal system. Delays in initiat-
ing and maintaining adequate fluid provision not
only has the capacity to result in hypovolaemia
and shock, but can also result in reperfusion inju-
ries, with catastrophic effects. Burned skin also
causes exaggerated evaporative and conductive
311
water loss and cannot contribute meaningfully to
thermoregulation. As children have a relatively
larger body surface area to circulating blood vol-
ume, these fluid shifts are exaggerated.
24.6 Emergency Management
Initial assessment of a patient with a burn injury
entails an advanced trauma life support (ATLS)
style approach, as taught by the Emergency
Management of Severe Burns (international) and
the Advanced Burn Life Support Courses (North
America), including a primary survey to identify
and treat immediately life-threatening condi-
tions, followed by a secondary head-to-toe sur-
vey. Smouldering or hot clothing should be
removed and the patency of the airway ensured.
For suspected inhalation injury and for all major
burns, 100% oxygen is given by face mask and
the need for definitive airway placement is deter-
mined. Progressive airway obstruction may
occur, and early intubation and mechanical venti-
lation may be indicated.
Intravenous fluid should be initiated and a uri-
nary catheter inserted. Burns should be immersed
in cool running water (18  °C) and/or covered
with wet compresses (e.g. Burnshield) for at least
30  minutes to relieve pain and discomfort.
Although this has the ability to reduce the ulti-
mate depth of the burn, children with larger burns
should be monitored closely for hypothermia.
Copious irrigation of the wound with water is
also indicated for most chemical burns and
should continue until the surface of the wound
has a neutral pH. For electrical burns, the offend-
ing source should be removed from the patient,
using a nonconducting object, and the patient’s
circulation and ventilation evaluated and
maintained.
Decompressive escharotomies are emergency
procedures and performed under sterile condi-
tions for constricting circumferential deep burns
of the arms, legs, abdomen, and chest. The pur-
pose of these procedures is to avoid irreversible
ischaemic damage to distal parts or the restriction
of chest excursion and breathing. The incisions,
placed longitudinally along medial and lateral
312
sides of the limbs, must divide the burn and
extend to and expose viable tissue. Practitioners
should be aware of important structures such as
nerves (e.g. superficial peroneal, ulnar) and blood
vessels. These procedures are usually performed
under general anaesthesia or sedation.
24.7 Burn Wound Assessment
The size of the burn injury is expressed as a per-
centage of the total body surface area (TBSA).
A number of methods of assessment are used,
often in conjunction with each other. Significant
differences in assessment are observed even
between experienced burn surgeons, particularly
because there are wide disparities in patient pro-
portions. Methods used include: (1) the open
adducted volar surface of the hand (including fin-
gers) of the patient representing 1% of the TBSA,
and (2) Regions of the body are allocated per-
centages of the TBSA depending on the age of
the patient (Fig.  24.1). Children have relatively
larger heads and smaller lower limbs than older
15%
19%
9½% 32% 9½%
9½% 32% 9½%
17% 17%
15% 15%
1–4 5–9
Fig. 24.1  The Rule of Nines in Adults and adjusted areas
according to age to determine extent of burn injury
according to the total body surface area (TBSA). Note
A. D. Rogers and H. Rode
children and adults. The Lund and Browder
charts are widely used, and some centres also
make use of computer-assisted three-dimensional
systems in an effort to derive more accuracy
(Strobel and Fey 2018).
The depth of the burn dictates whether conser-
vative wound care or surgical management
should be employed, and also correlates with
long-term appearance and function (Fig.  24.2).
Early excision and skin grafting for deep burns
has been shown to have unequivocal benefit in
reducing infection, shortening hospital stays, and
improving ultimate function, but overly aggres-
sive debridement of burns of indeterminate depth
may remove viable dermal elements, potentially
result in exaggerated scarring, and needlessly
creates a second painful wounded area (donor
site), which is itself a superficial partial-thickness
injury that must heal with dressings.
Superficial burns injure only the epidermis,
and are erythematous without blistering, and are
not included in the TBSA burn estimation.
Superficial partial-thickness burns are confined
to the epidermis and the papillary layer of the
10%
13%
9% 36% 9%
9½% 32% 9½%
18% 18%
18% 18%
Adult
10–14
(Rule of nines)
how as children grow, their head gets smaller as a propor-
tion of the total body surface area, and their lower extrem-
ities increase
24 Burns
Fig. 24.2  The layers of
the skin (left), and the
depth of burn injury
according to this (right).
First-degree burns are
not included in TBSA
calculation. Full-­
thickness (third-degree
burns) and deep Epidermis
partial-thickness burns
(deep second degree)
usually warrant surgical
management
Dermis
Subcutaneous
tissue
dermis and usually heal within two to three
weeks, while deep dermal/deep partial-thickness
burns involve the epidermis and dermis into the
reticular layer and are generally believed to
require surgery. Full-thickness burns have irre-
versible destruction of all elements of the skin
and may extend into subcutaneous structures
including muscle, bone, and tendon. Many burns
are a combination of injury depth, and accurate
assessment may only be made after a few days of
evolution.
Various staining and imaging techniques have
been utilized to aid in the early classification of
burn depth, with varying efficacy to corroborate
the clinical assessment, and few are in wide-
spread use. The diagnosis of superficial partial-
and full-thickness burns is usually straightforward
and surgeons can usually rely on their clinical
assessment, but differentiating deep from superfi-
cial partial-thickness burns can be challenging.
Laser Doppler imaging is perhaps the most popu-
lar technology, with an accuracy of depth deter-
mination of 54% on day 1, increasing to over
90% on day 5 of the burn.
313
Partial- Full-
thickness thickness
First- Second- Third-
degree degree degree
24.8 Airway and Ventilation
Patients with inhalational injury may experience
upper airway swelling, acute respiratory failure
(via several mechanisms), and carbon monoxide
intoxication. Oedema of the upper airway may
increase rapidly in children, especially during
major fluid resuscitation, and urgent preemptive
intubation is recommended in patients with signs
of upper airway injury, such as stridor, inspira-
tory grunting, wheezing, or tachypnoea. Deep
burns to the face and neck or extensive burns
(requiring significant volumes of fluid during
resuscitation) often also warrant intubation (Snell
et al. 2013).
Fibreoptic bronchoscopy is a reliable method
of diagnosing acute inhalation injury and the ana-
tomic level (supraglottic or infraglottic), as well
as the severity of large airway injury. It may also
assist during the intubation of patients with inha-
lation injury and facilitate pulmonary toilet. The
compressive effect of a full-thickness burn of the
neck or chest may contribute to respiratory com-
promise and may necessitate intubation, and
314
sometimes a surgical airway. Escharotomy of the
chest entails lateral incisions in the anterior axil-
lary lines that extend 2 cm below the clavicle to
the ninth or tenth rib. The incisions are then
joined to form a square across the anterior chest
(Figure), thus releasing the constricting effect on
breathing.
Carbon monoxide (CO) has an affinity for
haemoglobin more than 200 times that of oxy-
gen; levels should be obtained in all patients with
burns sustained within a confined space. 100%
oxygen should be administered until the carboxy-
haemoglobin (COHb) level is less than 10%
because the elimination half-life for COHb
depends on oxygen tension. In room air, the half-­
life of CO-bound Haemoglobin is 4 hours, which
decreases to 45  min with 100% oxygen.
Administration of 100% oxygen increases the
gradient for oxygen binding, and unbound CO is
exhaled through the lungs.
Cyanide poisoning usually occurs when plas-
tics burn and can cause severe hypoxaemia and
metabolic acidosis by uncoupling oxidative
phosphorylation and disrupting cytochrome
function. A commercially available product,
Hydroxycobalamin, a precursor of vitamin B12,
binds cyanide to form cyanocobalamin, which is
excreted in the urine. It should be noted that it
stains the skin and mucous membranes purple
and some reports suggest it can increase the risk
of renal dysfunction. It can be administered prior
to burn centre admission if transfer is prolonged,
but preferably following discussion with a burn
surgeon.
One of the leading causes of morbidity and
mortality in major paediatric burns is ventilator-­
associated pneumonia (VAP) (Rogers et  al.
2014). Acute respiratory distress syndrome
(ARDS) is also a virulent consequence of the
overwhelming systemic inflammatory state,
inhalational injury, fluid resuscitation, and fre-
quent infectious complications associated with
burn injury. Specific strategies to ameliorate the
impact of inhalation injury include lung protec-
tive ventilation and medical adjuncts like inhaled
N-acetylcysteine and heparin.
A. D. Rogers and H. Rode
24.9 Fluid Resuscitation
Adequate fluid resuscitation is one of the most
critically important aspects of early major burn
care and aims to maintain tissue and organ perfu-
sion and minimize systemic sequelae, especially
renal failure (Mason et  al. 2016; Klein et  al.
2007; Cartotto 2009). Fluid resuscitation should
begin promptly; delaying resuscitation in chil-
dren for more than two hours increases morbidity
and mortality substantially. The American Burn
Association has developed practice guidelines
for fluid resuscitation, which include the follow-
ing recommendations:
1. Formal fluid resuscitation for burns exceeding
20% TBSA.
2. A crystalloid solution at a volume of 2–4 ml
per kilogram body mass per % TBSA during
the first 24 h.
3. Fluid resuscitation titrated to maintain a urine
output of 1  ml per kg per hour in children
under 30 kg, and 0.5 ml per kg per hour over
30 kg.
Maintenance fluid containing glucose should
be administered to children in addition to the
resuscitation fluid. Fluid administered should be
carefully titrated to effect, as both insufficient and
excess fluid may be deleterious. Patients who may
require relatively more volume include those with
full-thickness burns, inhalational injuries, electri-
cal burns, and concomitant trauma, and for whom
there was a delay in initiating resuscitation. A
review of 28 burn centres found that a majority of
patients with major burns actually received more
than 4 ml/kg/% burn recommended by Baxter and
Shires in their landmark Parkland Formula.
Furthermore, as many as 73% of patients now
receive volumes between 4 and 8 ml/kg/% burn. A
syndrome of overhydration has come to be known
as ‘fluid creep’ and may manifest as acute respira-
tory failure, pneumonia, multi-organ failure, the
conversion of superficial into deeper burns, as
well as abdominal, limb, and orbital compartment
syndromes (Saffle 2007).
24 Burns
Common errors in fluid resuscitation may
stem from the initial assessment of burn size esti-
mation, fluid volume calculation, delay in initiat-
ing resuscitation, over-reliance on formula
without titrating to effect, as well as neglecting
the impact of sedating agents and analgesia on
haemodynamics. Blind adherence to formulae
without regular evaluation and ongoing
­monitoring using urine output and clinical (blood
pressure, pulse, sensorium) and biochemical tests
(lactate, haematocrit) is perhaps the most signifi-
cant concern. Clinical monitoring in infants may
be challenging; using the Sheridan endpoint cri-
teria to assess satisfactory clinical response to
resuscitation may be a rational strategy.
The best known formula is the Parkland
Formula, which estimates a 24-hour fluid volume
of 4 ml/kg/% burn for the first 24 hours, with half
administered in the first 8 hours after the burn and
the balance during the subsequent 16  hours.
Ringer’ s lactate is the most frequently used crys-
talloid. The role of colloids, and specifically
albumin, is controversial, and several schools of
thought exist (both for and against its use). Many
authorities now advocate for its introduction in a
5% formulation within 12 hours of a major burn,
and making up 1/3 of fluid infused.
Enteral resuscitation, a method pioneered by
paediatricians for children with gastroenteritis,
may be effective under certain conditions, and
may be particularly applicable in developing
countries. This strategy makes use of crystalloids
and polymeric feeds and is applicable for even
moderate-size burns without contra-indications.
Adjuncts to fluid resuscitation may include anti-
oxidants (to counteract lipid peroxidase and
oxygen-­derived free radicals), plasma exchange,
activated protein C (to decrease the capillary
‘leak’), various antioxidants like vitamins C and
E, and N-acetylcysteine. High dose intravenous
vitamin C has been advocated by some, but its
diuretic effect limits the use of urine output as an
endpoint, and recent evidence also suggests that
this may result in a higher incidence of renal fail-
ure. Further high level evidence is required to
determine its value when compared to standard
nutritional supplementation and fluid resuscita-
tion utilizing crystalloids and albumin.
315
24.10 Analgesia and Sedation
Pain and anxiety are important to address at each
stage of burn care. In small children, these may
be difficult to differentiate and therefore complex
to treat effectively and without side effects. The
goal is to have a comfortable patient who is able
to participate in rehabilitation, tolerate wound
care, and willingly undergo operative interven-
tions. The therapeutic plan for pain management
must be flexible and individualized to combat
acute, background, breakthrough, procedural,
and postoperative pain components (Shah and
Liao 2017; Pardesi and Fuzaylov 2017). The
COMFORT-B scale is the only tool validated in
patients with burns, with descriptors including
level of alertness, agitation, crying, movement,
muscle tone, and facial tension. The FLACC
(Face, Legs, Activity, Cry, and Consolability) and
Wong Baker FACES scale have not been for-
mally validated, and the visual analogue scale
(VAS), although widely used, is not believed to
be reliable.
Multiple neural pathways contribute to pain
and anxiety, so most patients benefit from a mul-
tipronged analgesic strategy using agents with
several different pathways of action.
Opioids remain the mainstay of pain control
for many patients with burn injuries. Long-acting
oral analgesic agents are used for background
pain and short-acting agents for breakthrough
pain. Gabapentin and pregabalin are anticonvul-
sants widely used in managing neuropathic pain,
especially associated with donor sites and pru-
ritic healing wounds; they are also believed to
reduce the incidence of hyperalgesia, as well as
opioid doses (Nieuwendijk et  al. 2018).
Procedural pain may need a combination of
agents. The NMDA receptor agonist Ketamine,
for example, has both analgesic and dissociative
effects depending on the dosage, and can be
administered via several routes. Alpha-2 ago-
nists, like oral clonidine or Dexmedetomidine
infusion, may be used in addition to or as an
alternative to benzodiazepines to manage anxiety
and pain. NSAIDs are not widely used in chil-
dren as a result of their side effect profile. In older
children, the patient controlled analgesia (PCA)
316
method affords some control in their own pain
control. Regional analgesic techniques, e.g., fas-
cia iliaca compartment blocks, may be utilized
for thigh skin graft donor sites, for example.
Agents commonly used in paediatric burn care
are tabulated.
The ability to offer conscious or deep seda-
tions when performing dressing changes for
extensive areas is a major advantage of a well-­
equipped burn intensive care setting, and likely
reduces the incidence of anxiety and pain syn-
dromes (Hansen et al. 2019). Agents used for this
purpose include combinations of propofol, ket-
amine, benzodiazepines, and opioids. Inhaled
nitrous oxide has been used for many years, but
frequently requires supplementation in more
painful procedures. Dressing changes can also be
undertaken with the aid of user-friendly virtual
reality devices, which have been shown to effec-
tively distract by way of games or appropriately
selected 3-dimensional images adapted to reduce
the real or anticipated painful experience (Schmitt
et  al. 2011). Child life specialists also offer a
great deal of value in developing strategies to
cope with and distract patients from uncomfort-
able procedures.
24.11 Nutrition
and the Hypermetabolic
Response
The degree of hypermetabolism is related to the
extent of injury. Following fluid resuscitation,
cardiac output normalizes and then increases to
supra-normal levels, with a simultaneous increase
in resting energy expenditure. In severe burns,
the patient’s core temperature and basal energy
expenditure may remain elevated for several
months by the ongoing stimulation of the hypo-
thalamus by inflammatory mediators. In order to
facilitate wound healing, maintain organ func-
tion, combat infection, and ultimately restore
homeostasis, significant increases in nutritional
support are necessary. Burn injury causes the
release of massive quantities of amino acids from
muscle, exacerbated by increases in cortisol and
decreases in growth hormone. Growth hormone
A. D. Rogers and H. Rode
treatment has been shown to increase protein
synthesis in muscle and accelerates wound heal-
ing after burn injury.
For major burns, nutritional support should be
initiated during the course of fluid resuscitation
(preferably within 12  hours of the burn) using a
feeding tube, and this is maintained as long as nec-
essary, as energy requirements can usually not be
met entirely by mouth (Shahi et al. 2020; Rousseau
et al. 2013). The aim is to gradually reduce intra-
venous fluid and energy supply intravenously and
increase enteral feeding, until 48 h, when intrave-
nous fluid requirements should be minimal.
Continuous nutritional support with negligible
interruption is integral to meet nutrient needs for
wound healing and preservation of weight and
nutritional parameters. Modern burn centres
implement volume-based feeding strategies,
which compensate for any interruptions in enteral
feeding over a 24-hour period. For most patients,
they should receive clear, glucose-­containing flu-
ids up until two hours prior to the procedure.
For intubated patients, feeds should generally
not be withheld preoperatively and should be
continued throughout surgical interventions,
unless the airway is to be changed. Although gas-
tric feeds are safe in many patients, positioning
the tip more distally into the small intestine
reduces the risk of aspiration of feed and safely
facilitates continuous feeds, even during proce-
dures. Passage of the tip of the tube beyond the
pylorus may be facilitated by the administration
of metoclopramide or erythromycin.
The patient’s caloric requirements may be
estimated using formulae and measuring the rest-
ing energy expenditure by indirect calorimetry at
least on a weekly basis. Table  24.2 outlines the
approach to determine kcal/per day requirements
using the Toronto and Schofield eqs. A nonpro-
tein kilocalorie-to-nitrogen ratio of 100:1 and at
least 2  g of protein/kg/day are recommended.
Carbohydrate should comprise less than 60% of
the total energy intake, and fat under 35%.
Measuring the prealbumin level is a more effec-
tive measure than albumin to assess the impact of
nutritional support. Zinc, copper, selenium, vita-
mins C, B1, D, E, and glutamine supplementa-
tion should be implemented (Joffe et al. 2016).
24 Burns
Table 24.2  Preferred equations for calculation of kcal/
day requirements in major burns
Age Equation Requirement (kCal/day)
Adults Toronto −4343 + (10.5 X %TBSA) + (0.23
X caloric intake) + (0.84 X REE
by Harris Benedict) + (114 X
to)-(4.5 X days after injury)
Girls Schofield (17 X weight in kg) + (1618 X
3–10y height in cm) + 371.2
Boys Schofield (19.6 X weight in kg) + (1033 X
3–10y height in cm) + 414.9
Girls Schofield (8365 X weight in kg) + (4.65 X
10–18y height in cm) + 200
Boys Schofield (16.25 X weight in kg) + (1372 X
10–18y height in cm) + 515.5
Many of the pharmacological strategies inves-
tigated in the area of burn care are believed to
ameliorate the negative consequences of the
hypermetabolic response to major burn injury,
and the consumption of resources to sustain it,
redirecting them towards anabolism and wound
healing. Non-selective beta blockers, like pro-
pranolol, believed to blunt hypermetabolism,
address insulin resistance and improve peripheral
lean mass, may be considered along with other
non-nutritional strategies like early surgery,
warm ambient temperatures, and oxandrolone
(LeCompte et al. 2017; Flores et al. 2016). Insulin
resistance and hyperglycaemia may be closely
observed and controlled, although moderate glu-
cose control with continuous insulin applied in
adults may pose a greater risk of hypoglycaemia
in the paediatric population.
24.12 Infection Prevention
and Control
Infection remains the main cause of mortality
after the resuscitation period of a severe burn.
Nosocomial infections are very common in these
patients and include ventilator-associated pneu-
monia, wound infection, and sepsis associated
with indwelling vascular lines, Foley catheters
and tubes necessary for their critical care (Rogers
et al. 2014). Infection prevention and control pro-
tocols, patient isolation, efficient surgery, optimal
antimicrobial dressing use, and reduced duration
317
of invasive devices, all contribute to limiting
morbidity from infections. Antibiotics are initi-
ated empirically for systemic infection according
to known resistance patterns, as well as the tim-
ing and location of infections, and adjusted sub-
sequently based on cultures.
Although admission surveillance wound
swabs may be valuable as part of a protocol, rou-
tine open wound swabs, taken without clinical
correlation of features of wound infection, should
not be used to inform clinical decisions.
Quantitative samples and biopsies may be more
rational in differentiating colonization from true
wound infection. A wide variety of topical anti-
microbial dressings are in use for the manage-
ment of burns and burn wound infections, and
decisions should usually be made based on the
dressing interface and the desired duration
between dressing changes, rather than the active
antimicrobial agent. All antiseptic dressings are
inherently cytotoxic, to varying degrees, and so
the choice of agent should be adjusted fairly reg-
ularly to balance efficacy with its potential cyto-
toxicity and potential impact on wound healing.
That being said, bacterial species that favour very
moist environments, like pseudomonas, may be
better managed using topical agents that oppose
this, like acetic acid 2% and povidone iodine
(betadine/inadine).
24.13 The Partial-Thickness Burn
Superficial partial-thickness burns are usually
healed within 3 weeks and are treated conserva-
tively, which may include cleaning, debridement
of blisters and adherent exudate or pseudo-
eschar, topical antimicrobials, and/or cytopro-
tective occlusive or semi-occlusive dressings.
Skin grafting will not improve functional out-
come, time to healing, or aesthetic results. Burns
initially classified as ‘indeterminate or mixed
depth’ are often best left for 7–10  days before
the final decision is taken as to the optimal man-
agement strategy. All burns should ideally either
be healed or grafted by 3  weeks to minimize
hypertrophic scar formation.
318
Timeous fluid resuscitation (if indicated) and
the early application of cool running water and
cold compresses (e.g. Burnshield) will reduce the
depth and therefore the severity of injury. As
soon as possible after stabilization, the child
should be washed with soap and water, and blis-
ters may be debrided (controversial, unless tense
and impairing range of motion they may be left).
To ensure wound healing, help to avoid desicca-
tion and bacterial infection, the burn wound
should then be covered with a dressing; these
should aim to maintain a clean wound environ-
ment, absorb excess exudate, facilitate easy
application and removal, and preferably have
some antimicrobial action. A variety of proteo-
lytic enzymes (collagenase, bromelain) may also
be used to remove superficial eschar and pseudo-­
eschar from burn wounds, but these are fairly unit
specific and not in widespread use.
A large number of dressings are available to
the burn surgeon for the management of partial-­
thickness burns. Certain dressings may have an
advantage over others in the management of
partial-­thickness burns and donor sites in terms
of time to wound healing, number of dressing
changes, ease of use, pain experienced, exudate
management, and cost. However, many random-
ized trials are flawed in their randomization strat-
egy, depth assessment, or outcome scoring
methods. Newer dressing modalities are more
expensive per unit but often require fewer dress-
ing changes than more traditional options. This is
particularly true for hydrocolloids, silicone, skin
substitutes, and nanocrystalline silver dressings
when compared to tulle gras and silver sulfadia-
zine, which are usually changed daily. Silver sul-
fadiazine (flamazine) has been in use for several
decades and remains an important dressing
modality globally in the treatment of burns. It
offers a cooling effect and is quick and easy to
apply, smothered first in gauze. Its efficacy is
unfortunately compromised by pseudo-eschar
formation and the need for regular dressing
changes (preferably twice daily), and therefore,
many newer dressings may perform favourably
when one considers ultimate staff and product
costs.
A. D. Rogers and H. Rode
Dressing choice may be influenced by the
adherence of the wound contact layer, its required
duration and interval, the wound’s moisture bal-
ance, and the presence or absence of necrosis,
slough, infection, or eschar. One should appreci-
ate that all antiseptics, including silver, are inher-
ently cytotoxic and should be used with caution.
Selected agents can be diluted to offer a more
biocompatible balance of antimicrobial efficacy
and keratinocyte survival. Negative pressure
wound therapy is an effective bolster for skin
grafts and incisions, and options to instil antisep-
tics have augmented its value in wound bed prep-
aration. Biobrane and Suprathel are definitive
temporary skin substitute that have been shown
to effectively facilitate spontaneous healing in
superficial partial-thickness burns with a single
application, and are able to reduce painful dress-
ing changes considerably (Rogers et al. 2011).
24.14 The Deep Burn
It is recommended that early total or serial exci-
sion of deep partial- and full-thickness burns be
performed as soon as haemodynamic stability is
restored and depth of the burn determined
(Herndon et al. 1989). In the case of large burns,
the percentage total body surface area (TBSA) to
be excised at each procedure depends on the sta-
bility of the patient, the burn size, donor sites,
anticipated blood loss and temperature control,
anaesthetic and critical care considerations, and
the experience and number of surgeons. Once
excised, usually by way of Humby (variations
include Braithwaite or Watson) or Goulian (also
known as Weck) bladed instruments (Fig. 24.3),
the deep wounds need to be covered with skin
grafts harvested with a dermatome from unburned
areas at that time or at a later surgery. These
grafts can be meshed and then expanded to obtain
greater wound coverage (Rode et  al. 2016).
Wherever possible, burn surgeons aim to obtain
complete wound healing within three weeks,
ostensibly to limit the development of scar hyper-
trophy and contracture and to reduce infection
risk.
24 Burns 319

a b

Fig. 24.3  Humby knife (a) and Weck/Goulian blade (b) enable expansion. The ridges on the derma-carrier or on
for the tangential excision of deep burns for large and the mesher itself, depending on the design, determine the
small/special areas, respectively. A dermatome (c) for har- mesh ratio
vesting split skin grafts, and skin graft mesher (d) to

A number of strategies, including tourniquet Table 24.3 Major indications for deceased donor
use and infiltration with epinephrine containing allograft skin in burn centre
solutions into donor and recipient sites, signifi- Indication Application
cantly reduce blood loss and transfusions Extensive burn Staged application at time of
excision and removal when
(Cartotto et al. 2000). The three ‘P’s’ should be
autograft applied
utilized to determine the adequacy of excision, Over wide meshed autograft (3:1
namely the presence of: 1.Patent blood vessels, or Meek)
2.Pearly white dermis, and 3. Pale yellow fat. Smaller burns Burn involving special areas
Bleeding should not be used as a primary end- (hand/face) to improve sheet
autograft take
point in this regard. Management of burn wound
Burn centres should ideally have ready access infection
to fresh or cryopreserved deceased donor Necrotizing soft ‘Test’ of wound bed/optimize
allograft skin. The three main indications include tissue infection general condition prior to
definitive closure
its use as a temporary biological cover for excised
Exfoliative skin Avoid repeated painful and
burns in the extensive burn, in the context of condition traumatic dressings (alternative to
infection, and in the staged management of sheet Biobrane)
grafting of important functional and aesthetic Complex wound ‘Test’ of wound bed/optimize
areas like hands and faces (Table  24.3). Its use general condition prior to
definitive closure
has been shown to improve autograft take, as the
opportunity to ‘fine-tune’ debridements and pre-
cisely secure autografts; the focus of the allograft Permanent skin substitutes, like Integra, Mat-
stage would be primarily burn wound excision riderm, and Biodegradable Temporizing Matrix
and haemostasis (Allorto et al. 2016). (BTM), are believed to add value to reduce the
320
negative sequelae of thin, widely meshed (e.g.
3:1 mesh ratio) autograft, in cases where repeat
harvesting of these areas is inevitable in exten-
sive burns. Biobrane, previously mentioned as a
temporary but definitive option in the context of
superficial partial-thickness burns, is a useful and
more rapidly applied alternative to allograft in
the excised burn, but would offer less value than
well adhered allograft in determining the ade-
quacy of debridement.
Meek micrografting is an extremely effective
means of widely expanding autograft from lim-
ited available donor sites to optimize coverage
for the very extensive burn, often over one of the
previously mentioned skin substitutes (BTM or
Integra). This technique does not mesh the skin
like standard 2:1 or 3:1, but cuts it into small
squares, which are elegantly applied with glue to
pre-folded gauze templates. When expanded, the
small squares are separated to cover a large area,
up to 9 times the size of the original donor site
surface area, and are initially applied to the
wound by securing the gauze with staples (Rode
et al. 2017).
24.15 Physical and Psychological
Rehabilitation
Dedicated occupational and physical therapists
should be involved from the time of admission
with burn injury to well beyond discharge, and
help to prevent or decrease deconditioning, main-
tain function, help patients to address or adapt to
disability, and reduce the negative sequelae of
scarring in the healing burn. This may involve the
development and application of adaptive devices,
prostheses and splints, and pressure garments.
The social worker is integral in offering com-
passionate and holistic patient care from admis-
A. D. Rogers and H. Rode
sion to discharge and beyond, not least when
communicating with patients and families regard-
ing major treatment decisions like advanced care
planning and end of life decision-making, but
also negotiating transitions between one phase of
care and the next. Psychological support is also
fundamentally important, and specialists in this
area are frequently required to complement the
invaluable contributions of the social worker in
facilitating acute management, rehabilitation,
and reintegration into society.
24.16 Reconstructive Burn Surgery
This important component of plastic surgery is
often undertaken after discharge from hospital,
and is usually offered in consultation with the
burn occupational and physical therapists
(Friedstat and Hultman 2014). Burn reconstruc-
tion addresses significant functional and aesthetic
considerations through reorienting and lengthen-
ing procedures like z-plasties (Fig. 24.4), and the
resurfacing of scarred areas with skin grafts, as
well as local, regional, and free flaps (Orgill and
Ogawa 2013; Hudson 2000). Traditionally, this
would occur only after a period of scar matura-
tion (6 months to one year or more) unless sig-
nificant consequences would result if not
addressed; a neglected lower lid ectropion, for
example, may result in ocular surface damage
and compromise vision. Splinting, pressure gar-
ment use, steroid injection, and silicone gel
sheeting are modalities that may reduce the need
for surgical intervention (Tredget et al. 2014). Of
all these strategies, pulsed-dye and fractional car-
bon dioxide laser therapy are arguably the most
promising methods of modulating the maturation
and hypertrophy of burn scars (Hultman et  al.
2014).
24 Burns 321

a b

Fig. 24.4  Release of right first webspace contracture (a) with z-plasties (b)

24.17 Outcomes tions vary considerably between one centre and

Several models have been developed to assist in
the prediction of mortality or futility after major
burn injury. The best known of these is the Baux
score, which is based on the sum of patient age
and body surface area; the revised version also
incorporates inhalation injury. The original pre-
dictions based on this score are now outdated as a
result of improvements in management strategies
over the last three decades, but it is still frequently
referenced (Baux 1961). Almost all children are
expected to survive their burn injuries, and mor-
tality rates have therefore ceased to be the promi-
nent outcome measure for burn centres (Rode
et al. 2011). Similarly, lengths of hospital stay are
dependent on a variety of patient and injury
related factors and may have little relationship to
functional outcomes, patient morbidity or
experience. Definitions for specific complica-
­ evidence-based strategies in a sustainable fashion.
the next, making the significance of these out-
comes questionable and comparisons difficult.
As such, patient reported outcomes are increas-
ingly recognized as an important measure of suc-
cess in terms of care of patients with burns, in
conjunction with more traditional outcome mea-
sures (Anzarut et al. 2005).
Burn centres aiming to improve outcomes
should support and develop quality improvement
systems. Such activities include regular audits,
multidisciplinary mortality and morbidity fora,
verification processes with independent organiza-
tions, as well as the reporting of outcomes to mul-
ticentre databases for the purposes of benchmarking
(Rogers 2020). Arguably the single most impor-
tant of these activities is the development and exe-
cution of local quality improvement interventions,
usually small-scale projects aimed to introduce
322
24.18 Conclusions
Burn Surgery is now appropriately regarded as a
sub-speciality area within plastic surgery, gen-
eral surgery, and paediatric surgery and can also
become the focus area of members of the multi-
disciplinary team including nurses, social work-
ers, psychologists, dieticians, physical and
occupational therapists, intensivists, pharma-
cists and intensivists, amongst others. Within an
environment that prioritizes the role of each of
these health workers, burn care has become
more organized, standardized, and therefore
more efficient. Within such a context, quality
improvement interventions can also become
entrenched and sustained, ultimately leading to
improvements in functional, psychological, and
aesthetic outcomes. This chapter outlines the
fundamental principles important in the care of
child who has sustained a burn injury, from ini-
tial assessment to critical care, surgery, and ulti-
mately rehabilitation.
References
Allorto NL, Rogers AD, Rode H (2016) ‘Getting under
our skin’: introducing allograft skin to burn surgery in
South Africa. S Afr Med J 106(9):865–866
Anzarut A, Chen M, Shankowsky H, Tredget EE (2005)
Quality-of-life and outcome predictors following mas-
sive burn injury. Plast Reconstr Surg 116:791–797
Baux S (1961) Contribution a l’Etude du traitement local
des brulures thermigues etendues. These, Paris
Cartotto R (2009) Fluid resuscitation of the thermally
injured patient. Clin Plast Surg 36:569–581
Cartotto R, Musgrave MA, Beveridge M, Fish J, Gomez
M (2000) Minimizing blood loss in burn surgery. J
Trauma 49:1034–1039
Flores O, Stockton K et  al (2016) The efficacy and
safety of adrenergic blockade after burn injury: a sys-
tematic review and meta-analysis. J Trauma Acute
Care Surg 80(1):146–155. https://doi.org/10.1097/
TA.0000000000000887
Friedstat JS, Hultman CS (2014) Hypertrophic burn scar
management: What does the evidence show? A sys-
tematic review of randomized controlled trials. Ann
Plast Surg 72:S198–S201
Hansen JK, Voss J, Ganatra H et al (2019) Sedation and
analgesia during pediatric burn dressing change: a
Survey of American Burn Association Centers. J Burn
Care Res 40(3):287–293. https://doi.org/10.1093/jbcr/
irz023
A. D. Rogers and H. Rode
Herndon DN, Barrow RE, Rutan RL, et al (1989) A com-
parison of conservative versus early excision: therapies
in severely burned patients. Ann Surg 209:547–552;
discussion 552–553
Hudson DA (2000). Some thoughts on choosing a
Z-­plasty: the Z made simple. Plast Reconstr Surg
106(3):665–71. Review. PMID: 10987477
Hultman CS, Friedstat JS, Edkins RE, Cairns BA, Meyer
AA (2014) Laser resurfacing and remodeling of hyper-
trophic burn scars: the results of a large, prospective,
before-after cohort study, with long-term follow-up.
Ann Surg 260:519–529; discussion 529–532
Joffe A, Anton N, Lequier L, et al (2016) Nutritional sup-
port for critically ill children. Cochrane Database Syst
Rev 2016(5):CD005144. Published 2016 May 27.
https://doi.org/10.1002/14651858.CD005144.pub3
Klein MB, Hayden D, Elson C et al (2007) The associa-
tion between fluid administration and outcome fol-
lowing major burn: a multicenter study. Ann Surg
245:622–628
LeCompte MT, Rae L, Kahn SA (2017) A survey of the
use of propranolol in burn centers: who, what, when,
why. Burns 43(1):121–126. https://doi.org/10.1016/j.
burns.2016.07.007. Epub 2016 Aug 27
Mason SA, Nathens AB, Finnerty CC et  al (2016)
Inflammation and the Host Response to Injury
Collaborative Research Program. Hold the pendulum:
rates of acute kidney injury are increased in patients
who receive resuscitation volumes less than predicted
by the parkland equation. Ann Surg 264(6):1142–
1147.PMID: 27828823
Nieuwendijk SMP, de Korte IJ, Pursad MM, van Dijk
M, Rode H (2018) Post burn pruritus in pediatric
burn patients. Burns 44(5):1151–1158. https://doi.
org/10.1016/j.burns.2018.02.022
Orgill DP, Ogawa R (2013) Current methods of burn
reconstruction. Plast Reconstr Surg 131:827e–836e
Palmieri TL, Taylor S, Lawless M, Curri T, Sen S,
Greenhalgh DG (2015) Burn center volume makes
a difference for burned children. Pediatr Crit Care
Med 16(4):319–324. https://doi.org/10.1097/
PCC.0000000000000366
Pardesi O, Fuzaylov G (2017) Pain management in pediat-
ric burn patients: review of recent literature and future
directions. J Burn Care Res 38(6):335–347. https://
doi.org/10.1097/BCR.0000000000000470
Rode H, Millar AJW, Castle B, Lyle J (2011) Ethical deci-
sion making in severe burn victims. SAMJ 101(1):17–19
Rode H, Martinez R et  al (2016) A review of the peri-
operative management of paediatric burns. S Afr
Med J 106(11):1114–1119. https://doi.org/10.7196/
SAMJ.2016.v106i11.10938
Rode H, Martinez R, Potgieter D, Adams S, Rogers AD
(2017) Experience and outcomes of micrografting
for major paediatric burns. Burns. 2017 Mar 15. pii:
S0305-4179(17)30111-0. https://doi.org/10.1016/j.
burns.2017.02.008
Rogers AD (2020) Quality Improvement and patient
safety in burns care. Wound Healing Southern Africa
13(1):7–12
24 Burns
Rogers AD, Adams S, Rode H (2011) The Introduction of
a protocol for the use of Biobrane for facial burns in
children. Plast Surg Int 2011:858093
Rogers AD, Deal C, Argent AC, Hudson DA, Rode H
(2014) Ventilator associated pneumonia in major pae-
diatric burns. Burns 40(6):1141–1148
Rousseau AF, Losser MR, Ichai C, Berger MM (2013)
ESPEN endorsed recommendations: nutritional ther-
apy in major burns [published correction appears in
Clin Nutr. 2013 Dec;32(6):1083]. Clin Nutr 32(4):497–
502. https://doi.org/10.1016/j.clnu.2013.02.012
Saffle JI (2007) The phenomenon of “fluid creep” in acute
burn resuscitation. J Burn Care Res 28:382–395
Schmitt YS, Hoffman HG, Blough DK et  al (2011) A
randomized, controlled trial of immersive virtual real-
ity analgesia, during physical therapy for pediatric
burns. Burns 37(1):61–68. https://doi.org/10.1016/j.
burns.2010.07.007
323
Shah AR, Liao LF (2017) Pediatric burn care: unique con-
siderations in management. Clin Plast Surg 44(3):603–
610. https://doi.org/10.1016/j.cps.2017.02.017
Shahi N, Skillman HE, Phillips R, et  al (2020) Why
delay? Early enteral nutrition in pediatric burn patients
improves outcomes [published online ahead of print,
2020 Aug 18]. J Burn Care Res iraa129. https://doi.
org/10.1093/jbcr/iraa129
Snell JA, Loh NH, Mahambrey T, Shokrollahi K (2013).
Clinical review: the critical care management of
the burn patient. Crit Care 17(5):241. https://doi.
org/10.1186/cc12706. Review. PMID: 24093225
Strobel AM, Fey R (2018) Emergency care of pediatric
burns. Emerg Med Clin North Am 36(2):441–458.
https://doi.org/10.1016/j.emc.2017.12.011
Tredget EE, Levi B, Donelan MB (2014) Biology and
principles of scar management and burn reconstruc-
tion. Surg Clin North Am 94:793–815
 Foreign Bodies
S. Shah, L. Nguyen, and R. Sun
25.1 Introduction
Children can put just about anything they can
grasp into their mouths or their noses and then
swallow it or aspirate it. The vast majority of for-
eign body (FB) ingestions or aspirations are acci-
dental. FB of the aerodigestive tract whether they
are aspirated, inserted, or ingested is potentially
dangerous. If certain objects are not diagnosed
early and removed promptly, they can result in
numerous complications, such as perforation,
obstruction, tissue necrosis, fistula formation,
ulcerations, massive bleeding, or pulmonary
infections. These complications carry a signifi-
cant morbidity and mortality. The incidence of
FB ingestions has increased in the last 2 decades.
The 2018 Annual Report of the American
Association of Poison Control Centers noted
66,519 foreign body ingestions in children
younger than 5  years of age. Overall estimates
are as high as 100,000 cases per year of FB inges-
tion in the United States (US).
From 2001 to 2009, estimated 111,914 chil-
dren ages 0–14 were treated in US hospital emer-
S. Shah (*)
Baylor College of Medicine, Houston, TX, USA
L. Nguyen
Montreal Children’s Hospital, Montreal, QC, Canada
e-mail: Luong.nguyen@muhc.mcgill.ca
R. Sun
Texas Children’s Hospital, Houston, TX, USA
e-mail: rcsun@texaschildrens.org
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_25
25
gency rooms for nonfatal food-related choking.
Hard candy was the most common type of FB
associated with choking. In 2001, the US Center
for Diseases Control (CDC) reported that esti-
mated 60% of choking episodes treated in
Emergency Departments were due to food items
such as peanuts, seeds, candy, gum, pieces of
fruit, vegetables, and hot dogs. Another 30%
were due to nonfood substances of which coins
accounted for a significant portion. Other non-
food items are plastic pieces, screws, pins, and
button batteries (Chapin et al. 2013).
Sixty-eight percent of the deaths in children
younger than 14  years reported to the US
Consumer Product Safety Commission were due
to nonfood substances. The remaining 32% of
deaths were caused by household items. The
majority of deaths occurred in children aged
3 years and older (Lelli Jr. 2005).
The diagnosis of a FB in the aerodigestive
tract may be challenging because of the difficulty
in obtaining a reliable history from children,
especially when they are very young. Children
may present with a wide range of symptoms
including choking, shortness of breath, coughing,
drooling, stridor, wheezing, pain, fuzziness, chest
pain, abdominal pain, and fever. Others may
remain asymptomatic but are brought in, because
a parent or caretaker witnessed the ingestion or
aspiration.
In clinical practice, most children (80%) had
been witnessed to choke on an identifiable object,
325
326
but only 52% of events of airway FB were diag-
nosed early. Estimated 40% of foreign body
ingestions are not witnessed, and in many cases,
the child never develops symptoms. In a retro-
spective review, only 50% of children with con-
firmed foreign body ingestion were
symptomatic.
Objects that have passed the esophagus, once
they reach the stomach, do not cause symptoms
unless complications occur. They are usually
eliminated spontaneously—even sharp objects—
with normal bowel movements. Therefore, the
spontaneous passage of FB would be common
and unreported if the child remained asymptom-
atic. Prior to the 1930s, the mortality associated
with FBs was very high. Currently, it is about
1–2%. In recent years, the development of mod-
ern instruments and types of equipment has dra-
matically improved the techniques for the
removal of foreign bodies, even in the small
child. During the same period, the ability to make
a better diagnosis of foreign body ingestion or
aspiration and their complications have improved,
reducing the mortality and morbidity in these
children (Puri and Hollwarth 2006).
25.2 Airway Foreign Bodies
Differential Diagnosis
Epiglottis, asthma, pneumonia, upper respiratory
infection, peritonsillar abscess, bronchiolitis, and
pharyngitis.
25.2.1 Ear
The child can place a multitude of relatively
small objects in the ear canal. Tissue, foam, or
paper fragments are the most common FBs
retrieved from the external ear canal followed by
beads, round toys, fragments of food materials,
and erasers.
The majority of children with FBs in the ear
canal were seen on the day of the accident (72%)
but up to 8% were seen more than a week later. In
S. Shah et al.
the vast majority of cases, the child was asymp-
tomatic. Some may complain of pain, otorrhea,
bleeding or decrease of hearing. FBs in the ear
canal can be successfully removed by using small
forceps or irrigation of the ear canal with gentle
suction by using a small suction tip. Removal of
an FB in the ear canal is sometimes challenging.
Without the cooperation of the child, it can be
extremely difficult. In one study, retrieval by
Emergency Room staff was successful in a small
number of patients (7%). With the majority of
cases referred to Ear, Nose, and Throat (ENT)
specialists. General anesthesia was required
when the child did not cooperate or the FBs were
beads or solid objects which were difficult to
remove without anesthesia.
25.2.2 Nose
Most of the nasal FBs may be visualized directly
by using a rhinoscope or an otoscope. Common
FBs placed in the nasal cavity are beads, pieces of
paper, foam, tissue, and food matter (nuts, seeds,
dried beans, and other pieces of food). Other mis-
cellaneous objects included stones, screws, button
batteries, pellets, and erasers. Nasal FBs, if not
removed early, can cause foul-­smelling nasal dis-
charge, airflow obstruction, and rhinitis. In one
study, 80% of patients present on the same day,
8% present the following day, and 10% are dis-
covered incidentally following the development
of foul-smelling nasal discharge (Pyman 1971).
Plastic and other inert materials may be toler-
ated for a relatively long period until reactive
granulation tissue develops to produce a nasal
obstruction. A delay in diagnosis of button batter-
ies in the nasal cavity may lead to septal necrosis,
with perforation and destruction of the cartilage
causing deformity of the nose (saddle-nose defor-
mity). Most nasal FBs can be removed in the
Emergency Department by Emergency Medicine
Physicians with good visualization and appropri-
ate instrumentation. Long-standing objects such
as button batteries may be more complicated,
requiring assistance from ENT specialists.
25  Foreign Bodies
25.2.3 Throat
Throat FBs are most often encountered in chil-
dren between 4 and 8  years of age. The most
common presentation is local pain followed by
dysphagia, vomiting, drooling, crying, and
hematemesis. Most throat FBs are fish bones.
Other FBs include other bones (chicken and
pork) and toys. Throat examination can success-
fully visualize the FB in the throat warranting its
removal. Radiographs are indicated when the
FBs cannot be found. A lateral view of the neck
may be helpful to visualize and locate the
FB. However, less than 10% of radiographs con-
firm the diagnosis of the FB.  Often Emergency
Medicine Physicians are able to remove FBs in
the throat in the Emergency Department setting.
In one study, 23% of cases were discharged after
the successful removal of the FB in the Emergency
Department; the rest being referred to ENT spe-
cialists (Kim et al. 1973).
25.2.4 Larynx
Laryngeal FBs occur less frequently than in bron-
chial foreign bodies. They may be large and bulky
and present with acute, life-threatening obstruc-
tion. Smaller, thin, sharp, or triangular-­ shaped
objects are at risk of embedding in the laryngeal
mucosa or of becoming caught between the vocal
folds (Fig. 25.1). Some can wedge in the larynx
and cause an inflammatory reaction, mimicking
an infectious disease or incomplete upper airway
obstruction. The diagnosis may be difficult due to
presentation with nonspecific symptoms and
some objects not appearing on plain radiographs.
Flexible fiberoptic laryngoscopy is indicated in
any child suspected of having a laryngeal FB
whether it is a life-threatening clinical picture or
chronic upper airway symptoms.
25.2.5 Trachea and Bronchi
Children often play, run, and laugh while eating.
In addition, the mechanism that coordinates
swallowing and respiration is still not mature
327
Fig. 25.1  Eggshell as a foreign body in the larynx
(Courtesy of Dr. J. Manoukian)
enough to provide perfect control to avoid foods
or other materials from entering the respiratory
tract. Approximately 75% of tracheal and bron-
chial FBs in the children occur in those younger
than 3 years. Most FBs are small enough to pass
through the larynx and enter the trachea. Because
the trachea is relatively large beyond the level of
the cricoid cartilage, only a small percentage of
them (3–12%) lodge in the trachea. The main
stem bronchus is the most common area where
FBs become lodged.
Among children aged 5 years or younger, aspi-
ration of food items such as nuts, seeds, and grains
is more frequent than nonfood materials (79%
versus 15%). In contrast, children older than
5  years are far less likely to aspirate food items
(12%) compared to nonfood items (88%). In older
children, pins, thumbtacks, paper clips, and
screws account for over half of the cases. Children
have a tendency to put these objects in their mouth
as a temporary repository, while the hands are
otherwise occupied (for example, holding a pen
cap in the mouth while writing) the objects are
then accidentally aspirated. An accurate history
and a good physical exam are the key factors in
evaluating a child who has possibly aspirated an
FB.  Some children may remain asymptomatic,
while others may have a cardiopulmonary arrest.
A sudden choking episode, even unwitnessed, is
328
considered to be an important part of the history
which may lead to the correct diagnosis of FBs of
the airway. Any symptoms, including choking,
drooling, difficulty swallowing, fatigue, and
respiratory distress, may warrant evaluation with
laryngoscopy and bronchoscopy. The presence of
these symptoms with the right history has a sensi-
tivity ranging from 50 to 90% in various series.
The clinical triad of wheezing, coughing, and
diminished or absent breath sounds is often con-
sidered to be diagnostic of FB aspiration. Although
each of the individual components of the triad
may be present in 75% of the patients, all three are
present in only 31% of cases. Kim noted that the
triad was more often present in patients in which
the discovery of the airway FB was late. The diag-
nosis of an airway FB is often delayed, Wiseman
reported only 46% of patients were diagnosed on
the first day. Pyman described six distinct types of
clinical pictures depending on the degree of air-
way obstruction, location, and duration of the FB:
1. Sudden wheezy breathing (most frequent).
2. Cough alone with or without pyrexia.
3. Hemoptysis.
4. Stridor (seen only in cases of laryngeal and
subglottic FB).
5. Life-threatening asphyxia.
6. Patient with no symptoms.
a b
Fig. 25.2 (a) Hyperinflation of right lung secondary to a foreign body in the right mainstem bronchus. (b) Foreign
body identified on bronchoscopy Source: (Courtesy of Dr. J. Manoukian)
S. Shah et al.
Radiographic evaluation is helpful in a patient
suspected of having an airway FB. It must include
anteroposterior and lateral views of the extended
neck, and anteroposterior and lateral views of the
chest. Chest radiographs should be obtained dur-
ing both inspiration and expiration to appreciate
air trapping which is present in 62% of patients
when the bronchial FB acts as a one-way valve
(Figs. 25.2a and b). Other radiographic findings
may include mediastinal shift, radiopaque
objects, or atelectasis. In some late-diagnosed
airway FBs, radiographs may show signs of
pneumonia, lung abscess, bronchiectasis, and
empyema (Wiseman 1984; Pyman 1971).
However, radiographs may still be completely
normal in 18.5% of patients with an airway foreign
body. Computerized tomography (CT) scan can
also be helpful to evaluate the airway if the patient
is stable and further testing is warranted to possibly
avoid an unnecessary procedure. CT scans are con-
sidered 100% sensitive and 66.7% specific for
identifying an airway FB. However, laryngobron-
choscopy is the procedure of choice for both diag-
nosis and therapeutic removal of the airway FB.
In life-threatening asphyxia, laryngobron-
choscopy can be an emergency procedure. After
induction of general anesthesia, the patient is
placed in a supine position with the head
extended. A laryngoscope is first inserted to visu-
25  Foreign Bodies
alize and to expose the larynx for insertion of the
bronchoscope. Great care is taken to protect the
eyes, lips, teeth, tongue, and other laryngeal
structures during the bronchoscopy. The bron-
choscope is passed into the upper trachea, and
ventilation is performed through the broncho-
scope. The tracheobronchial tree is then com-
pletely inspected, and the FB is removed.
Failure of bronchoscopic removal of FBs may
result from:
–– Inexperienced endoscopist.
–– Poor vision associated with bleeding, granula-
tion tissue and edema.
–– Broken FB (especially nuts, vegetables) that
migrate distally and become impacted.
Fragmentation of these FBs at the time of
removal may be avoided by using Fogarty catheters
instead of forceps. The Fogarty balloon is passed
distally; the balloon is inflated and then withdrawn;
the FB can be retrieved in the oropharynx.
In some cases, a second attempt may be rea-
sonable if the first attempt was unsuccessful and
if the patient is clinically stable. In cases refrac-
tory to bronchoscopic retrieval, a thoracic
approach may be considered in some patients. In
addition, in the right setting, extracorporeal
membrane oxygenation (ECMO) availability
may need to be considered for surgical retrieval.
Potential complications of endoscopic
retrieval include pneumothorax and pulmonary
and airway hemorrhage. Patients that have an air-
way FB that has been lodged for a prolonged
period of time are at higher risk for hemorrhage
due to associated airway inflammation.
25.3 Upper Gastrointestinal Tract
Foreign Bodies
Ingestion of a FB is commonly encountered by
physicians working in the Emergency Room. The
management depends on the type of FB ingested,
its location along the gastrointestinal (GI) tract,
and the condition of the child. Eighty percent of all
patients seen after ingestion of an FB are children
(Stack and Munter 2006). Children aged between
329
6  months and 3  years are the most affected
(Orsagh-Yentis et al. 2019). Most ingested FBs
pass spontaneously. Estimated 40% of foreign
body ingestions in children are not witnessed, and
in many cases, the child never develops symptoms
(Chen and Beleri 2001). Objects with an irregular
shape that are very sharp, that are unusually long
or objects, and that can liberate toxic products like
button batteries may require removal.
Complications of FB ingestion include steno-
sis, perforation, tracheoesophageal fistula, bleed-
ing, and bowel obstruction. Although FB ingestion
is less dangerous than FB aspiration, they are the
cause of 1500 deaths per year in the U.S (Center
for Disease Control 2004).
Differential Diagnosis
Pharyngitis, esophagitis, esophageal stricture,
esophageal spasm, and achalasia.
25.3.1 Esophageal Foreign Bodies
The esophagus has three naturally narrow areas:
(1) the cricopharyngeus muscle, (2) the middle
third of the esophagus at the level of the left
mainstem bronchus, and (3) the lower esophageal
sphincter. FBs tend to lodge in one of these three
areas. Coins are by far the most frequent FB
swallowed by a child followed by fragments of
food materials, metal, plastic, pins, and bones. A
significant geographic and cultural influence
exists. For example, in Asia, the most frequent
FBs are fish bones, while in western countries,
coins may account for up to 89% of ingested FBs.
Twenty-two percent of patients older than
5 years with esophageal FBs were noted to have
anatomic abnormalities, such as a repaired
esophageal atresia, a vascular ring, a cartilagi-
nous rest, an esophageal stricture, and/or duplica-
tion cyst (Lee 2018).
On the other hand, a small-size FB lodged in
the esophagus may be completely asymptomatic.
Radiographic studies are the simplest method to
assess a child suspected of having an FB in the
esophagus. The plain radiograph should include
the neck, the chest, and the abdomen, and should
be both AP and lateral views (Fig. 25.3 and b).
330
a b
Fig. 25.3 (a) AP X-ray of a coin ingestion. (b) Lateral X-ray of a coin ingestion
Radiopaque FBs can be easily seen on plain
radiographs. If food materials or other radiolu-
cent FBs are suspected as the cause of dyspha-
gia and drooling, a barium esophagram is
indicated.
If an FB is lodged in the esophagus and not
detected early, it can cause complications.
Depending on the nature of the object and the
duration of its presence, these include:
–– Button batteries: esophageal burn, stricture,
perforation, aortoesophageal, tracheoesopha-
geal fistula, and retropharyngeal abscess.
–– Sharp objects: perforation, abscess, retropha-
ryngeal abscess, and mediastinitis.
–– Others: stricture and esophagitis.
Therefore, foreign bodies in the esophagus
should be removed. Three main techniques have
been described for the removal of FBs in the
esophagus:
–– Extraction by using Foley catheter balloon.
–– Bougienage.
–– Endoscopic retrieval.
S. Shah et al.
The first two techniques are limited to smooth
objects such as coins. The choice between the
three techniques depends on factors such as:
–– Size and shape of the FB.
–– History of esophageal abnormalities.
–– How long the FB has been lodged in the
esophagus.
–– Experience and preference of treating physician.
Radiopaque FBs can be easily seen on plain
radiographs. If food materials or other radiolu-
cent FBs are suspected as the cause of dysphagia
and drooling, a barium esophagram is indicated.
If an FB is lodged in the esophagus and not
detected early, it can cause complications.
Depending on the nature of the object and the
duration of its presence, these include:
–– Button batteries: esophageal burn, stricture,
perforation, aortoesophageal, tracheoesopha-
geal fistula, and retropharyngeal abscess.
–– Sharp objects: perforation, abscess, retropha-
ryngeal abscess, and mediastinitis.
–– Others: stricture and esophagitis.
25  Foreign Bodies
Therefore, foreign bodies in the esophagus
should be removed. Three main techniques have
been described for the removal of FBs in the
esophagus:
–– Extraction by using Foley catheter balloon.
–– Bougienage.
–– Endoscopic retrieval.
The first two techniques are limited to smooth
objects such as coins. The choice between the
three techniques depends on factors such as:
–– Size and shape of the FB.
–– History of esophageal abnormalities.
–– How long the FB has been lodged in the
esophagus.
–– Experience and preference of treating
physician.
Foley catheter retrieval is generally successful
for removing smooth objects like coins located in
the esophagus. It can be performed in an outpa-
tient setting with or without fluoroscopic guid-
ance. Full resuscitation equipment should be
available during the procedure. With the patient
lying down in lateral decubitus position, the
Foley catheter is inserted into the esophagus
through the mouth. Under fluoroscopy, the tip is
passed further down, beyond the location of the
FB.  The balloon is inflated with contrast and
carefully pulled back to bring the FB back into
the mouth, so it can be retrieved.
Success rates are reported up to 96%. This
technique is not typically recommended for a coin
that has been lodged in the esophagus for more
than 2–3 days due to the potential for impaction.
Bougienage is a simple method for pushing
smooth objects into the stomach with the expec-
tation that they will then be eliminated spontane-
ously (up to 95%). Bougienage is generally
recommended in a select group of patients:
–– A single coin or impacted meat ingested less
than 24 hours since the ingestion.
–– No esophageal abnormalities.
–– No respiratory distress.
Endoscopic retrieval under general anesthesia
is the most widely accepted technique for the safe
331
removal of sharp or impacted objects in the
esophagus with a success rate approaching 100%.
At the time of the retrieval, the esophagus and its
mucosa can be carefully inspected. And any asso-
ciated esophageal stenosis can be dilated in the
same setting.
For FBs that are present in the esophagus for
an unknown duration, endoscopic removal is the
only acceptable procedure.
A thoracic approach is sometimes required for
complications such as mediastinal FBs, an aorto-
esophageal fistula, tracheoesophageal fistula, or
mediastinitis.
25.3.1.1 Disk or Button Batteries
Disk or button batteries can cause damage to the
esophagus by direct corrosive effects, voltage
burns, and pressure necrosis. This is due to the
release of a low-voltage electric current or sec-
ondary to leakage of the alkali solution which
can lead to liquefactive necrosis. Injury to the
wall of the esophagus may happen within few
hours after the ingestion. They are more com-
monly ingested than cylindrical batteries by
young children.
Only 10% of all patients were symptomatic,
but fatalities have been reported. More than 3500
incidents of button battery ingestions are reported
each year to the US Poison Control Center.
Complications of esophageal button batteries
include pressure necrosis and burns to the esoph-
agus leading to stenosis or perforation, aorto-
esophageal fistula, and tracheoesophageal fistula
(American Association of Poison Control Centers
2001).
Button batteries are often mistaken for coins.
An anterioposterior and lateral chest radiograph
can help distinguish the two as button batteries
will have a double contour rim. (Fig. 25.4).
Emergency endoscopic removal of all esopha-
geal button batteries should be done, because burns
to the esophagus can occur as early as 2 hours after
ingestion. During the procedure, the esophagus is
carefully inspected. After the battery is removed, it
is important to perform a completion esophagos-
copy to reinspect the esophageal mucosa.
A current management algorithm published
by Rosenfeld et  al. on how to manage battery
ingestion is shown in Fig. 25.5 (Rosenfeld et al.
2019).
332 S. Shah et al.

Fig. 25.4  AP X-ray of

foreign body ingestion
that has a double
contour rim

Battery Ingestion
Treatment Algorithm
Immediate removal
Battery in YES (if significant
esophagus? corrosion consider
esophagram)
Inpatient
Observation
NO vs
Laproscopic or
YES Endoscopic Removal

YES NO
Battery in
Symptomatic? Symptomatic?
stomach?
(Beyond
stomach) NO Outpatient
management if
NO battery not passed in
YES 10-14 days check
AXR

Outpatient
NO AXR in 5-7
days is
YES
Endoscopic removal battery
beyond
stomach?

Fig. 25.5  Algorithm for management and treatment of battery ingestion

25.3.1.2 Coins in the Esophagus tomatic. In asymptomatic patients with a coin in

Coins lodged in the upper two-thirds of the
esophagus will require removal as the spontane-
ous passage is very rare.
However, most esophageal coins are in the lower
esophagus and these patients are usually asymp-
the lower esophagus, a conservative approach may
be used for the first 24 hours. This will consist of
serial radiographs. Surprisingly, in up to 89% of
patients, the coin passes spontaneously to the stom-
ach (Figs. 25.6a and b) (Uyemura 2005).
25  Foreign Bodies 333

a b

Fig. 25.6 (a) Coin lodged in the lower esophagus in an asymptomatic patient. (b) Spontaneous passage to the bowel
after 24 h of observation

25.4 Subdiaphragmatic Foreign

Bodies

Smooth or blunt objects that reach the stomach

will usually be evacuated spontaneously without
complications. In one series of 1259 children
who had ingested FBs of all kinds in the subdia-
phragmatic region, 97% of the FBs were passed
spontaneously. Although the incidence of inter-
vention for all ingested FBs is only about 1%, the
rate for ingested sharp objects is 15–30%. These
include toothpicks, bones, nails, pins, needles,
and sharp toys.
Objects longer than 5 cm tend to lodge in the
stomach or duodenum, because they cannot nego-
tiate through the C loop of the duodenum. Fig. 25.7  Large phytobezoar extracted from the stomach
during surgery
Therefore, they will need to be removed by endos-
copy. For infants, objects larger than 2  cm in
diameter can have difficulty traversing the pyloric by plants and vegetable matter (phytobezoars)
channel and thus may require endoscopic retrieval (Fig. 25.7), by hair (trichobezoars), or by lactose
(Kramer et al. 2015). Bezoars are FBs that have (lactobezoars). These can cause obstruction
accumulated over time in the GI tract and often symptoms, failure to thrive, vomiting, nausea, and
occur in the stomach. They are commonly formed abdominal pain. Bezoars can be retrieved by
334
endoscopy or eliminated by enzymatic fragmen-
tation. Laparoscopy or laparotomy may be neces-
sary to remove very large bezoars.
For button batteries lodged in the stomach, the
corrosive effects are less common than when
lodged in the esophagus. Most of these patients
can be followed by serial radiographs for
24 hours. Upper endoscopy to remove the battery
is indicated if the battery does not pass through
the stomach within 24  hours or if the patient is
symptomatic. Symptoms may include nausea,
vomiting, or abdominal pain. The risk of compli-
cations increases when the battery is still charged.
This may cause the batteries to adhere to the gas-
tric wall and cause ulcerations.
Intestinal FBs rarely cause complications. The
most common location for the failure of progres-
sion is the ileocecal valve, the only point of ana-
tomic intestinal narrowing. Most of the time, the
FBs, especially smooth objects, can be passed
without difficulty. Sharp objects carry the risk of
perforation; however, most still pass spontane-
ously. Sharp object ingestions should be followed
with serial abdominal exams and radiographs.
Complications such as perforation and bowel
obstruction have been reported. If there is the
failure of progression of an intestinal FB on serial
radiographs, then surgery may be indicated.
Differential Diagnosis
Appendicitis, intussusception, bowel obstruction,
peptic ulcer disease, gastroenteritis, pancreatitis,
and cholecystitis.
a b
Fig. 25.8 (a) X-ray demonstrating magnetic buckyballs together, (b) enteroenteric fistula, and (c) retrieval of the string
of magnet balls
S. Shah et al.
25.4.1 Magnetic Foreign Bodies
Approximately 1600 pediatric cases of rare-earth
magnet ingestion occurred in the US in 2019,
which is a sixfold increase from 2016 following a
reversal of a 4-year ban on these magnetized balls
(Frankel 2019). These rare-earth magnets are
uniquely dangerous and recognized to be ten times
stronger than a refrigerator magnet. The Consumer
Product Safety Commission, initially responsible
for the product ban, has set out a task force to
make these magnets safe and less appealing to
children (Centers for Disease Control and
Prevention 2006). Such magnets are dangerous
when separated in the gastrointestinal tract. If sep-
arated, the magnets may attract across adjacent
segments of the bowel causing enteroenteric fistu-
lae complicated by perforation or bowel obstruc-
tion (Fig. 25.8a, b, c). Management of these cases
ranges from observation to surgery. Some studies
have found that up to 40% of magnet ingestions
will require an endoscopic or abdominal surgical
intervention (Waters et al. 2015).
Sola et al. published an algorithm to help guide
operative and nonoperative management of mag-
net ingestion (Fig. 25.9) (Sola Jr. et al. 2018).
25.4.2 Genitourinary Foreign Bodies
Vaginal FBs are most frequent between the age of
5 and 8 years. Vaginal bleeding and blood-stained
or foul-smelling secretions are the main clinical
c
25  Foreign Bodies 335

Magnet in YES Non-urgent removal

esophagus? (No esophagram)
Endoscopic removal

NO
YES

Outpatient
Multiple magnets management with
In stomach?
YES or magent +2nd NO bowel regimen. If
metallic object magnet not passed
co-ingestion?* in 14 days check
AXR.

(Beyond
NO
stomach)
YES Laparoscopic
removal

Symptomatic or
obstructive signs
on AXR?
Inpatient observation If failure to progress
with surgical in 48 hours consider
consultation, bowel laproscopic or
regimen and serial colonoscopic
NO AXRs removal

*If a single magnet vs. multiple magnet ingestion cannot be definitively differentiated by history and radiographic findings, then the
patient should be treated as an inpatient for suspicion of ingestion of multiple magnets.

Fig. 25.9  Algorithm for management and treatment of magnet ingestion

manifestations of foreign objects in the vagina. 25.5 Conclusion

Most vaginal FBs are not radiopaque. Therefore,
examination under anesthesia with a vaginoscope
is essential. It allows for both diagnosis and for
removal. The presence of vaginal FB may also be
an indication of sexual abuse. In adolescents, a
vaginal foreign body may be complicated by a
vesicovaginal fistula.
Self-insertion of FBs into the urethra and the
bladder is extremely rare in children. However,
in adolescents and in young adults, especially in
patients with psychiatric disorders, a multitude
of objects have been reported requiring removal
from the urethra and/or bladder. Some such
objects include a needle, pin, portion of a can-
dle, thermometer, toothbrush, and metal hook.
To avoid embarrassment, patients tend to seek
treatment late, often waiting until the problem
becomes symptomatic from urethritis, cystitis,
or hematuria. Cystoscopy gives the diagnosis in
difficult cases and allows for treatment.
Foreign body aspiration and ingestion are com-
mon in children and can be asymptomatic. The
diagnosis and management depend on the ana-
tomic location of the foreign object. Some objects
such as magnets and batteries have potentially
serious complications. Because of this, the
workup and treatment algorithms were created to
facilitate healthcare providers to manage these
ingestions in a timely manner.
References
American Association of Poison Control Centers (2001)
Annual Report 19; 5:335–395
Center for Disease Control (2004) 53 (SS07); 1–57
Centers for Disease Control and Prevention (CDC) (2006)
Gastrointestinal injuries from magnet ingestion in
children  – United States, 2003–2006. MMWR Morb
Mortal Wkly Rep. 55(48):1296–1300
336
Chapin MM, Rochette LM, Annest JL et  al (2013)
Nonfatal choking on food among children 14
years or younger in the United States, 2001–2009.
Pediatrics
Chen MK, Beleri EA (2001) Gastrointestinal foreign bod-
ies. Pediatr Ann 30:736–742
Frankel T (2019) Number of children swallowing danger-
ous magnets surges as industry largely polices itself.
Washington Post. Dec 25 2019
Kim IG, Brummitt WM, Humphry A et al (1973) Foreign
body in the airway. Laryngoscope 83:347–354
Kramer RE, Lerner DG, Lin T et al (2015) Management
of Ingested Foreign Bodies in Children: a clini-
cal report of NASPGHAN Endoscopy Committee. J
Pediatr Gastroenterol Nutr 60(4):562–74
Lee JH (2018) Foreign Body Ingestion in Children. Clin
Endosc 51(2):129–136
Lelli JL Jr (2005) Foreign bodies. In: Ashcraft KW,
Holcomb GW III, Murphy JF (eds) Pediatric Surgery.
Elsevier Saunders, Philadelphia, pp 137–145
Orsagh-Yentis D, McAdams RJ, Roberts KJ et al (2019)
Foreign-body ingestions of young children treated in
S. Shah et al.
US Emergency departments: 1995–2015. Pediatrics.
143(5)
Puri P, Höllwarth ME (eds) (2006) Pediatric surgery.
Springer, Berlin, Heidelberg
Pyman C (1971) Inhaled foreign bodies in childhood. J
Otolaryngol Soc Aust 3:170–180
Rosenfeld EH, Sola R, Yu Y (2019) Battery ingestions in
children: variations in care and development of a clini-
cal algorithm. J Pediatr Surg 53(8):1537–1541
Sola R Jr, Rosenfeld EH, Yu YR et al (2018) Magnet for-
eign body ingestion: rare occurrence but big conse-
quences. J Pediatr Surg 53(9):1815–1819
Stack LB, Munter DW (2006) Foreign bodies in the
gastrointes-­
tinal tract. Emerg Med Clin N Am
14:493–521
Uyemura M (2005) Foreign body ingestion in children.
Am Fam Physician 72(2):287–291
Waters AM, Teitelbaum DH, Thorn V et al (2015) Surgical
management and morbidity of pediatric magnet inges-
tions. J Surg Rese 199(1):137–140
Wiseman NE (1984) The diagnosis of foreign body aspi-
ration in childhood. J Pediatr Surg 19:531–535
 Physical and Sexual Child Abuse
Michael E. Höllwarth
26.1 Introduction
There are many aspects of child abuse, with the
most common forms being physical and sexual
abuse, emotional or physical neglect, and
Munchausen by proxy syndrome. However, all
aspects of intentional or unintentional malnutri-
tion and hunger in childhood on the one hand, as
well as the excess of everything possible on the
other, should be included in the list of child
abuse. In this chapter, we refer primarily to the
symptoms, clinical signs, and diagnostic proce-
dures in children suspicious for physical or sex-
ual child abuse and for Munchausen by proxy
syndrome.
26.2 Historical Overview
Child abuse is as old as mankind (deMause 1998;
Lynch 1985; Lascaratos and Rebelakou 2000;
Thearle and Gregory 1988). There are probably
only a few ethnic groups among which children
had the same value and position as any other
member of the community (deMause 2009).
Over the centuries, children were often seen as
low-value members of society, and as a result,
abuse was not recognized as a criminal act
M. E. Höllwarth (*)
University Clinic of Paediatric and Adolescent
Surgery, Medical University of Graz, Graz, Austria
e-mail: michael.hoellwarth@medunigraz.at
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_26
26
(Adams 2016, 2017). Clerical sexual abuse is
well known in the catholic church over the centu-
ries (Rashid and Barron 2019). In 1962, Kempe
et  al. published a paper with the title “The bat-
tered child syndrome” that gained international
interest (Kempe et al. 1962). But still today, child
abuse occurs in nearly all countries, unrelated to
whether they are in a so-called developing state
or highly industrialized and they still present a
major public health problem.
26.3 Incidence
The real incidence of child abuse is difficult to
estimate due to the rather high number of unre-
ported cases, especially when sexual abuse occurs.
In 2018, the U.S.  Department of Health and
Human Services studied 678,000 victims (+4%
when compared to 2014). A total of 3,534,000
children received a child protective service inves-
tigation. The majority of victims suffered from a
single type of maltreatment (84.5%); 60% were
neglected only, 10.7% were physically abused,
and 7% were sexually abused. 15.5% were vic-
tims of more maltreatment types. The youngest
children are the most vulnerable to maltreatment
and 28.7% were younger than 3  years. Parents
were the perpetrators in 77% of child fatalities.
More than 80% of the perpetrators were between
the ages of 18 and 44 years. The report estimates
that 1770 died of abuse or neglect corresponding
337
338
to 2.39 per 100,000 children. Of these deaths,
41% occurred in children under 1 year of age, and
85% occurred in children under 6  years of age.
Factors that are typically associated with mal-
treatment are low socioeconomic status, violence
in the family, breakdown of families, chronic
child morbidity, parental mental health issues, and
parents who were abused during their own child-
hood (Child Maltreatment Report 2018 of the
U.S. Department of Health and Human Services).
A meta-analysis of childhood sexual abuse esti-
mated 127/1000  in self-reported studies and
4/1000  in informant studies. Sexual abuse was
more common among females when compared to
males. The lowest rates were found in Asia; the
highest reported for females in Australia and for
males in Africa (Stoltenborgh et al. 2011).
The UNICEF Report 2014 “Hidden in Plain
Sight: A statistical analysis of violence against
children” is the largest-ever compilation of data
on the subject of violence against children. The
report sheds light on the prevalence of different
forms of violence against children, with global
figures and data from 190 countries. In this
report, it is estimated that in 2012, 95,000 chil-
dren and adolescents younger than 20 years were
victims of homicide worldwide. Furthermore, on
average, six in ten children worldwide between
the ages of 2 and 14 years are subjected to corpo-
ral punishment and psychological aggression by
their caregivers. One in ten girls (120 million
worldwide) has experienced sexual abuse at an
age between 10 and 14  years. According to
Letourneau, the economic costs of sexual abuse
in the United States are approximately $9.3 bil-
lion, estimating 20 cases of fatal and 40,387 non-
fatal cases in 2015 (Letourneau et al. 2018).
A retrospective study, using the Pediatric
Hospital Information System of 38 hospitals,
reported that the rates of physical abuse and non-
accidental severe head injury increased by 0.79%
and 3.1% per year between 2000 and 2009, while
all-cause injuries rates declined by 0.8% per year
(Wood et al. 2012).
Recently, child maltreatment in Germany was
investigated with a questionnaire including 2510
participants between 14 and 94 years, and results
were published by Witt et al. (2017). The results
M. E. Höllwarth
showed that 2.6% reported severe emotional
abuse, 3.3% severe physical abuse, 2.3% severe
sexual abuse, and 7.1% severe emotional neglect.
26.4 Physical Abuse
Physical abuse is not accidental, but intentional
and violent maltreatment caused either by harmful
actions or by the omission of helpful support by a
person who is responsible for the child’s welfare,
leading to a temporary or permanent injury of a
child under the age of 18 (UNICEF Report 2014).
The annual statistics of the Child Protection
Group of the Departments of Pediatrics and
Pediatric and Adolescent Surgery in Graz/
Medical University showed that in three-quarters
of cases, the children were younger than 10 years
with the highest incidence between 0 and 4 years
(Fig. 26.1). The primary diagnoses in two-thirds
of the patients after admission to the hospital are
injuries, while symptoms of somatization or sui-
cide are comparably rare—in contrast to patients
with a history of sexual abuse (Fig. 26.2). Typical
Fig. 26.1 Age distribution in children after physical
abuse shows the highest incidence between 0 and 4 years
(Höllwarth ME.; Physical and Sexual Child Abuse. In Puri
P., Höllwarth ME, editors. Pediatric Surgery: Diagnosis
and Management. Heidelberg, Springer; 2009)
others 16% somatisation 16%
suicide 5%
injuries 63%
Fig. 26.2  The leading symptoms in patients with physi-
cal abuse are injuries, while suicide or somatization of
symptoms is comparably rare
26  Physical and Sexual Child Abuse 339

risk factors for child abuse in this age group are

very young and inappropriately reacting mothers
or fathers, who have difficulties in coping with
stressful situations, or parents overwhelmed with
problems, such as premature babies, handicapped
children, constantly screaming babies, or difficult
living and social circumstances.
The history presented by the caregivers is
often vague and may change from 1  day to the
other, or it is not consistent with the age of the
child. It has been shown that the absence of a his-
tory of trauma, changes in history, or a history
that blamed home resuscitative efforts is strongly
associated with child abuse. Furthermore, the
experience has shown that child abuse is likely
when the reported trauma does not match with
the severity, pattern, or localization of injury. If
an injury is suspicious for physical abuse, a care-
ful anamnesis by a person who is experienced in
child abuse cases and has the skills of interview-
ing is important (Palusci and Fischer 2012).
It is of utmost importance to know which type
of behavior in caregivers or which posttraumatic
findings raise a red flag regarding physical abuse
of a child. The most suspicious behaviors of care-
givers and findings are listed in Table 26.1. Fig. 26.3  Multiple cutaneous injuries and bruising with
hematoma. The boy was beaten with a belt. (Höllwarth
ME, Physical and Sexual Abuse. In PuriP, Höllwarth M,
26.4.1 Soft-Tissue Injuries editors. Pediatric Surgery: Diagnosis and Management.
Heidelberg, Springer; 2009)

The most common manifestations of child abuse are

cutaneous injuries. The physical examination of the
patient may show multiple cutaneous injuries, with
Table 26.1  Most suspicious behaviors of caregivers and
findings
Suspicious behavior of the caregivers
Often changing doctors or hospitals
Delayed medical consultation
Changing history presented from the caregivers
Injury has not been observed at all and there is no
explanation
The demeanor of the caregivers is inadequate
Suspicious findings at the exam of the child
Injuries not typical for the age of the child
Several injuries of different age
Formed injuries or marks
Sudden and unexpected change of the behavior of the
child
swellings of the soft tissue or diffuse or formed
bruises, abrasions, lacerations, petechiae, and burns,
often of different age (Figs.  26.3 and 26.4).
Accidental injuries in infants and children are usu-
ally localized around the prominent parts of the
body—forehead, chin, hips, shins, and anterior or
lateral parts of the extremities. Thus, suspicious inju-
ries are localized on the side of the face, ears and
neck, trunk, and inside or dorsal side of arms or legs,
unless the history of the accident is appropriate.
As regards bruises, the inflicting instruments
often may be discerned from the shape of the skin
lesion. The human hand leaves parallel lines rep-
resenting the spaces between the fingers.
Cutaneous lesions of the upper arms or the thorax
may result from hard pressure applied during
violent handling.
340
Fig. 26.4  Bruising on the frontal lateral side of the fore-
head. A punch of hear has been tuned out in this child.
Höllwarth ME. Physical and Sexual Child Abuse. In Puri
P. Höllwarth M, editors. Pediatric Surgery: Diagnosis and
Management. Heidelberg, Springer
In adult bites, only one arch is often seen,
which usually reaches from one canine tooth to
the other, while child bites can reach the second
primary molar and involves both arches.
Photographs should be taken and saline swabs
should be used to pick up saliva for DNA analy-
sis. Ligatures typically cause lesions around the
neck, the wrists, or the ankles.
Burns represent around 10% of all physical
abuse cases. The peak age of burn victims is from
13 to 24 months. Cigarette burns account for up
to 5% of burns. They are round, often in clusters
(Fig.  26.5). The classic lesions occur when the
child is held in extremely hot water, resulting in
sharp demarcation lines on the margin of the hot
water-skin contact line—e.g., stocking or glove
immersion burns of the feet or hands. Immersion
in the bathtub produces a typical doughnut lesion
of the buttocks, sparing the parts that are in direct
contact with the base of the tub. In contrast, if a
M. E. Höllwarth
Fig. 26.5  Typical cigarette burn close to the breast
child unintentionally enters a hot bathtub, the
resulting burn has no sharp edges and is
unilateral.
26.4.2 Skeletal Injuries
Skeletal trauma is the second most common con-
sequence of child abuse. A skeletal survey in
children under 2 years is mandatory in all cases
of suspected physical abuse. In older children, we
perform a bone scan to detect older skeletal inju-
ries, which may not be seen on plain radiographs.
80% of abuse fractures are seen in children
younger than 18 months; 25% of them are skull
fractures.
Fractures of the extremities occur in 44%.
Accidental fractures in children under 2 years of
age do occur, but are relatively rare in compari-
son to older children. Among them, the most fre-
quently injured long bones are the femur, the
humerus, and the tibia. Metaphyseal lesions of
the long bones are highly suspicious signs of
26  Physical and Sexual Child Abuse 341

due to acceleration or deceleration forces during

shaking of the baby or twisting an individual
limb—“the limbs are handles for mishandling the
child”.
Rib fractures may be diagnosed only coinci-
dentally when chest radiographs are obtained for
other reasons. In infants, they are always caused
by violence against the child, unless the trauma
has a consistent history, e.g., a severe car acci-
dent. These fractures in child abuse are often
bilateral and are usually localized close to the rib
head or neck at the dorsal part of the thorax, or in
the lateral parts of the ribs (Fig. 26.7). It has been
demonstrated that they are caused by anterior–
posterior compression of the thoracic cage by
placing the thumbs in front, the palms laterally,
and the fingers along the spine during shaking or
other abusive acts.

Fig. 26.6 Corner fracture at the distal metaphysio-­

epiphysial region of the femur

child abuse (Fig. 26.6). They occur very close to

the growth plate and are known as bucket handle
or corner fractures. Sites of predilection are the
distal femur, proximal and distal tibiae and fibu-
lae, and proximal humerus. They occur due to
acceleration or deceleration forces during shak-
ing the baby or twisting an individual limb—“the
limbs are handles for mishandling the child”.
Lesions of the long bones are highly suspicious
signs of child abuse (Fig. 26.6). They occur very
close to the growth plate and are known as bucket
handle or corner fractures. Sites of predilection
are the distal femur, proximal and distal tibiae Fig. 26.7  Old rib fracture with a visible callus in an
and fibulae, and proximal humerus. They occur infant. This injury is typical for child abuse in infancy
342 M. E. Höllwarth

Skull fractures can be simple or multiple and

cross suture lines. The latter fractures are more
likely to be associated with abuse or caused by a
high-energy injury. The explanation of the injury
by the caregivers should be consistent with the
type of the fracture, but—in contrast to metaphy-
seal fractures or rib fractures—it is a difficult task
to decide whether or not it is a consequence of
child abuse.

26.4.3 Shaken Baby Syndrome

Shaken Baby Syndrome (SBS) occurs in babies

usually younger than 6  months, but it has also
been described in significantly older children.
The injury results from violently shaking the
baby, with or without additional impact (Mian
et  al. 2015). The usual trigger for shaking is
inconsolable crying by the infant. A study has
shown that in infants aged 6 months, nearly 6%
of parents reported taking at least one action to
stop infants crying by smothering, slapping, or
shaking. One in five had taken more than one of
these three actions and the cumulative rates rose
steadily from 1 to 6 months and were highest for
shaking. The risks were highest for parents from
nonindustrialized countries, those with either no
job or a job with short working hours, and those
who judged their infant’s crying to be excessive.
The first description of a shaken baby syn-
drome was published by Guthkelch in 1971, who
wrote that the injury resulted in a triad of subdu-
ral hematoma, retinal hemorrhages, and encepha-
lopathy (Guthkelch 1971). The shaking results in
a direct trauma to the brain by pushing against
the skull due to constant acceleration/decelera-
tion by severely shaking the baby, resulting in
tears of the subdural veins leading to a subdural
hematoma or subarachnoid hemorrhages, in
breaking of axons in deeper brain structures and
in a reduced oxygen supply to the brain. These
effects have been considered controversial and a
large Swedish meta-analysis showed no direct
evidence for this triad (Lynoe et al. 2017; Cohen
2019). Additionally, the trauma may cause sig-
Fig. 26.8  Extensive subdural hematoma in a case of a
shaken baby syndrome (SBS)
nificant injury to the brainstem, leading to anoxic
encephalopathy leading to severe neurological
symptoms, such as lethargy, irritability, convul-
sions, and respiratory arrest. A CT scan is the first
imaging modality to prove or exclude intracranial
lesions and skull fractures (Fig. 26.8). Associated
injuries can be seen, such as bruising of the tho-
racic skin and rib fractures. Retinal hemorrhages
(RH) are the most commonly found additional
lesions in SBS, characterized by extensive hem-
orrhages involving several layers of the retina. In
contrast, vaginal delivery may cause fine prereti-
nal petechiae, which usually resolves without
residuals within 10–14  days after delivery. Our
current strategy is to confirm an SBS in babies
with unexplained subdural hematoma—with or
without a fracture—when an RH is present.
Bilateral macular retinoschisis is associated with
a sevenfold and unilateral with a fourfold increase
of poor neurological outcome or death (Chong
et al. 2018).
26  Physical and Sexual Child Abuse
26.4.4 Abdominal and Thoracic
Injuries
Abdominal injuries are the second most common
cause of fatality in child physical abuse. The his-
tory from the caregivers is usually very vague and
they seek medical care rather late. The injuries
are more common in younger children and mor-
tality is significantly higher (53%), when com-
pared with unintentional accidents. In most cases,
the trauma results from violent hits with the fists
or the legs into the middle part of the abdomen.
Therefore, injuries to hollow viscera in the
midline (stomach, small bowel, and bladder)
­ group for abused children”, which consists of
occur more often than injuries to solid organs, in
contrast to other accidents.
Pharyngeal, hypopharyngeal, and esopha-
geal injuries are rare manifestations of child
abuse, due to perforation with blunt instruments.
Radiographs show subcutaneous or retropha-
ryngeal or mediastinal emphysema. In half of
the reported cases, additional rib fractures are
present.
26.4.5 Prevention Strategies
in Physical Abuse
An important prerequisite for effective preven-
tion of physical child abuse is to recognize endan-
gered persons or families. In those family
systems, shown in Table 26.2, an increased risk
for child abuse exists. Early recognition allows
early support for endangered families and
strengthens their abilities for childrearing. Family
Table 26.2  Endangered families
– Families with hyperactive children
– Inconsolable infant crying
– Child abuse in the family history
– A very dominant family member
– Background of permissive violence
– Parental mental ill-health
– Low socioeconomic status
– Unemployment and financial problems
– Drug abuse
– Transcultural problems between family members
– Social isolation
343
doctors, pediatricians, and gynecologists are the
professions that may recognize endangered fami-
lies. At the time when a family has a new baby,
there is “a window of opportunity” to support
parenting abilities. Many studies have docu-
mented the effectiveness of home-visiting pro-
grams and early intervention, which are
performed in most countries by social worker
organizations. If a child abuse case has already
occurred and has been documented in the chil-
dren’s hospital, we try to establish a firm and
trustful relation to the family. All children’s hos-
pitals in Austria must have, by law, a “Working
pediatricians and pediatric surgeons, psycholo-
gists, pediatric nurses, and social workers. This is
a multidisciplinary team that takes the responsi-
bility of appropriate evaluation and adequate
strategy to achieve the best result for the child
(Herbert and Bromfield 2019). The treatment
strategies include education, paternal support,
psychological help, and community support pro-
grams. However, these programs only come into
reality if the family accepts therapeutic measures
and help, as well as control, such as home-­visiting
programs. If these programs turn out to be unsuc-
cessful or if the family refuses to cooperate with
the social system of the appropriate State
Organizations, it becomes necessary to evaluate
for an adequate and safe place for the child/chil-
dren outside of the family, to prevent ongoing
physical abuse of the child/children.
26.5 Sexual Abuse
Sexual abuse is the encroachment of immature
children or adolescents for sexual actions, which
they cannot completely understand, and are
therefore not able to agree to, because they do not
have the ability to understand the significance of
consent and/or sexual actions yet, which offend
the family taboos (WHO). It is a universal prob-
lem, occurs in all ages and in all socioeconomic
classes, and usually has life-long consequences
for the children involved (Singh et al. 2014).
In contrast to physical abuse, the average age
of the victims is significantly older and most of
344
the children have no signs of genital or anal
injury if they are not examined acutely (Adams
et  al. 2018). The presenting symptoms of the
patients are usually unspecific, such as sleep dis-
turbances, abdominal pain, constipation, enure-
sis, weight loss, and phobias, and unexplained
change of the behavior of the child (Figs.  26.9
and 26.10) (Vrolijk-Bosschaart et  al. 2017;
Adams et al. 2018).
Diagnosis and treatment of these children are
a very difficult problem for the family doctor or
pediatrician in private practice, due to the involve-
ment of family members. A single person is
always overcharged with the circumstances
around sexual abuse, even if he/she is familiar
with the problem and well trained in recognizing
the subtle symptoms.
Therefore, it is indicated in all suspicious
cases of sexual abuse to transfer the patient under
an unspecific diagnosis to a specialized pediatric
center with a specialized team. Evaluation of vic-
tims of sexual abuse involves careful question-
ing, individualized psychological techniques and
tests, evidence collection procedures for forensic
purposes, and specialized examination tech-
niques, which are best provided by a group of
trained staff with expertise in the evaluation and
treatment of these children (Hauet-Wiedemann
Fig. 26.9  Age distribution in patients with sexual abuse
Fig. 26.10  Typical complaints when a child is admitted
after sexual abuse. Somatisation symptoms are leading
M. E. Höllwarth
et  al. 2018; Joki-Erkkilä et  al. 2018; Vrolijk-­
Bosscharart et  al. 2018; Herbert and Bromfield
2019).
A careful evaluation of the patient’s history
and complaints will often yield important infor-
mation. The interviews with the patient and the
caregiver need a quiet environment and a good
understanding of the child’s needs appropriate to
the age.
A full physical examination is always indi-
cated when abuse or neglect is suspected. We
recommend performing the examination of the
anogenital region of children up to 10–12 years
under anesthesia, unless the patient agrees to
have that examination under normal circum-
stances in a quiet atmosphere, using interna-
tional guidelines (Adams et al. 2016, 2018). The
inspection of the genitalia should be done as
gentle as possible. Colposcopy and documenta-
tion with photographs are necessary, while spec-
ulum and digital examination are usually not
required. It is important to know that in most of
the abused children, the anogenital aspects and
findings are normal, which does not exclude sex-
ual abuse at all. Signs of acute damage disappear
rapidly and even tears in the posterior fourchette
can heal with minimal scar tissue. Tears in the
hymen can only be seen if penetration happened,
but a recent study of pregnant adolescents
showed that only 2 of 36 had evidence of pene-
tration. Suspicious findings are vulvovaginitis,
recurrent urinary tract infection, and sexually
transmitted diseases. However, most causes for
vulvovaginitis are poor hygiene and thread-
worms, but the possibility of sexual abuse must
be considered if vulvovaginitis is persistent or
recurrent after adequate treatment. Similarly,
most recurrent urinary tract infections are caused
by anomalies of the urogenital tract.
Vaginal and perianal sampling of swab, blood
tests, and secretions is important. The presence of
semen, sperm, acid phosphatase, or a positive
culture for gonococci, Chlamydia trachomatis, or
positive serologic tests for syphilis or HIV make
the diagnosis of sexual abuse nearly certain.
Positive findings for Trichomonas vaginalis, ano-
genital warts, and papillomas—type HPV-6,
HPV-11, HPV-16, and HPV-18—and genital
26  Physical and Sexual Child Abuse 345

Table 26.3  In most cases, the sexual abuse occurs within to put the child under massive emotional pressure
the family by a relative, or a stepfather of the child to retract everything that has been said. In cases
Perpetuator n % where the results of the examinations are nega-
Biological father 27 14 tive, it is difficult to provide significant help to
Adoptive father 1 1
the child. Only regular “medical” controls may
Stepfather 15 8
Biological mother 1 1
protect it from further abuse. When clear evi-
Biological brother 23 12 dence exists, involvement in the civil, juvenile, or
Stepbrother 2 1 family court system is necessary and the child
Biological sister 3 2 must be protected in a safe environment.
Biological grandfather 8 4
Step grandfather 3 2
Uncle 48 25 26.6 Munchausen By Proxy
Male cousin 30 16
Female cousin 3 2
Munchausen by proxy (MBP) is defined as a situ-
Brother-in-law 7 4
ation when the illness of the child is simulated by
Other male relative 15 8
Other female relative 3 2 the caregiver and the child is presented several
Total 189 times to one or more different healthcare systems
for medical assessment (Abeln and Love 2018).
This syndrome describes a strange form of child
localization of herpes simplex are suspicious for abuse and defines circumstances in which:
sexual abuse.
When a child is sexually abused by a person (a) Illness of the child is simulated or produced
outside the family, there is the outrage on the part by a parent (mostly the mother).
of the parents, with emphasis on the guilt and (b) The child is brought persistently for medical
responsibility of the abuser. A difficult problem assessment and care, often resulting in mul-
exists when parents are in the process of divorce tiple medical procedures.
and allege that their child is being sexually (c) Knowledge about the etiology is denied by
abused by the other parent during custodial visits. the perpetrator.
However, in most cases, incest occurs in a seem- (d) Acute symptoms and signs in the child abate
ingly normal family (Table 26.3). when it is separated from the perpetrator.
The experience shows that in many situations,
the “non-offending” caregivers—most often the Most of the abusers are female, and in over
mothers—deny any knowledge of what was 95% of cases, it is the mother (Yates and Bass
­happening, sometimes for years. Thus, it seems 2017). The perpetrators are often familiar with
likely that either subconscious forces block the medical needs and agree easily with all proposed
ability to acknowledge what happened and may clinical investigations, even surgical procedures.
be dictated by the emotions and the shame which The mothers have often had obstetric complica-
comes over the family—“that cannot happen in tions (23.5%), or a history of child maltreatment
my family”—or the family members are under (30%) (Yates and Bass 2017). The difficulty
such massive pressure by the perpetrator that comes from the problem of gathering evidence
they do not dare to consider the reality. Denial of supporting the theory that the illness of the child
events is often the first defense. Thus, most often is simulated or fake. Careful findings and detailed
neither, the nonoffending family members nor notes in the charts are essential to finally formu-
the child inform about suspected child abuse. late the diagnosis when the case is brought to
Sometimes, the child may start to slowly confide court. A clear understanding of the parent’s psy-
in one staff member on the ward and to tell him/ chopathology is still not available and more
her some secrets. However, as soon as the family information is needed about the long-term results
is confronted with the suspicion, they often begin for these children.
346 M. E. Höllwarth

26.7 Conclusion Hauet-Wiedemann M et  al (2018) Agressions sexualles

sur mineurs: Étude d’une population d’enfants ayant
consulté dans une unite medico-judiciaire hospitalière
In conclusion, cases of physical and sexual abuse entre 2011 et 2015. Arch Pediatr 25:247–250
are some of the most difficult problems—clini- Herbert JL, Bromfield L (2019) Better together? Are view
cally and emotionally—in pediatric medicine. of evidence for multi-disciplinary teams responding
to physical and sexual child abuse. Trauma Violence
There is an absolute need that these children are Abuse 20:214–228
treated by a team of experts, first to come to a Joki-Erkkilä M et al (2018) Child sexual abuse – initial sus-
definite diagnosis, and second to be able to help picion and legal outcome. Forensic Sci Int 291:39–43
the child—with or without the family (Herbert Kempe C et  al (1962) The battered child syndrome.
JAMA 181:17–24
and Bromfield 2019). Prevention strategies Lascaratos J, Rebelakou EP (2000) Child sexual abuse:
include the early recognition of endangered fami- historical cases in the Byzantine Empire (324–1453
lies for physical abuse and appropriate school-­ A.D.). Child Abuse Negl 24:1085–1090
based programs to teach the children how to Letourneau EJ et  al (2018) The economic burden of
sexual abuse in the United States. Child Abuse Negl
recognize impermissible kinds of physical touch- 79:413–422
ing, to fend off sexual advances, and to report Lynch MA (1985) Child abuse before Kempe: an histori-
abuse to a trusted person. These programs are cal literature review. Child Abuse Negl 9:7–15
effective in increasing children’s knowledge and Lynoe N et  al (2017) Insufficient evidence for “shaken
baby syndrome” – a systematic review. Acta Paediatr
skills. Good epidemiological data are needed 106:1021–1027
based on annual nationwide studies, not only to Mian M et  al (2015) Shaken baby syndrome: a review.
explore the incidence but even more importantly, Fetal Pediatr Pathol 34:169–175
to evaluate the effectiveness of measures taken to Palusci VJ, Fischer H (2012) Child abuse and neglect  –
a diagnostic guide. Manson Publishing, Corringham
reduce the numbers or even to prevent child abuse. Road, London
Rashid F, Barron I. (2019) Why the focus of clerical chld
sexual abuse has largely remained on the catholic
References church amongst other non-catholic christian denomi-
nations and religions. J Child Sex Abuse 28: 564–585
Singh MM et  al (2014) An epidemiologic overview of
Abeln B, Love R (2018) An overview of Munchausen
child sexual abuse. J Family Med Prim Care 3:430–435
Syndrome and Munchausen Syndrome by proxy. Nurs
Stoltenborgh M et al (2011) A global perspective on child
Clin North Am 53:375–384
sexual abuse: meta analysis of prevalence around the
Adams KA (2016) The psychohistory of child maltreat-
world. Child Maltreat 16:79–101
ment among antebellum slaveholders. J Psychohist
Thearle MJ, Gregory H (1988) Child abuse in nineteenth
44:2–23
century Queensland. Child Abuse Negl 12:91–101
Adams KA (2017) The psychohistory of child maltreat-
U.S.  Department of Health and Human Services (2018)
ment among antebellum slaveholders. J Psychohist
Child Maltreatment 2018. Administration for
44:178–199
Children, Yought and Families. 330 C Street, SW,
Adams JA et al (2016) Updated guidelines for the medical
Washington, DC 20201
assessment and care of children who may have been
UNICEF Report (2014) Hidden in Plain Sight: a statistical
sexually abused. J Pediatr Adolesc Gynecol 29:81–87
analysis of violence against children
Adams JA et al (2018) Interpretation of medical findings
Vrolijk-Bosschaart TF et  al (2017) Physical symptoms
in suspected child sexual abuse: an update For 2018. J
in very young children assessed for sexual abuse: a
Pediatr Adolesc Gynecol 31:225–231
mixed method analysis from the ASAC study. Eur J
Chong CF et al (2018) Predictors of long-term neurologi-
Pediatr 176:1365–1374
cal outcome in non-accidental head injury. Eye (Lond)
Vrolijk-Bosscharart TF et al (2018) Clinical practice: rec-
32:608–614
ognizing child sexual abuse – what makes it difficult?
Cohen MC (2019) The Swedish shaken baby syndrome
Eur J Pediatr 177:1343–1350
report and review: what does the latest knowledge tell
Witt A et al (2017) Child maltreatment in Germany: prev-
us? Acta Paediatrica Editorial 108:198–199
alence rates in the general population. Child Adolesc
deMause L (1998) The History of child abuse. J Psychohist
Psychiaty Ment Health 11:47–56
25:216–236
Wood JN et  al (2012) Local macroeconomic trends and
deMause L (2009) Child abuse, homicide and raids in
hospital admissions for child abuse, 2000–2009.
tribes. J Psychohist 36:192–211
Pediatrics 130:e358–e364
Guthkelch AN (1971) Infantile subdural haematoma
Yates G, Bass C (2017) The perpetrators of medical child
and its relationship to whiplash injuries. Br Med J
abuse (Munchausen Syndrome by proxy) – a system-
2:430–431
atic review of 796 cases. Child Abuse Negl 72:45–53
 Part III
Head and Neck
 Pierre Robin Sequence
Udo Rolle, Aranka Ifert, and Robert Sader
27.1 Introduction and Historical
Overview
The French dental surgeon Pierre Robin described
is a triad of micrognathia, glossoptosis, and the
upper airway obstruction, and this was subse-
quently named the Pierre Robin sequence (PRS).
The majority of patients with PRS also have a
cleft of the secondary palate. A U-shaped cleft
palate is a typical feature that has been reported
in 73–90% of patients (Cote et  al. 2015). The
ptotic tongue with a retropositioned base against
the posterior pharyngeal wall could result in the
upper airway collapse during inspiration, which
worsens in the supine position in newborns with
PRS (Khansa et  al. 2017). This severe airway
obstruction might result in feeding difficulties,
U. Rolle (*)
Department of Pediatric Surgery, University Hospital
Frankfurt, Frankfurt/Main, Germany
Department of Pediatric Surgery, Goethe University
Frankfurt, Frankfurt, Germany
e-mail: udo.rolle@kgu.de
A. Ifert
Carolinum, Institute of Dentistry, Frankfurt, Germany
R. Sader
Department of Oral, Maxillofacial, and Plastic Facial
Surgery, Goethe University Frankfurt,
Frankfurt, Germany
e-mail: Robert.sader@kgu.de
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_27
27
gastroesophageal reflux, and failure to thrive
(Cladis et al. 2014). Remarkedly, newborns with
or without PRS may have similar birth weights,
but the growth of PRS infants’ lags behind unaf-
fected infants, even when the upper airway has
been addressed in early infancy (Li and Scott
2020).
27.2 Incidence
The incidence of Pierre Robin Sequence is vari-
able; the highest incidence is in the USA with 1
case per 3120 live births. European countries
have reported incidence rates of 1  in 8060 live
births in Germany, 1  in 8500 live births in the
UK, and 1 in 14,000 live births in Denmark.
Approximately half of the cases with PRS are
syndromic, and the remaining cases occur as an
isolated anomaly. Stickler syndrome, Velocardio-
facial syndrome, and Treacher-Collins syndrome
are the three syndromes most commonly associ-
ated with PRS (Cote et al. 2015).
PRS is associated with mortality rates ranging
from 3.6% to 21%, with the highest mortality
rates among patients with associated anomalies
or syndromic cases (Logjes et al. 2018). Recently,
the overall mortality rate was reported to be
16.6% in isolated cases. PRS patients with car-
diac or central nervous system anomalies had
higher mortality, with rates estimated at 39% and
33%, respectively (Costa et al. 2014).
349
350
27.3 Etiopathogenesis
The cause and pathophysiology of PRS remain
unclear (Cote et al. 2015). In the last decades, the
inciting factor in Pierre Robin sequence was con-
sidered to be hypoplasia of the mandible, which
is caused either from a primary growth distur-
bance or from hyperflexion of the neck. In the
last years, more and more a functional-based
thinking took place according to the rule “form
follows function” and that the hypoplasia of the
mandible was not the reason for PRS but the
­consequence of a functional disturbance of the
tongue (Sader 2009).
The development of PRS takes probably place
before 9 weeks in utero. The three major hypoth-
eses to explain the sequence of events in the
Pierre Robin sequence based on the mandibular
growth (Tan et al. 2013) are as follows: (a) hypo-
plastic mandible; (b) oropharyngeal and muscu-
lar deficiencies; and (c) compression of the
mandible in utero. The most widely accepted
theory is the hypoplastic mandible theory.
Meckel’s cartilage is thought to present as the
primary defect. This embryonic structure is
involved in the formation and growth of the man-
dible. The mandible develops hypoplastic, which
subsequently leads to a small mouth volume.
Furthermore, an abnormal position of the tongue
and secondary impairment of palatal closure
occur (Hanson and Smith 1975; Cote et al. 2015).
The second theory, the oropharyngeal and mus-
cular deficiency hypothesis, is based on hypoto-
nia of the oropharyngeal muscles, which could
result in hypoplasia of the mandible. Third, the
mandible compression theory plays a role in a
small proportion of neonates with Pierre Robin
sequence.
In opposite to these theories published by Tan
et al. (2013), it becomes more and more evident
that the problem of PRS may be primarily based
on the malpositioning of the tongue and not being
a growing problem of the mandible (Buchenau
et al. 2017). It is known that the tongue muscle
and its movements are one of the most important
factors for stimulating the growth of the mandi-
ble. It was seen in general that in cleft palate
U. Rolle et al.
babies the tongue is usually repositioned being
the reason for the general coincidence of a cleft
palate and a slight mandibular retrognathia. This
retroposition also has a known strong functional
effect on the later speech production. By this ret-
ropositioning of the tongue, consequently cleft
palate babies often have a retropositioning of the
articulation zones which causes the known artic-
ulation problem to be treated by logopaedic
speech therapy. This functional problem of the
tongue is increased in babies with PRS, maybe
by a more wide cleft of the soft palate. The func-
tional theory of the development of PRS now
states that the cleft of the soft palate causes the
retropositioning of the tongue creating as a sec-
ondary consequence the hypoplasia of the man-
dible (Müller-Hagedorn et al. 2017). As a clinical
proof for this theory a sufficient treatment con-
cept of PRS was developed using a modified
palatal plate influencing the position of the
tongue and eliminating the functional airway
problems in PRS (Sader 2009, Buchenau et  al.
2017).
27.4 Genetics
Twins reveal a high proportion of
PRS.  Additionally, family members of PRS
patients show a higher incidence of cleft lip and
palate (Gangopadhyay et  al. 2012). A critical
chondrogenic regulator, the SOX9 gene, has been
linked to non-syndromic RPS in families with
more than one affected member (Cote et al. 2015).
A possible genetic component in the aetiology of
PRS seems probable due to the relationship of
PRS with many different other syndromes. The
most common syndromic PRS cases are associ-
ated with different genes; for example, Stickler
syndrome is associated with mutations in COL
genes (Acke et  al. 2012). Velocardiofacial syn-
drome arises from a microdeletion of chromo-
some 22q11.2 (Buchanan et  al. 2014), and
Treacher-Collins syndrome is associated with
mutations in the TCOF1, PLOR1C, and POLR1D
genes (Buchanan et  al. 2014; Trainor and
Andrews. 2013; Kadakia et al. 2014).
27  Pierre Robin Sequence
27.5 Pathology
The three essential components of Pierre Robin
sequence are as follows:
1. Micrognathia or retrognathia.
2. Glossoptosis, possibly accompanied by air-
way obstruction.
3. Cleft palate (usually U shaped, but a V shape
is also possible).
It is notable that cleft palate does not have a
complete penetrance and is only seen in approxi-
mately 80% of PRS patients (Sadewitz 1992).
27.6 Diagnosis
The main reason for making the diagnosis in
patients with PRS is clinical. Affected newborns
present with typical signs comprising micro- or
retrognathia, glossoptosis, and cleft palate. The
initial clinical symptoms could be variable.
Some infants with PRS present with severe
airway obstruction such as stridor, retractions,
and even cyanosis at birth, whereas others only
exhibit minimal respiratory symptoms. Naturally,
airway obstruction in PRS requires early and
proper management, since it may lead to hypoxia,
heart failure, failure to thrive, and cerebral
impairment. Usually, syndromic PRS cases are
more severe and carry a worse prognosis than
non-syndromic cases. In contrast, it is expected
that patients with non-syndromic Pierre Robin
sequence will show catch-up growth of the
mandible.
Airway obstruction and cleft palate will lead
to feeding difficulties in patients with PRS. Cleft
palate is thought to inhibit the formation of ade-
quate intraoral pressure required for the extrac-
tion of milk from the breast or the bottle (Cote
et al. 2015).
27.7 Differential Diagnosis
Non-syndromic Pierre Robin Sequence has no 27.8.1.1 Nasopharyngeal Tube
relevant differential diagnosis. Stickler syn- An endotracheal tube could be converted into a
drome, Velocardiofacial syndrome, and Treacher-­ nasopharyngeal tube by cutting it to the appropri-
351
Collins syndrome are the three syndromes most
commonly associated with PRS (Cote et  al.
2015).
27.8 Management
27.8.1 Airway Management
Airway obstruction in PRS is related to glossop-
tosis and occurs at birth or immediately after and
might take up to 3  weeks to become apparent
(Ogborn and Pemberton 1985). Most neonates
present with isolated PRS that is not combined
with any of the syndromes. The latter patients
will typically present with significantly more
clinical problems, such as airway and feeding
difficulties, as mentioned before. Airway obstruc-
tion in PRS seems to be related to the narrowing
or complete obstruction of the pharyngeal space
by the posteriorly displaced tongue. This airway
obstruction could occur intermittently. Compli-
cations and unfavourable outcomes of PRS are
directly related to delayed or inappropriate air-
way management (Myer et al. 1998). Therefore,
special attention is required in PRS patients.
Increased respiratory effort, stridor, subcostal
retractions, and cyanotic or apneic spells are typ-
ical clinical signs of the upper airway obstruc-
tion. If a child is otherwise asymptomatic,
choking attacks, cyanosis during feeding, or
repeated aspiration events may be related to
intermittent airway problems.
PRS infants may also have a short or collaps-
ing epiglottis, laryngomalacia, and/or segmental
tracheal stenosis, and endoscopy is advised to
assess the child. The persistence of airway diffi-
culties requires further intervention (Albino et al.
2016; Kam et al. 2015).
Every child with symptoms of airway obstruc-
tion should be nursed prone with the head to one
side. The head should be maintained in a level
position to prevent either glossoptosis or gastro-
esophageal reflux (GER). This position allows
successful oral feeding in affected children.
352
ate length, inserting it via the nasal route and
securely strapping it in place. This provides a
secure nasopharyngeal airway, which bypasses
the oral pharynx and the obstruction due to
glossoptosis. The nasopharyngeal airway is a
­ formed in newborns to prevent tracheostomy or at
very effective, temporary form of airway man-
agement within the intensive care unit (ICU).
Patients with nasopharyngeal tubes inserted
should not be sent home, as dislodgement of the
tube can result in an acute airway obstruction.
27.8.1.2 Endotracheal Tube
Endotracheal intubation serves as short-term sup-
port if opening of the nasopharyngeal airway is
not successful or during resuscitation or
anaesthesia.
27.8.1.3 Tongue–Lip Adhesion/
Glossopexy
In this technique, the tongue is essentially sutured
to the lower lip. After the child has demonstrated
catch-up growth, the tongue–lip adhesion can be
released (Broucqsault et  al. 2018). The efficacy
of the tongue–lip adhesion technique remains a
controversial issue.
Glossopexy consists of suturing the tongue
base to the mandible. Permanent glossopexy is
difficult to achieve due to the relatively soft con-
sistency of the mandible. Therefore, this tech-
nique is also controversial.
27.8.1.4 Tracheostomy
Tracheostomy is not the treatment of choice and
should be avoided if possible. Tracheostomy is
indicated if all other techniques fail. Tracheostomy
should be performed by an appropriately skilled
surgeon who is familiar with infantile airways.
Tracheostomy requires close monitoring but
enables oral feeding. Tracheostomy could be
removed after the child’s airway obstruction has
resolved, which usually occurs within the first
year of life.
27.8.1.5 Distraction Osteogenesis
of the Mandible
Distraction osteogenesis is a new and promising
technique. The mandible is near the angle of the
mandible on both sides. A specialized mechanical
U. Rolle et al.
device distracts these two portions by approxi-
mately 1.5–2 mm every day. Using this technique,
the mandible is gradually elongated over a period
of 2–3 weeks. Mandibular distraction can be per-
a later stage to remove a tracheostomy tube.
Distraction osteogenesis has been introduced
only during the last 5–10 years. Therefore, long-­
term follow-up results of this promising tech-
nique are not available. Nevertheless, the
distraction osteogenesis technique should be
reserved for severe cases of non-syndromic and
syndromic PRS, since physiologic catch-up
growth of the mandible occurs in most cases of
non-syndromic PRS.
Recently, the first long-term outcomes have
been reported. Patients undergoing external man-
dibular distraction show an improvement in sleep
outcomes but still require polysomnographic
follow-up (Ehsan et al. 2020).
27.8.1.6 Tongue Positioning
and Stimulation Plate
During the last decade, a new, nonsurgical tech-
nique was developed by orthodontists that guar-
antees, in most cases, a free airway space and
treats the hypoplastic mandible causally. First, a
palatal plate is produced; this is similar to the
feeding plate for cleft palate newborns but with a
dorsal spur that goes directly to the epiglottis
(Fig.  27.1a–c). In individual cases, endoscopic
control is necessary during positioning to avoid
irritation of the epiglottis. The basic principle is
to position the tongue anteriorly so that the air-
way is kept patent. Moreover, via functional
stimulation of the tongue, the mandible starts to
grow during the following months and is usually
normal when palatal closure is performed at the
age of approximately 6  months (Fig.  27.2a–e).
Feeding is also supported, but problems remain
in some cases (Brosch et al. 2006).
27.8.1.7 Noninvasive Ventilation
There is growing evidence that noninvasive respi-
ratory support (NRS) could improve breathing
patterns and respiratory outcomes for infants
with severe upper airway obstruction due to
PRS. Subsequently, the rate of necessary trache-
27  Pierre Robin Sequence
a b
Fig. 27.1 (a–c) Stimulation plate
ostomies was reduced. Some authors consider
this the first-line treatment (Leboulanger et  al.
2010; Amaddeo et al. 2016).
27.8.2 Nutritional Management
In 38–62% of the cases, children with Pierre
Robin sequence could present with feeding diffi-
culties (Evans et al. 2011). Initial treatment con-
sists of bottle-feeding in a prone position with the
head slightly elevated. This method of feeding is
appropriate in children with catch-up growth of
the mandible.
If this is not satisfactory, gavages or feeding
tubes can be used temporarily to improve nutri-
tion. If feeding is still not successful, the child
might need a gastrostomy, which can be removed
after gaining the ability to be fed orally.
It has been shown clearly that infants with
Pierre Robin sequence require adequate caloric
353
intake. It is important to achieve the maximum
growth rate of the mandible since the resolution
of the airway problems is directly related to
mandibular growth. Only recently has the
increased work of breathing been appreciated as
an important component of calorie consump-
tion. It may be necessary to provide these chil-
dren with several times the normal caloric
requirement of an infant to compensate for up to
a tenfold increase in respiratory work. Indeed,
failure to gain weight despite maximum nutri-
tional intake should suggest the need for more
aggressive airway management. The availability
of total parenteral nutrition should prevent any
instances of failure to thrive, but it is rarely
needed if other aspects of the condition are
managed correctly.
It has been additionally proven that PRS
infants have a higher incidence of GER, and even
empiric reflux treatment may be indicated to
improve breathing and feeding.
354
a b
d e
Fig. 27.2 (a) Pierre Robin sequence patient before the
insertion of a stimulation plate. (b) PRS patient with stim-
ulation plate. (c) PRS patient after 2 months of treatment
27.8.3 Management of Cleft Palate
Cleft palate is present in at least 80% of patients
with Pierre Robin sequence. Cleft palates are
typically repaired while patients are infants. A
palatal plate can be used in patients with a cleft of
the hard palate to improve feeding. The plate also
corrects the tongue position by moving it
­anteriorly. In patients with a cleft of the soft pal-
ate alone, a palatal plate has no positive effect on
feeding, but it can improve the tongue position
and stimulate mandibular growth. To enhance
this effect, Castillo Morales has suggested that
the plate can be modified by an anterior stimulus
(Hohoff and Ehmer 1999).
U. Rolle et al.
c
with a stimulation plate. (d) PRS patient after 4 months of
treatment with a stimulation plate. (e) PRS patient after
closure of cleft palate at the age of 8 months
Surgical protocols differ from centre to centre,
and cleft closure is performed not only by differ-
ent techniques (i.e., Langenbeck, Furlow,
Wardill) but also at different ages, ranging from 4
to 36 months.
It is currently assumed that early surgery will
provide a better chance of normal palatal func-
tion and speech development.
27.8.4 Management
of Micrognathia/Retrognathia
The first described functional therapy for micro-
gnathia was the use of the orthodontic palatal
27  Pierre Robin Sequence
plate to achieve growth stimulation of the man-
dible. It was not clear until recently whether the
growth potential of the mandible after this stimu-
lation is sufficient to achieve normal dimensions.
However, it has been shown, based on physical
examinations until age 5, that the mandible can
barely regain its growth in relation to a normal
population (Daskalogiannakis et al. 2001). A ret-
rospective longitudinal study that evaluated
cephalograms and lateral photographs of
American patients with Pierre Robin sequence
and cleft of the soft palate showed that the
­mandible achieved only partial catch-up growth
and, in adults, a smaller maxilla, mandible, and a
narrow respiratory airway space persisted
(Figueroa et al. 1991). Studies in the Finnish pop-
ulation showed the same result (Laitinen et  al.
1997). Increased mandibular growth was seen
during the first 2 years of life, but normal cranio-
facial dimensions were never achieved. At the
young adult stage, even if the patient’s profile
appeared less retrognathic due to masking by the
overlying soft tissues or the patient’s teeth
showed neutral occlusion, the cephalograms
revealed retrognathia and caudal-dorsal rotation
of the mandible. Thus, it seems, in accordance
with today’s knowledge, that the microgenia in
Pierre Robin sequence can be balanced only par-
tially by growth processes. Frequently, orthodon-
tic therapy is necessary in childhood. In severe
cases, surgical advancement of the mandible
combined with a genioplasty can also be
beneficial.
27.8.5 Management of Further
Associated Malformations
27.8.5.1 Skeletal Anomalies
Approximately 11–21% of children with Pierre
Robin sequence have limb defects (Williams
et  al. 1981). Common anomalies are talipes
equinovarus, syndactyly, short or absent digits,
and hypoplastic long bones. Occipito-atlanto-­
axial instability has also been described, empha-
sizing the need for very experienced clinicians to
undertake the intubation of such patients.
Orthopaedic and radiological consultation should
355
be sought in children with suspected skeletal
problems. Rare neuromuscular defects can also
occur, resulting in a tendency for glossoptosis to
persist despite mandibular growth (Carey et  al.
1982).
27.8.5.2 Ear Problems
Malformations of the ear have a frequency of
10.5% and consist of defects in the auditory
capacity and anomalies of the shape of the ear.
One main concern is the frequently recurring
infections of the middle ear, which also occur in
patients with a cleft palate, and are based on dis-
turbed function of the Eustachian tubes.
Therefore, hearing screening must be performed
at birth. At a later date, control of the middle ear
tube function, if necessary, must be achieved via
the placement of grommets (Handžić et al. 1996).
27.8.5.3 Cardiovascular Anomalies
Intrinsic cardiac defects are found in up to 20%
of infants with Pierre Robin sequence (Pearl
1982). Septal defects are common, but more
complex lesions can also occur. A thorough car-
diovascular examination should be performed in
PRS babies, particularly since airway difficulties
may aggravate the cardiac status (Dykes et  al.
1985).
27.8.5.4 Ocular Anomalies
Retinal detachment and micrognathia occur as
part of Stickler syndrome (Opitz et al. 1972), but
10% of infants with non-syndromic PRS also
have eye defects, such as strabismus, ptosis, and
microphthalmia. More severe defects, such as
cataracts and congenital glaucoma, have also
been reported, and ophthalmologic consultation
is recommended in all cases (Smith and Stowe
1961).
27.8.6 Nasal Obstruction
Choanal atresia is a rare accompaniment of Pierre
Robin sequence (Borovik and Kveton 1987), but
it may complicate respiratory difficulties in small
infants who do not breathe through their mouths.
It is important to ensure nasal patency, especially
356
if one nostril is to be utilized for a nasogastric
feeding tube. Choanal obstruction by itself can
lead to glossoptosis, with consequences identical
to those of Pierre Robin sequence (Cozzi and
Pierro 1985).
27.9 Conclusion
In isolated Pierre Robin sequence, the long-term
outcome is directly related to the quality of the
management at the onset of symptoms. With
adequate nutrition, mandibular growth will
­ Amaddeo A, Abadie V, Chalouhi C et  al (2016)
achieve normal or near-normal proportions, and
glossoptosis will resolve.
The previously documented high incidence of
mental retardation in PRS patients is almost cer-
tainly due to unrecognized episodes of hypoxia,
and with good airway management, this compli-
cation is uncommon.
Undiagnosed hypoxia may also lead to pul-
monary vasoconstriction, with resultant pulmo-
nary hypertension and cor pulmonale. Some
instances of sudden death in patients with Pierre
Robin sequence were likely due to this problem.
The presence of cardiomegaly on a chest X-ray
should alert the physician to the possibility that
hypoxic episodes have been overlooked, and
appropriate steps should be taken immediately.
Although airway patency improves with
growth, there remains a potential for obstruction,
particularly after invasive procedures such as
intubation or cleft palate repair (Hatch 1989). In
some children, obstruction may occur during
sleep, causing occasional apnoea with potentially
hazardous consequences (Frohberg and Lange
1993). A degree of mandibular hypoplasia may
persist for several years, resulting in malocclu-
sion and the need for dental treatment (Sheffield
et al. 1987).
The overall mortality rate in infants with
Pierre Robin sequence is approximately 25%.
The majority of deaths (70%) occur in children
with associated anomalies, particularly those
with cardiac defects or an underlying syndrome.
These facts must be considered when counselling
parents of affected children. With good medical
U. Rolle et al.
and nursing care, the prognosis for children with
isolated Pierre Robin sequence should be excel-
lent (Bull et al. 1990).
References
Acke FR, Dhooge IJ, Malfait F et  al (2012) Hearing
impairment in Stickler syndrome: a systematic review.
Orphanet J Rare Dis 7:84
Albino FP, Wood BC, Han KD, Yi S, Seruya M, Rogers
GF, Oh AK (2016) Clinical factors associated with the
non-operative airway management of patients with
Robin sequence. Arch Plast Surg 43(6):506–511
Continuous positive airway pressure for upper airway
obstruction in infants with Pierre Robin sequence.
Plast Reconstr Surg 137(2):609–612
Borovik HR, Kveton JF (1987) Pierre Robin syndrome
combined with unilateral choanal atresia. Otolaryngol
Head Neck Surg 96:67–70
Brosch S, Flaig S, Bacher M, Michels L, de Maddalena
H, Reinert S, Mauz P (2006) The influence of the
Tübingen soft palate plate and early cleft closure on
swallowing and Eustachian tube function in children
with Pierre Robin sequence. HNO 54:756–760
Broucqsault H, Lamblin MD, Hosana G, Fayoux P (2018)
Evaluation of the efficacy of tongue-lip adhesion in
Pierre-Robin-Sequence. Eur Ann Otorhinolaryngol
Head Neck Diseases 135:155–158
Buchanan EP, Xue AS, Hollier LH Jr (2014) Craniofacial
syndromes. Plast Reconstr Surg 134(1):128e–153e
Buchenau W, Wenzel S, Bacher M, Müller-Hagedorn S,
Arand J, Poets CF (2017). Functional treatment of
airway obstruction and feeding problems in infants
with Robin sequence Arch Dis Child Fetal Neonatal
102(2):142–146. https://doi.org/10.1136/archdischild-­
2016-­311407. Epub 2016 Jul 19
Bull MJ, Givan DC, Sadove AM, Bixler D, Hearn D
(1990) Improved outcome in Pierre Robin sequence:
effect of multidisciplinary management. Pediatrics
86:294–301
Carey JC, Fineman RM, Ziter FA (1982) The Robin
sequence as a consequence of malformation, dys-
plasia, and neuromuscular syndromes. J Pediatr
101:858–864
Cladis F, Kumar A, Grunwaldt L et  al (2014) Pierre
Robin sequence: a perioperative review. Anesth Analg
119(2):400–412
Costa MA, Tu MM, Murage KP et  al (2014) Robin
sequence: mortality, causes of death, and clinical out-
comes. Plast Reconstr Surg 134(4):738–745
Cote A, Fanous A, Almajed A et al (2015) Pierre Robin
sequence: review of diagnostic and treatment chal-
lenges. Int J Pediatr Otorhinolaryngol 79(4):451–464
Cozzi F, Pierro A (1985) Glossoptosis-apnoea syndrome
in infancy. Pediatrics 75:836–843
27  Pierre Robin Sequence
Daskalogiannakis J, Ross RB, Tompson BD (2001) The
mandibular catch-up growth controversy in Pierre
Robin sequence. Am J Orthod Dentofac Orthop
120:280–285
Dykes EH, Raine PAM, Arthur DS, Drainer IK, Young
DG (1985) Pierre Robin syndrome and pulmonary
hypertension. J Pediatr Surg 20:49–52
Ehsan Z, Weaver KN, Pan BS, Huang G, Hossain MM,
Simakajornboon N (2020) Sleep outcomes in neo-
nates with Pierre Robin sequence undergoing exter-
nal mandibular distraction: a longitudinal analysis.
Plast Reconstr Surg; Aug 19. https://doi.org/10.1097/
PRS.000000000007289
Evans KN, Sie KC, Hopper RA, Glass RP, Hing AV,
Cunningham ML, Robin sequence. (2011) From diag-
nosis to development of an effective management
plan. Pediatrics 127:936–948
Figueroa AA, Glupker TJ, Fitz MG, BeGole EA (1991)
Mandible, tongue, and airway in Pierre Robin
sequence: a longitudinal cephalometric study. Cleft
Palate-Craniofac J 28:425–434
Frohberg U, Lange RT (1993) Surgical treatment of Robin
sequence and sleep apnea syndrome; case report
and review of the literature. J Oral Maxillofac Surg
51:1274–1277
Gangopadhyay N, Mendoonca DA, Woo AS (2012) Pierre
Robin sequence. Sem Plast Surg 26:76–82
Handžić J, Ćuk V, Rišavi R, Katić V, Katušić D, Bagatin
M, Štajner-Katušić S, Gortan D (1996) Pierre Robin
syndrome: characteristics of hearing loss, effect of age
on hearing level and possibilities in therapy planning.
J Laryngol Otol 110:830–835
Hanson JW, Smith DW (1975) U-shaped palatal defect in
the Robin anomalad: developmental and clinical rel-
evance. J Pediatr 87:30–33
Hatch DJ (1989) Anaesthesia for paediatric surgery.
In: Summer E, Hatch DJ (eds) Textbook of paediat-
ric anaesthesia practice. Baillière Tindall, London,
pp 275–304
Hohoff A, Ehmer U (1999) Short-term and long-term
results after early treatment with the Castillo Morales
stimulating plate  – a longitudinal study. J Orofac
Othop 60(1):2–12
Kadakia S, Helman SN, Badhey AK et al (2014) Treacher
Collins syndrome: the genetics of a craniofacial dis-
ease. Int J Pediatr Otorhinolaryngol 78(6):893–898
Kam K, McKay M, MacLean J, Witmans M, Spier S,
Mitchell I (2015) Surgical versus nonsurgical inter-
ventions to relieve upper airway obstruction in
children with Pierre Robin sequence. Can Respir J
22(3):171–175
Khansa I, Hall C, Madhoun LL et al (2017) Airway and
feeding outcomes of mandibular distraction, tongue-­
lip adhesion, and conservative management in Pierre
Robin sequence: a prospective study. Plast Reconstr
Surg 139(4):975e–983e
357
Laitinen SH, Heliövaara A, Ranta RE (1997) Craniofacial
morphology in young adults with the Pierre Robin
sequence and isolated cleft palate. Acta Odontol
Scand 55:223–228
Leboulanger N, Picard A, Soupre V, Aubertin G,
Denoyelle F, Galliani E, Roger G, Garabedian EN,
Fauroux B (2010) Physiologic and clinical benefits of
noninvasive ventilation in infants with Pierre Robin
sequence. Pediatrics 126:e1056–e1063
Li L, Scott AR (2020) Weight gain in infants with Pierre
Robin Sequence in the first year of life. Otolaryngol
Head Neck Surg 2020, May 19; https://doi.
org/10.1177/0194599820925454
Logjes RJH, Haasnoot M, Lemmers PMA, Nicolaije
MFA, van den Boogaard MJH, van der Molen ABM,
Breugem CC (2018) Mortality in Robin sequence:
identification of risk factors. Eur J Pediatr 177:781–789
Müller-Hagedorn S, Buchenau W, Arand J, Bacher
M, Poets CF (2017). Treatment of infants with
Syndromic Robin sequence with modified pala-
tal plates: a minimally invasive treatment option.
Head Face Med 30;13(1):4. https://doi.org/10.1186/
s13005-­017-­0137-­1
Myer CM, Reed JM, Cotton RT, Willging JP, Shott SR
(1998) Airway management in Pierre Robin sequence.
Otolaryngol Head Neck Surg 118:630–635
Ogborn MR, Pemberton PJ (1985) Late development
of airway obstruction in the Robin anomalad (Pierre
Robin syndrome) in the newborn. Aust Paediatr J
21:199–200
Opitz JM, France T, Herrman J, Spranger JW (1972) The
stickler syndrome. N Engl J Med 286:546–547
Pearl W (1982) Congenital heart disease in the Pierre
Robin syndrome. Pediatr Cardiol 2:307–309
Sader R (2009) Lippen-Kiefer-Gaumenspalten. Pädiatrie
up2date 2:183–205. https://doi.org/10.1055/
s−0029−1214683
Sadewitz VL (1992) Robin sequence: changes in think-
ing leading to changes in patient care. Cleft Palate
Craniofac J 29(3):236–253
Sheffield LJ, Reiss JA, Strohm K, Gilding M (1987) A
genetic follow-up study of 64 patients with the Pierre
Robin complex. Am J Med Genet 28:25–36
Smith JL, Stowe FR (1961) The Pierre Robin syndrome
(glossoptosis, micrognathia, cleft palate). Pediatrics
27:128–133
Tan TY, Kilpatrick N, Parlie PG (2013) Developmental
and genetic perspectives on Pierre Robin sequence. Am
J Med Genet C Semin Med Genet 163C(4):295–305
Trainor PA, Andrews BT (2013) Facial dysostoses: etiol-
ogy, pathogenesis and management. Am J Med Genet
C Semin Med Genet 163C(4):283–294
Williams AJ, Williams MA, Walker CA, Bush PG (1981)
The Robin anomalad (Pierre Robin syndrome) – a fol-
low up study. Arch Dis Child 45:663–668
 Choanal Atresia
28
R. Ben Speaker, Michael Harney, and John Russell

28.1 Introduction

Infants are obligate nasal breathers for the first

4–6  weeks of life. Bilateral choanal atresia is,
therefore, an airway emergency. Failure to recog-
nize and promptly treat this condition may result
in death by asphyxia. The condition was first
described by Johann Roederer in 1755 and Emmert
reported the first successful treatment of the condi-
tion in 1854, using a curved urethral sound to
puncture atretic choanae in a child. The posterior
choanae connect the nasal cavities to the nasophar-
ynx. An image of the posterior choanae can be
seen in Fig.  28.1, which shows the view from a
120° endoscope viewed through the mouth.
Choanal atresia is an uncommon condition affect-
ing 8 in 100,000 births. The rate of unilateral-to-
bilateral atresia is 2:1. The atretic plate can be Fig. 28.1 Left choanal atresia viewed through 120°
endoscope following punctured by the small urethral
bony (30%), or bony and membranous (70%). The sound
bony narrowing can arise from any combination of
the sphenoid superiorly, pterygoid plates laterally,
the vomer medially, or hard palate inferiorly.
28.2 Etiopathogenesis

The precise reason for the development of the

R. B. Speaker
Royal College of Surgeons Ireland, Dublin, Ireland
M. Harney
Bon Secours Hospital, Cork, Ireland
J. Russell (*)
Department of Pediatric Otorhinolaryngology,
Children’s Health Ireland, Crumlin, Dublin, Ireland
e-mail: john@drjrussellent.com
atretic plate is not known and may arise from sev-
eral etiologies. It may represent persistence of the
buccopharyngeal membrane, persistence of the
nasobuccal membrane, persistent mesoderm in
the nasal cavity creating adhesions, or a defect in
neural crest migration.

© Springer Nature Switzerland AG 2023 359

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_28
360 R. B. Speaker et al.

28.3 Pathophysiology plinary to preoperative work-up is essential in

these patients, with the involvement of the cardi-
Choanal atresia is associated with additional ologists, ophthalmologists, geneticists, nephrolo-
anomalies in 50% of cases. It can be associated gists, neonatologists, and audiological scientists.
with CHARGE syndrome (25% cases), Treacher-­ Treacher-Collins is associated with retrognathia
Collins syndrome, Aperts, Di George sequence, which may have additional implications for the
VATER syndrome, or trisomy 18. CHARGE syn- newborn airway. Ten-to-fifteen percent of chil-
drome is a nonrandom association of Coloboma dren with Treacher-Collins syndrome have cho-
(of iris, choroid, and/or microphthalmia), Heart anal atresia.
defects (septal defects and/or conotruncal lesion),
choanal Atresia, mental or growth Retardation,
Genito-urinary abnormalities (cryptorchidism, 28.4 Pathology
microphallus, and/or hydronephrosis), and Ear
anomalies (hearing loss, external, middle, or Obstruction can be unilateral or bilateral.
inner ear may be involved). In addition to these Bilateral obstructions are present in the early
cardinal features, additional minor characteristics neonatal period. Infants are obligate nasal breath-
include characteristic face and hand dysmorphol- ers, as the tongue is in opposition to the palate.
ogy, hypotonia, orofacial clefting, dysphagia, and Attempts at inspiration result in sternal retraction
tracheoesophageal abnormalities. Traditionally, a and intercostal recession. A failure to adequately
diagnosis of CHARGE syndrome requires the ventilate results in cyanosis. If the child starts to
presence of at least three cardinal characteristics cry, the mouth opens and the apnea/cyanosis set-
or two cardinal with three or more minor charac- tles. Once the crying ceases, the obstructed
teristics; however, CHARGE syndrome may breathing pattern recommences. This cyclical
have quite a varied presentation and several diag- cyanosis is typical of choanal atresia. This pattern
nostic criteria exist. The presence of bilateral may only become conspicuous during feeding.
choanal atresia indicates a poor prognosis in Occasionally, the infant may present with failure
infants diagnosed with CHARGE. Choanal atre- to thrive secondary to an inability to maintain a
sia is present in 57% of infants with satisfactory airway during feeding, yet not pres-
CHARGE. Although CHARGE is inherited in an ent with cyclical cyanosis or choking during
autosomal dominant fashion, the majority of feeding. A unilateral atresia may not be detected
cases are sporadic. Males and females are equally during the neonatal period. It may present later in
affected. Several different genetic abnormalities life with unilateral obstruction and discharge as
have been identified in infants with CHARGE; the mucus that is produced on this side of the
most cases are due to a mutation or deletion of nasal cavity cannot be transported to the postna-
the chromodomain helicase DNA-binding pro- sal space.
tein-­7 (CHD7) gene. Thirty percent of neonates
with this condition will require a tracheostomy
for airway difficulties, in addition to choanal 28.5 Diagnosis
atresia. They may recurrently aspirate which can
be from a number of different causes; glossopha- If clinical suspicion of choanal atresia arises, the
ryngeal or vagus nerve paresthesias, causing an diagnosis is further suspected if the child fails to
insensate larynx or a vocal cord palsy. They may mist up a stainless steel spatula or mirror that is
have laryngomalacia or suffer from a laryngeal held under the nose. Failure to pass a size 8 cath-
cleft. They may have esophageal atresia or a tra- eter 5.5  cm from the anterior choana is highly
cheoesophageal fistula. Correspondingly, the suggestive of atresia. Nasal airflow can usually
mortality rate in this group is higher; children be detected with the use of a stethoscope. The
with choanal atresia with CHARGE association atretic plate can be visualized directly by nasal
have a 20% perioperative mortality. A multidisci- endoscopy. A computed tomography (CT) scan
28  Choanal Atresia
should be performed to confirm the atresia, to
rule out other diagnoses, to determine the propor-
tion of membranous to bony narrowing, and to
determine the main sites of bony narrowing. The
nose should be suctioned and decongested prior
to performing the CT to avoid the presence of
additional soft-tissue shadowing.
28.6 Differential Diagnosis
The differential diagnosis represents all other
causes of bilateral nasal blockage in the neonate,
including piriform aperture stenosis, encephalo-
cele, dermoid cysts, septal abnormalities, benign
nasal tumors, nasolacrimal duct cysts, and muco-
sal disease with turbinate hypertrophy.
28.7 Emergency Treatment
Prompt recognition of respiratory distress is
essential. Airway obstruction can be relieved by
the insertion of an oropharyngeal airway, which
may be secured in place with tape across the
mouth. The oropharyngeal airway length can be
estimated by approximating the distance from the
angle of the mandible to the center of the two
incisors. This airway can be left in for several
weeks, if necessary. A McGovern nipple is an
alternative option. This is a nipple that has
enlarged perforations at the tip, which allows the
child to breathe. Endotracheal intubation pro-
vides a safe and secure airway, though this will
require an intensive-care unit bed until definitive
repair can be performed. Intubation may prove
difficult in infants with retrognathia associated
with Treacher-Collins syndrome. An orogastric
tube is required for the child to feed until defini-
tive surgical repair can be attempted.
28.8 Management
The timing of surgery is controversial and should
vary in response to the acuity of the patient’s pre-
sentation. Some authors suggest that the rule of
10s is adhered to; that the child is 10 lb., 10 weeks
361
old, and has 10 g of hemoglobin. In bilateral atre-
sia, repair should be undertaken as soon as it is
safe to do so; this is largely dictated by the pres-
ence or absence of cardiac co-morbidities in the
child. Recent work by the International Pediatric
Working Group (IPOG) suggests deferring sur-
gery for unilateral atresia in the stable infant until
6 months of age. Several different operative tech-
niques have been described. They can be divided
into endoscopic, transnasal, transpalatal, subla-
bial transseptal, and transmaxillary. The most
popular approach is currently the endoscopic
approach. The transmaxillary approach is mainly
of historical interest only and will not be dis-
cussed further here.
28.9 Endoscopic Technique
The author’s technique of choice is the endo-
scopic approach. A tonsil (Boyle–Davis) mouth
gag is inserted. A suture is placed through the
uvula and clipped to provide retraction to the soft
palate. A 120° endoscope is passed into the naso-
pharynx and a view of the obstructed choanae is
obtained on the video monitor (Fig.  28.1). The
nasal cavity is decongested with 1:10,000 adren-
aline patties, and the atretic plate is injected with
1% lignocaine with 1:200,000 adrenaline. The
atretic plate is perforated with a small urethral
sound, using the 120° endoscope which is passed
through the mouth to view the atretic plate at the
time of perforation. The sound is passed to the
medial and inferior aspect of the atretic plate to
avoid inadvertent entry to the skull base or sphe-
noid. Even with thick atretic plates, the sound can
usually be passed without much force being
required. Progressively larger sounds are intro-
duced until the sound can no longer pass through
the anterior choana. The drill is then inserted in
through the nose until it reaches the nasopharynx.
A drill which has a protective sheath over the
shaft is useful to prevent trauma to the nasal cav-
ity. The drilling is performed medially over the
vomer and laterally over the pterygoid plates.
The posterior aspect of the vomer is then removed
by inserting a back-biting forceps into the naso-
pharynx and removing the posterior half of the
362
vomer. This creates a common cavity posteriorly
which minimizes the chance of restenosis.
Although it has largely fallen out of favor, some
centers will apply mitomycin C (0.25  mg/ml)
patties topically to the choanae for 5 min at the
end of the procedure. The area is then cleaned
with saline-soaked patties. Mitomycin C is an
aminoglycoside produced by Streptomyces sp.,
which was originally used as a cytotoxic
­chemotherapeutic agent. It crosslinks DNA lead-
ing to apoptosis. Fibroblasts are particularly sen-
sitive to this action at the doses described above;
theoretically, this reduces scar formation and in
turn the likelihood that a revision will be neces-
sary. While its role in subglottic stenosis and
glaucoma surgery has been clearly demonstrated,
evidence for the use of mitomycin C in choanal
atresia repair is dwindling and it should not be
employed routinely (Kwong 2015). The role of
stenting is controversial. Stents can be associated
with columellar necrosis, nasal vestibule excoria-
tion, blockage, and intranasal synechia. In addi-
tion, stents may act as a foreign body contributing
to restenosis. Stents are avoided at our institution.
If they are employed, they are fashioned from
nasopharyngeal airways which are cut to a length
just longer than the distance from the nares to the
choana. Prior to placement in the nose, three silk
sutures are placed through the external hub of the
NPA and tied. These are then secured to the
patient’s face with tape. Postoperatively, the
stents are kept patent with saline drops.
Betamethasone drops may be used in the nasal
cavities to minimize edema, though benefit from
their use has not been demonstrated. Some
authors advocate the use of antibiotics for the
duration of the stent placement to minimize the
risk of infection, which may increase the risk of
stenosis. The role of antibiotics, however, is also
unclear. If stents are in place, they should be
removed under anesthesia at 6 weeks postop. Any
granulations can be removed with the microde-
brider. Dilatations are often required, with the
largest reported series observing an average of
4.9 procedures being required per patient (Samadi
R. B. Speaker et al.
et al. 2003). We perform all dilatations with the
use of the 120° scope and the monitor. There are
several variations on this technique described;
however, at our institution, the majority of revi-
sions are performed using high-pressure balloon
dilation only. Some authors have described the
use of the drill alone to refashion the posterior
choanae. The sparing of mucosal flaps is also fre-
quently described which are placed over raw sur-
faces at the end of the procedure to improve
epithelialization. However, the author’s experi-
ence is that these flaps are usually not viable by
the end of the procedure. CT guidance systems
are routinely used by certain authors. We feel that
this technology is not necessary in routine cases.
28.10 Transpalatal
The transpalatal technique involves making a
U-shaped incision in the hard palate, 5 mm from
the dental arch, and raising a posteriorly based
subperiosteal flap to gain access to the nasophar-
ynx. The blood supply to the flap is from the
greater palatine arteries which enter posterolater-
ally. Care must be exercised in this area. The
inferior aspect of the vomer is visualized and
removed. The lateral atretic plates are then
removed using a drill or a trucut forceps. The flap
is closed using a two-layered closure. This tech-
nique is useful if the nasopharynx is small and
the skull base low which may be found in chil-
dren with Treacher-Collins syndrome. The bony
narrowing can be enlarged starting from a known
safe area minimizing the chance of a skull base
breach. There are several drawbacks to this tech-
nique, however. The postoperative pain is higher,
possibly requiring a longer hospital stay. Risks of
palatal fistula and reduced growth of the midface
have been described. This is felt to be secondary
to resection of the palatine bone growth plate.
Resulting dental malocclusion has been described
in 50% of patients undergoing this technique. For
these reasons, this technique should be reserved
for children with difficult skull base anatomy.
28  Choanal Atresia
28.11 Sublabial Transseptal
This technique is used by some authors in infants
with abnormal intranasal anatomy or craniofacial
anomalies. A sublabial incision is made through
the perichondrium, which is elevated over the
premaxilla to expose the lower piriform aperture.
The mucosa of the nasal floor and the mucoperi-
chondrium from the septum is elevated. The pos-
terior bony septum and the atretic plates laterally
are then resected.
28.12 Conclusion
Prompt recognition of this condition is essential.
Emergency airway management requires the
insertion of an oropharyngeal airway to stabilize
363
the patient. There are several different techniques
available to reopen the atretic plates. Revision
surgery in the form of repeated dilatations is
often required.
References
Kwong KM (2015) Current updates on choanal atresia.
Front Pediatr 3:52. Published 2015 Jun 9. https://doi.
org/10.3389/fped.2015.00052
Samadi DS, Udayan K, Handker SD (2003) Choanal atre-
sia: a twenty-year review of medical comorbidities
and surgical outcomes. Laryngosocope 113:254–258
 Thyroglossal and Branchial Cysts,
Sinuses, and Fistulas
Michael E. Höllwarth
29.1 Introduction
Remnants of embryological structures in the
neck are derived either from the thyroglossal duct
or from the branchial arches, and constitute the
most common congenital anomaly in this ana-
tomical region, accounting for up to 60% of all
excised neck masses in children.
Thyroglossal duct cysts are slightly more
common when compared with branchial cleft
anomalies (55% vs. 45%) and present as a mid-
line neck mass below the hyoid bone. Sinuses
drain into the foramen cecum of the tongue. Most
commonly, they cause clinical problems in the
first decade of life and more than half of cases are
diagnosed before the age of 5. However, they
may cause problems in older children too.
Anomalies of the branchial arches are the sec-
ond most common congenital lesion of the head
and neck in children. Although branchial cysts
and sinuses are most often operated in child-
hood, they are sometimes detected as a clinical
problem not before adulthood and they may
occur within the family. They are located along
the anterior border of the sternocleidomastoi-
deus muscle; however, they drain into very dif-
ferent regions depending on their origin from
M. E. Höllwarth (*)
University Clinic of Paediatric and Adolescent
Surgery, Medical University of Graz, Graz, Austria
e-mail: michael.hoellwarth@medunigraz.at
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_29
29
one of the branchial clefts. Knowledge of the
embryology and the anatomical variants, as well
as neck anatomy, is crucial to avoid surgical mis-
takes, thereby creating nerve lesions and/or a
significant morbidity.
29.2 Etiology
Thyroglossal Duct: Remnants of the thyroglossal
duct usually present as cysts or a midline mass
below the hyoid bone. The thyroid anlage is a
part of the second branchial arch located on the
midline floor of the pharynx (Sadler 2006). The
endodermal cells descend on the 17th embryonic
day, as a duct from the region of the later foramen
cecum down to the neck, finally forming the two
lateral lobes of the thyroid gland. Primarily, the
duct is located posterior to the first pharyngeal
arch, but passes anterior to the second and third
arch to the thyroid membrane and thyroid carti-
lage anterior, through, or posterior to the hyoid
bone (Fig. 29.1). The duct obliterates in the fifth
gestational week, and the foramen cecum at the
base of the tongue on the one end and the pyrami-
dal lobe of thyroid at the other end are the normal
remnants. However, if duct cells persist, they can
form a cyst or a sinus, and the latter might be con-
nected to the foramen cecum at the base of the
tongue, but rarely appears as a fistula with an
external opening in the middle of the neck, except
after perforation of a putrid cyst. Approximately
365
366 M. E. Höllwarth

thyroid glands. Cysts, sinuses, and fistulas result

in persisting epithelial cells within the mesoderm
and are lined by squamous or columnar epithelia
(Waldhausen 2006). Between 75% and 90% of
the anomalies derive from the second branchial
cleft, while 8–20% arise from the first cleft.

29.3 Pathology

Thyroglossal duct remnants account for more

Fig. 29.1  Schematic drawing of a thyroglossal cyst. The
duct runs through, in front, or behind the foramen cecum
to the base of the tongue
50% of these cysts are close or just inferior to the
border of the hyoid bone. Aberrant thyroid tissue
may be found anywhere along the path of the
descent of the duct.
than half of all congenital anomalies in the neck
region (El-Gohary and Gittes 2011). Three-­
quarters present as cysts and 25% as sinuses, with
or without infection (Foley and Fallat 2006).
Infected cysts form a red and tender tumor in the
middle of the neck (Fig. 29.2). When elements of
the duct persist after the descent of the thyroid,
most of them become clinically apparent before
the age of 20. Rarely, the rest of the duct may
persist within the thyroid at the one end or within
the tongue on the other end (Huang et al. 2015;
El-Korbi et al. 2018). Ectopic thyroid tissue may
be found in 25% of the wall of the cysts. In some

Branchial Cysts and Sinuses  Five branchial

arches form in the fourth and fifth week of devel-
opment by condensation of the mesoderm five
parallel bars containing an artery, a nerve, a carti-
lage, and a muscle anlage (Sadler 2006). They
are separated outside by clefts and internally by
four pharyngeal pouches. Embryologically, they
are precursors of the face, the neck, and the phar-
ynx (Adams et  al. 2016). Congenital branchial
cysts and fistulas are remnants of these embry-
onic structures, which have failed to regress com-
pletely. The first arch, cleft, and pouch form the
mandible ventrally and the maxillary process of
the upper jaw dorsally, the external ear, parts of
the Eustachian tube, and the tympanic cavity. The
most common branchial cysts and fistulas derive
from the second branchial pouch, which forms
the tonsillar fossa and the palatine tonsils. The
third arch forms the inferior parathyroid glands
and the thymus, while the fourth arch migrates Fig. 29.2  Infected thyroglossal cyst in the middle of the
less far down and develops into the superior para- neck
29  Thyroglossal and Branchial Cysts, Sinuses, and Fistulas
cases, the duct cysts may either contain papillary
cancer or squamous cell carcinoma. In cases with
papillary cancer, additional nodes can exist
within the thyroid gland, and/or regional affected
nodes may be present (Thompson et al. 2017, Van
Vuuren et al. 1994).
Although branchial cleft anomalies are present
at birth, they may not appear until a ­fluid-­filled
cyst is formed or becomes infected. Bilateral cysts
or sinuses can be observed in 10–15% of patients.
If a fistula of the second branchial cleft persists,
saliva is discharged periodically or continuously
at the skin ventral to the sternocleidomastoid
muscle. Although the majority of branchial anom-
alies are singular events, some cases of coinci-
dence in families point to a genetically determined
abnormality. Simple cystic remnants are present
commonly in adolescence and adulthood, while
sinuses and fistulas are usually diagnosed in
infancy and early childhood. In principle, clinical
manifestation–no matter what age—should be
taken as an indication for elective excision before
infectious complications supervene.
29.4 Diagnosis and Differential
Diagnosis
Thyreoglossal duct cysts are found in 60% of
cases in the midline, at or below the hyoid bone.
However, according to their origin, 24% are
located above the hyoid and 8% are intralingual
(El-Korbi et al. 2018). In the latter case, they can
cause respiratory distress or even sudden infant
death when located at the base of the tongue (Diaz
et al. 2005). Differential diagnosis must exclude
dermoid/epidermoid cysts, thymus cysts, lym-
phatic malformations, and metastatic diseases
(Patel and Bhatt 2019). During palpation, unin-
fected cysts are often ballotable and can be moved
slightly from side to side, but not up or down. Due
to their origin in the foramen cecum, the thyroid
cysts move upward during swallowing or when
the tongue protrudes. However, this clinical sign
is difficult to observe in small infants. The differ-
ential diagnosis includes complete ectopic thyroid
gland or parts of the thyroid, dermoid cysts, lipo-
mas, sebaceous cysts, submental lymphadenitis,
367
thymus cysts, or tumor. As mentioned above, the
latter group of pathologies may be moved up and
down with palpation, but do not move during
swallowing unless the pathology is in close con-
nection with the hyoid bone or the thyroid gland.
Dermoid cysts, in particular, may be located in
close vicinity to the hyoid bone and should there-
fore be treated in a similar way as thyroglossal
duct cysts to avoid a recurrent pathology. An
ultrasound investigation is often recommended as
a preoperative workup to verify the cystic struc-
ture of the pathology. However, the information is
limited, because it is difficult to exclude other cys-
tic pathologies, such as infected or uninfected der-
moid cysts or a lymph-­ node abscess. If the
ultrasound shows a solid structure, an ectopic thy-
roid gland may be present when the bilateral thy-
roid lobes are missing in their typical location.
Some authors state categorically that a thyroid
scan must be performed in all cases to exclude
ectopic thyroid tissue, which would result in
hypothyroidism if resected surgically (Radkowski
et  al. 1991; Wadsworth and Siegel 1994).
However, this anomaly is very rare (1–2%), and in
the case of a seemingly solid tumor, a frozen sec-
tion during surgery may be diagnostic and prevent
the resection of the whole gland.
Remnants of the second branchial cleft have an
incidence of more than 50%–90% and are the
most common branchial anomalies typically pre-
sented as a painless, smooth, slowly enlarging
mass along the anterior border of the sternoclei-
domastoid muscle (Bajaj et al. 2011). Unilateral
cysts are found on the right side in 89% of cases
(Maddalozzo et  al. 2012). The majority of the
anomalies are cysts with or without an additional
sinus tract; they may be painful and fluctuate in
size from time to time (Lee et al. 2018). The diag-
nosis is easier when clear mucous or saliva or pus
is discharged from an external opening (Fig. 29.3).
The internal fistula enters the supratonsillar
fossa and can be seen easily in adults. Injection of
water-soluble contrast material into the neck fis-
tula under X-ray control shows the extension of
the tract up to the pharynx. From the supratonsil-
lar fossa, the tract passes over the hypoglossal
nerve and behind the bigastric muscle through
the bifurcation of the carotid artery and in front
368
Fig. 29.3  Discharge of pus from the external opening of
a second bronchial cleft fistula
of the superior thyroid artery. The most common
complication is an infection of the cyst leading to
an abscess. More rare presentations are stridor,
tumor feeling in the throat with dysphagia, and
hypoglossal nerve palsy. Differential diagnoses
include suppurative lymphadenitis or dermoid
cyst, vascular anomalies, such as cystic hygroma,
or subcutaneous hemangioma. Some authors
suggest performing preoperative fine-needle
aspiration cytology in unclear cases (Lee et  al.
2018). Remnants of the first branchial cleft occur
with an incidence of approximately 5%–25% and
are more common in females (Golf et al. 2012).
They form small cysts at the posterior border of
the parotid gland or a sinus anterior to the ear in
close proximity to branches of the facial nerve.
Within the parotid gland, the relation to the facial
nerve is crucial (D’Souza et al. 2002). Most clas-
sifications divide them into two types. One type
is of ectodermal origin and occurs medial to the
concha but often extending into the postauricular
crease and ending at the osseous–cartilaginous
junction of the external meatus, thus, is superfi-
cial to the facial nerve. The other type has a sinus
opening below the angle of the mandible and
extends upward through the facial nerve ending
in or around the external auditory meatus (Work
1972). Infection of the cysts or sinus is common,
and pus may be discharged from the ear.
Diagnosis, as well as surgical therapy, are diffi-
cult and must avoid any damage to the facial
nerve. Cysts and sinuses are usually superficial,
while fistulas are deep to the facial nerve (Olsen
M. E. Höllwarth
et  al. 1980; Magdy and Ashram 2013). Thus,
either an ultrasound study or MRI is recom-
mended to demonstrate the course of the tract and
the relation to the nerve. The remnants of the first
branchial cleft need to be distinguished from the
preauricular pits, cysts, and sinuses, which are
remnants of the auditory tubercles. They are
located anterior to the tragus of the ear, are often
bilateral, and tend to be inherited. Very rarely,
there exists a coexistence of the first and second
branchial fistulas (Liu et al. 2020).
Remnants of the third and fourth branchial
cleft are uncommon and occur usually on the left
side (Golf et al. 2012). The external presentation
of the former anomaly is similar to the remnants
of the second branchial cleft, but the sinus tract of
the third cleft passes between the hypoglossal
and glossopharyngeal nerve and posterior to the
carotid artery to finally end up above the superior
laryngeal nerve within the pyriform sinus. The
epithelial tracts of the very rare remnants of the
fourth branchial cleft pass inferior to the superior
laryngeal nerve and enter also the pyriform sinus,
but on the left side in nearly all cases. Typical
symptoms are recurrent respiratory tract infec-
tion, hoarseness, and painful swallowing. If the
anomalies present as recurrent abscesses at the
left side of the thyroid gland, or in close vicinity
to the pyriform sinus, they may cause life-­
threatening respiratory stridor in neonates and
infants or acute unilateral thyroiditis in children
and adults. The appropriate diagnosis is difficult,
and the cyst may evoke a false impression of
acute thyroiditis. Most often, these anomalies are
treated, as they were harmless local lymphoid
abscesses, a mistake that makes later surgical
excision of the tract very difficult. MRI of the
neck helps to identify the origin of such lesions.
In an acute suppurative phase, external pressure
onto the mass may ensue a visible evacuation of
pus into the piriform fossa during laryngoscopy.
29.5 Therapy
Surgery of the thyroglossal cyst or duct must
always include resection of the middle part of the
hyoid corpus whether or not the surgeon has the
29  Thyroglossal and Branchial Cysts, Sinuses, and Fistulas
Fig. 29.4  The surgical procedure of a thyroglossal cyst
must include resection of the middle part of the hyoid
bone
feeling that the duct ends at the bone (Thompson
et al. 2016) (Fig. 29.4). The procedure was first
described by Sistrunk in 1920 based on embryo-
logical studies (Sistrunk 1920; Isaacson 2020).
In order to resect the hyoid bone, the upper
rim and the lower rim have to be freed from the
straight neck muscles—omohyoid and sternohy-
oid muscle—and the middle 1–2  cm can be
excised with strong scissors. In some cases, a
clear continuity of the duct behind the bone can
be seen, which then has to be resected up to the
base of the tongue including the high ligature of
the fistula. The procedure described by Sistrunk
consists of a fingertip of the index finger of the
surgeon directed downwards on the foramen cae-
cum, in order to guide the dissection (Righini
et al. 2016). If the pathology presents primarily
as an abscess, antibiotic therapy and a horizontal
incision are recommended. The use of a drain
depends on the size of the abscess. Some authors
recommend antibiotic therapy alone or an addi-
tional needle aspiration of the pus in order to
avoid seeding of ductal epithelium. We have not
seen any disadvantage after a primary surgical
369
incision of the abscess. The parents have to be
informed that a second surgical procedure with
excision of the duct will become necessary as
soon as the local inflammation is under control.
Recurrence only occurs when the middle part of
the hyoid corpus has not been resected appropri-
ately. Bleeding is minimal and a small suction
drain is rarely needed (Lillehei 2012).
Reconstruction of the straight neck muscles is
recommended to form the anterior aspect of the
neck.
Surgery of the lateral neck cysts or fistulas
starts in the case of a remnant of the second
branchial cleft with the excision of the visible
pathology at the anterior border of the sternoclei-
domastoid muscle. If an external opening exists,
a probe may be inserted, and methylene blue can
be injected to make the fistula easier to visualize
(Fig.  29.5). The duct is then isolated carefully
from the surrounding tissue avoiding any rough
procedure, which could damage the carotid arter-
ies within the bifurcation or the hypoglossal
nerve (Fig. 29.6). In order to isolate and ligate the
duct close to the supratonsillar region in older
children, it is useful to make a second stepladder
incision 3–5 cm above the first incision. As soon
as the excision is close to the supratonsillar fossa,
the anesthetist is asked to push the fossa down
with his finger in order to enable the surgeon to
ligate the central part of the fistula as close as
possible to the fossa (Höllwarth 2006). Drainage
Fig. 29.5  If injection of methylene blue is possible, the
fistula is easily visible during surgery
370
Fig. 29.6  The fistula
runs through the carotid
bifurcation and close to
the hypoglossal nerve.
Therefore dissection
must be close to the
fistula
of the wound is rarely necessary. Recurrences
only occur when the duct ruptures during the pro-
cedure and cannot be closed by ligature.
Surgery of the first branchial cleft remnants is
difficult, and knowledge of the embryology and
anatomy is crucial to avoid surgical mistakes. At
the beginning of the procedure, it is essential to
expose the facial nerve and all its horizontal and
lower branches. The surgical procedure should
start with the identification of the main trunk of
the nerve by means of a retroauricular incision.
The exit of the nerve from the stylomastoid fora-
men can be found in a triangle formed by the
sternomastoid muscle, the biastric muscle, and
the external ear canal. Once the branches of the
facial nerve are identified, minor or major parts
of the parotid gland must be resected to identify
the tract of the fistula. Intraoperative monitoring
of both mechanically elicited activity and elec-
trically evoked responses to the facial nerve
stimulation with a nerve integrity monitor is use-
ful (Magdy and Ashram 2013). The opening of
the fistula to the external ear canal should be
included in the resection to avoid any recur-
rences. Rarely, the duct opens into the middle ear
or runs parallel to the Eustachian tube. The
whole procedure is even more difficult in cases
with previous infection.
M. E. Höllwarth
Hypoglossal nerve
Carotid bifurcation
The special anatomy in the rare case of rem-
nants of the third and fourth branchial cleft and
the risk to damage the superior or the recurrent
laryngeal nerve makes the surgical procedure dif-
ficult. Several approaches have been recom-
mended, each of them bears its special risks
(Madana et  al. 2011; Pereira et  al. 2004).
Recently, closure of the internal opening at the
piriform sinus by injecting sclerosing material
(OK-432) endoscopically into the tract has been
recommended, which may be an especially help-
ful solution after previous repeated infections
(Nicoucar et al. 2010, 2009; Roh et al. 2006; Kim
et  al. 2009; Nixon and Healey 2011). In three
patients with a fourth branchial pouch anomaly
under 1 year of age, the sinus opening in the left
piriform fossa was successfully endoscopically
coagulated with a monopolar diathermy in order
to seal the opening (Bajaj et al. 2011).
29.6 Conclusion
Thyroglossal cysts and branchial anomalies are
common in children. Conservative therapy can-
not be recommended. Surgical therapy seems to
be simple, but there are several pitfalls. Therefore,
a detailed knowledge of the embryology and
29  Thyroglossal and Branchial Cysts, Sinuses, and Fistulas
pathology is essential. Rarer causes of these mal-
formations deriving from the first, third, and
fourth branchial arch, as well as unusual presen-
tations of the thyroglossal cyst, often need addi-
tional diagnostic procedures, such as ultrasound
and CT/MRI.
References
Adams A, Mankad K, Offiah C, Childs L (2016) Branchial
cleft anomalies: a pictorial review of embryologi-
cal development and spectrum of imaging findings.
Insights Imaging 7:69–76
Bajaj Y, Ifeacho S, Tweedie D et  al (2011) Branchial
anomalies in children. Int J Pediatr Otorhinolaryngol
75:1020–1023
D’Souza AR, Uppal HS, Zeitoun H et al (2002) Updating
concepts of first branchial cleft defects: a literature
review. Int J Pediatr Otorhinolaryngol 62:103–109
Diaz MC, Stormorken A, Christopher MC (2005) A thy-
roglossal duct cyst causing apnea and cyanosis in a
neonate. Pediatr Emer Care 21:35–37
El-Gohary Y, Gittes G (2011) Congenital Cysts and
Sinuses of the neck. In: Puri P (ed) Newborn surgery,
3rd edn. Hodder Arnold, London
El-Korbi A, Bouatay R, Houas J et al (2018) A rare loca-
tion of thyroglossal duct cyst in a newborn. Pan Afr
Med J 31:104
Foley DS, Fallat ME (2006) Thyroglossal duct and other
congenital midline cervical anomalies. Sem Pediatr
Surg 15:70–75
Golf CJ, Allred C, Glade RS (2012) Current manage-
ment of congenital branchial cleft cysts, sinuses,
and fistulae. Curr Opin Otolaryngol Head Neck Surg
20:533–539
Höllwarth EM (2006) Branchial Cyst and Sinus. In:
Puri P, Höllwarth M (eds) Pediatric surgery. Springer
Surgery Atlas Series. Springer, Berlin, Heidelberg
Huang LD, Gao SQ, Dai RJ, et al (2015) Intra.throid thy-
roglossal duct cyst: a case report and review of the
literature. Int J Clin Exp Pathol 8:7229–7233. eCol-
lection 2015
Isaacson G (2020) Sistrunk centennial: evolution of a
classic operation. Laryngoscope 130:E46–E47
Kim MG, Lee NH, Ban JH et al (2009) Sclerotherapy of
branchial cleft cysts using OK-432. Otolaryngol Head
Neck Surg 141:329–334
Lee DH, Yoon TM, Lee JK, Lim SC (2018) Clinical study
of second branchial cleft anomalies. J Craniofac Surg
29:e557–e650
Lillehei C (2012) Neck Cysts and Sinuses. In: Coran AG,
Adzick NS, Krummel TM, Laberge J-M, Shamberger
RC, Caldamone AA (eds) Pediatric surgery, 3th edn.
Elsevier
Liu D, Li G, Wang J, Pei G (2020) Unusual coexis-
tence of first and second branchial fistulas: clini-
371
cal case and review of the literature. J Int Med Res
48:300060520944303
Madana J, Yolmo D, Kalaiarasi R et al (2011) Recurrent
neck infection with branchial arch fistula in children.
Int J Pediatr Otorhinolaryngol 75:1181–1185
Maddalozzo J, Rastatter JC, Dreyfuss HF et  al (2012)
The second branchial cleft fistula. Int J Pediatr
Otorhinolaryngol 76:1042–1045
Magdy EA, Ashram YA (2013) First branchial cleft anom-
alies: presentation, variability and safe surgical man-
agement. Eur Arch Otorhinolaryngol 270:1917–1925
Nicoucar K, Giger R, Pope HG Jr et al (2009) Management
of congenital fourth branchial arch anomalies: a
review and analysis of published cases. J Pediatr Surg
44:5–10
Nicoucar K, Giger R, Jaecklin T et al (2010) Management
of congenital third branchial arch anomalies: a
systematic review. Otolaryngol Head Neck Surg
142(21–28):e2
Nixon PP, Healey AE (2011) Treatment of a branchial
sinus tract by sclerotherapy. Dentomaxillofac Rdiol
40:130–132
Olsen KD, Maragos NE, Weiland LH (1980) First bran-
chial cleft anomalies. Laryngoscope 90:423–436
Patel S, Bhatt AA (2019) Thyroglossal duct pathology and
mimics. Insight Imaging 10:12
Pereira KD, Losh GG, Oliver D et al (2004) Management
of anomalies of the third and fourth branchial pouches.
Int J Pediatr Otorhinolaryngol 68:43–50
Radkowski D, Arnold J, Healy GB et  al (1991)
Thyroglossal duct remnants: preoperative evaluation
and management. Arch Otolaryngol Head Neck Surg
117:1378–1381
Righini C-A, Hitter A, Reyt E, Atallah I (2016)
Thyroglossal duct suregy. Sistrunk procedure. Eur
Ann Otorhinolaryngol Head Neck Dis 133:133–136
Roh JL, Sung NH, Kim H et al (2006) Treatment of bran-
chial cleft cyst with intracystic injection of OK-432.
Acta Otolaryngol 126:510–514
Sadler TW (2006) Medical embryology, 10th edn.
Lippincott Williams & Wilkins
Sistrunk WE (1920) The surgical treatment of cysts of the
thyroglossal tract. Ann Surg 71:121–124
Thompson LDR, Herrera HB, Lau SK (2016) A clinico-
pathologic series of 685 thyroglossal duct remnant
cysts. Head Neck Pathol 10:465–474
Thompson LDR, Herrera HB, Lau SK (2017) Thyroglossal
duct cyst carcinomas in pediatric patients: report of
two cases with a comprehensive literature review.
Head Neck Pathol 11:442–449
Van Vuuren PA, Balm AJ, Gregor RT et  al (1994)
Carcinoma arising in thyroglossal duct remnants. Clin
Otolaryngol 19:509–515
Wadsworth DT, Siegel MJ (1994) Thyroglossal duct cysts:
variability of sonographic findings. Am J Roentgenol
163:1475–1477
Waldhausen JHT (2006) Branchial cleft and arch anoma-
lies in children. Sem Pediatr Surg 15:64–69
Work WP (1972) Newer concepts of the first branchial
cleft defects. Laryngoscope 82:1581–1593
 Tracheostomy
Lina Woods, Thom E. Lobe, and John Russell
30.1 Introduction and Incidence
Tracheostomy may be required for several con-
genital and acquired conditions of infancy and
childhood. Pediatric surgeons should have a good
understanding of appropriate indications, surgi-
cal technique, and perioperative care, and should
be competent to perform the procedure under
urgent or elective conditions.
The incidence of pediatric tracheostomy is
reported at approximately 2% in intensive care set-
tings in the United Kingdom, with a range of 0.13–
5.66% (Wood et  al. 2012). Similar data from a
large multicenter North American study revealed a
rate of 1.37% of pediatric ICU admissions receiv-
ing a new tracheostomy, while 0.6% had a trache-
ostomy already in place (Edwards et al. 2016).
While neonates who require urgent attention
to their airway are occasionally encountered, the
commonest indication for tracheostomy is in
those requiring long-term ventilation. Recent
L. Woods
Department of Pediatric Otorhinolaryngology,
Children’s Health Ireland, Dublin, Ireland
T. E. Lobe
University of Tennessee Health Science Center,
Memphis, TN, USA
e-mail: Lobet2@ihs.org
J. Russell (*)
Department of Pediatric Otorhinolaryngology,
Children’s Health Ireland, Crumlin, Dublin, Ireland
e-mail: john@drjrussellent.com
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_30
30
consensus guidelines on perioperative pediatric
tracheostomy care have helped to guide clini-
cians where areas of debate exist with this proce-
dure (Strychowsky et al. 2016).
Opinions about pediatric tracheostomy vary
widely among pediatric intensivists. Some
believe that the procedure is underutilized, while
others think that the risks outweigh the benefits
(Principi et al. 2008). Indeed, when compared to
adult tracheostomies, pediatric tracheostomies
are associated with an increased rate of compli-
cations (Mahida et al. 2016). Complication rates
are also higher in younger children and those
with significant underlying disease (Funamura
et al. 2017; Mahida et al. 2016). Rates of decan-
nulation are significantly lower in children com-
pared to adults, and the time to decannulation can
average around 2  years (de Trey et  al. 2013;
Mahadevan et  al. 2007; Ozmen et  al. 2009;
Watters 2017). Therefore, it is important to con-
sider whether or not the indication for insertion is
appropriate.
This chapter will discuss tracheostomy indica-
tions, surgical techniques, and maintenance care
in infants and children.
30.2 Etiopathogenesis
Tracheostomy is one of the oldest reported surgi-
cal procedures in the medical literature (Carroll
and Pahor 2001; Jackson 1909). The procedure
373
374
was standardized in the published literature
through the work of Chevalier Jackson and has an
important role in pediatric airway management.
Early pioneers emphasized the importance of
postoperative care in reducing associated mor-
bidity and mortality (Woods et  al. 2019). The
incidence of pediatric tracheostomy has varied
over the past 100  years; however, the improved
survival of premature, critically ill, and neuro-
logically impaired infants has led to an increasing
number of pediatric tracheostomies in recent
years (Dal’Astra et al. 2017; Doherty et al. 2018;
Gergin et  al. 2016; Wetmore et  al. 1999).
Furthermore, indications for tracheostomy have
expanded to include use during ex utero intrapar-
tum treatment (EXIT), which was first described
in 1989 and is now an established practice for
patients with prenatally diagnosed suspicion of
airway obstruction.
30.3 Pathophysiology
The leading indications for tracheostomy in
childhood before mass vaccination were infective
causes of airway compromise, such as croup, epi-
glottitis, and diphtheria. However, the need for
tracheostomy in infective airway disease is now
rare. Advances in neonatal respiratory care, par-
ticularly in the case of premature infants needing
prolonged ventilation, as well as increased use of
endotracheal intubation, have led to increased
demand for tracheostomy. Indeed, children with
complex medical comorbidities can now survive
longer, with tracheostomy facilitating home ven-
tilation. It has even been suggested that earlier
tracheostomy in ventilated patients may be asso-
ciated with improved outcomes (Lee et al. 2016;
Pizza et al. 2017).
The commonest indications for tracheostomy
in childhood are now those requiring prolonged
mechanical ventilation due to respiratory or neu-
rological failure or those with upper airway
abnormalities. Upper airway obstruction can be
due to congenital anomalies, such as those with
severe Pierre Robin sequence and other craniofa-
cial syndromes, or acquired complications of
prolonged endotracheal intubation or trauma. In
the case of prolonged ventilation, there are no set
L. Woods et al.
criteria regarding the timing of tracheostomy in
children. Infants can tolerate endotracheal intu-
bation for longer than adults and some premature
infants remain intubated for months before tra-
cheostomy is considered, even in cases of multi-
ple failed extubations. However, earlier
tracheostomy may be helpful for weaning by
decreasing the work of breathing. Additionally, a
longer duration of ventilation before tracheos-
tomy is associated with increased morbidities
and stay in intensive care (Holloway et al. 2015;
Zhu et al. 2015). It has been suggested that after
2 weeks of intubation in a child that is stable on a
ventilator, evaluation for tracheostomy should be
considered (Watters 2017).
30.4 Pathology
Chronic lung disease, neurological impairment,
and upper airway anomalies have been demon-
strated as the most common underlying comorbid
conditions in children requiring tracheostomy in
retrospective multicenter research (Berry et  al.
2009). Approximately one-third of children requir-
ing tracheostomy are infants and the majority of
patients have significant comorbidities with
requirement for nutritional support (Deutsch 2010;
Mahida et al. 2016). Older children rarely require
a tracheostomy for a congenital problem, but are
more likely to need one due to acquired conditions
resulting from infection, tumor, or trauma.
Coinciding with advancements in prenatal
imaging, the role of EXIT has increased signifi-
cantly and indications have expanded (Norris et al.
1989), although the majority of indications for tra-
cheostomy placement are in cases of CHAOS
(congenital high airway obstruction syndrome),
large head and neck mass, or severe retrognathia
(Bence and Wagner 2019; Kumar et al. 2019).
Although children are significantly less likely
to suffer severe effects of COVID-19, tracheos-
tomy is generally not indicated for patients with
respiratory failure secondary to COVID-19.
During the pandemic, it has been advised that
elective tracheostomy for ambulatory patients be
delayed whenever possible pending local
resource availability, while endoscopic mini-
mally invasive airway procedures are preferred
30 Tracheostomy
whenever possible to avoid the need for intuba-
tion or tracheostomy placement. However, the
potential risks and benefits of multiple proce-
dures must be weighed against tracheostomy
placement (Bann et al. 2020).
30.5 Diagnosis
Pediatric tracheostomy is resource-intensive,
often requiring several days of ICU-level care
with mechanical ventilation (Strychowsky et  al.
2016). Most infants needing a tracheostomy
already have an endotracheal tube in place. Each
patient should undergo an appropriate history
and a physical examination that includes an
assessment of neck mobility and anatomical
landmarks. The neck skin should be assessed,
and any preoperative skin condition treated
appropriately. The surgeon should ascertain
coagulation status and complete blood count by
laboratory investigation. Further investigation,
such as electrolyte status and nutritional status,
will be guided by patient comorbidities.
Tracheostomy in children has a significant
impact on parents and caregivers, and requires
early engagement, guidance, and counseling.
Adequate information and education regarding
the procedure including benefits, risks, and the
expected postoperative course is recommended.
Discussion should include the relatively high in-­
hospital mortality rate of children undergoing
tracheostomy at approximately 10%, although a
minority of these deaths are related to the trache-
ostomy itself, and the tracheostomy-related com-
plication rate of 15–19% (Berry et al. 2009).
Care needs, dependencies, and family impact
should be addressed preoperatively. Dedicated
clinical nurse specialists are a particularly impor-
tant part of the team managing children with tra-
cheostomies and can provide significant
emotional support. Preoperative evaluation with
specialists in communication and feeding can be
of value where appropriate.
Cases requiring EXIT need significant prena-
tal planning to ensure optimal chances of suc-
cess. Prenatal planning, along with EXIT
procedures, requires a large multidisciplinary
team comprising pediatric otolaryngologists,
375
radiologists, specialist obstetricians, maternal
and fetal anesthetists, maternal/fetal medical spe-
cialists, and neonatologists, among others. As
there is a high rate of adverse events associated
with EXIT, thorough preplanning may help to
reduce these risks. Appropriate multimodality
imaging in the form of ultrasound and MRI can
greatly assist decision-making, as well as prena-
tal echocardiography.
30.6 Differential Diagnosis
An alternative treatment option to tracheostomy
is long-term endotracheal intubation. However,
this carries the risk of complications, such as sub-
glottic stenosis. It also requires sedation in an
appropriate intensive care setting. In some cir-
cumstances, patients who might historically have
required a tracheostomy can now sometimes be
managed with alternative airway devices, for
example with the use of a nasopharyngeal airway
in patients with Pierre Robin syndrome.
30.7 Management
30.7.1 Technique
Prior to performing a tracheostomy on a child, it
is important to consider the differences between
adult and pediatric airways. In general, infants
have shorter and fatter necks with a larynx that is
one-third the size of an adult and lies more supe-
rior in the neck around the third or fourth cervical
vertebra before descending at around 2 years of
age. The hyoid often overlies the thyroid carti-
lage notch and the narrowest portion of the air-
way in the child is at the cricoid cartilage.
Relevant imaging should be reviewed before
starting the case to identify potential hazards; for
example, deviation of the airway from a large
neck mass.
The selection of appropriate tube size is impor-
tant prior to surgery. This should be appropriate
for both the size of the airway and the indication
for placement of the tracheostomy tube. A tube
that is too small may not permit sufficient air
exchange, and a tube that is too big might cause
376 L. Woods et al.

tracheal mucosal injury and stenosis. Appropriate

tube size can be estimated based on the formula
for the internal diameter of an endotracheal tube
for children over 1 year old (age in years/4 + 4 mm).
Cuffless tubes are usually used in infants; how-
ever, specially designed cuffed tubes can be used
in certain instances to reduce air leaks.
Please see Figs.  30.1, 30.2, 30.3, 30.4, 30.5,
30.6 demonstrating the surgical approach. Under
general anesthetic with an endotracheal tube in

Fig. 30.3  The midline is dissected until the trachea can

be visualized easily

Fig. 30.1  The patient is placed supine on the operating

table. The neck should be extended sufficiently to allow
complete access to the neck

Fig. 30.4  A monofilament stay suture is placed on either

side of the midline in the anterior trachea

situ, the patient is positioned with the neck

Fig. 30.2  Incision is made in the lower neck crease,
about one centimeter below the cricoid cartilage
extended and a shoulder roll in place in the supine
position. Depending on the indication, it may be
appropriate to carry out laryngotracheobronchos-
copy prior to the incision to determine the status
of the trachea and the original indication for tra-
30 Tracheostomy 377

The whole of the child’s neck should be

exposed and prepared in a sterile fashion and
draped to allow sufficient access for the surgeon
and anesthesiologist. A horizontal skin incision is
marked approximately one centimeter below the
cricoid cartilage or midway between the cricoid
cartilage and sternal notch, and local anesthetic
infiltrated. Following skin incision, subcutaneous
fat is removed to improve visualization and
access, and also to reduce the incidence of false
passage. The strap muscles are divided in the
midline to the level of the trachea and retracted
laterally. There are usually few blood vessels
encountered if dissecting in the midline, but
bipolar cautery is used throughout the procedure
to ensure that the surgical field is kept dry. The
thyroid isthmus is identified and divided, if nec-
essary, with bipolar diathermy. Intermittent pal-
pation during the case can help to establish the
midline (Fig. 30.7).
Fig. 30.5  A vertical incision is made through the tracheal
wall between the stay sutures
Nonabsorbable stay sutures, such as 4–0
prolene or equivalent, are placed on either side of
the intended incision site. These sutures are not
tied onto the tracheal wall, but can be tied at their
ends, which should be left 6–8 cm in length. At
the end of the case, the end of these sutures will
be taped securely to the anterior chest wall and
labeled ‘left’ and ‘right’. Color coding can be
considered. These sutures will be used to locate

Fig. 30.6  Stomal maturation sutures are placed between

the trachea and the skin after the removal of subcutaneous
fat at the beginning of the case

cheostomy. Close coordination with anesthesiol-

ogy colleagues is important during tracheostomy.
The patient should be prepared in such a way as
to allow easy access for tube removal by the Fig. 30.7  The inserted tracheostomy tube is connected to
anesthesiologist. the ventilator circuit and the tube is secured to the patient
378
the tracheal incision in the event of a postopera-
tive emergency in case the newly placed trache-
ostomy dislodges.
Maturation sutures should be considered. In
this case, four-quadrant absorbable maturation
sutures are placed between the trachea and skin
edges. These are placed further afield than the
stay sutures and create an environment for early
tract formation to reduce risk from decannula-
tion. Maturation sutures are often used as an extra
safety measure in pediatric tracheostomy cases to
ensure adequate tract formation and have been
shown to reduce complications including acci-
dental decannulation rates (Woods et  al. 2019).
There has been a concern for the increased inci-
dence of tracheocutaneous fistula with the use of
maturation sutures; however, various studies
reveal no difference in fistula rates between sto-
mas with and without maturation sutures (Woods
et al. 2019).
After communication with the anesthesiolo-
gist to ensure that they are prepared to begin tube
withdrawal, a tracheal incision is then made
between the stay sutures. Debate exists as to the
orientation of the tracheal incision in children. A
vertical incision through the second-to-fourth tra-
cheal rings is most frequently used in infants and
young children to avoid the potential for tracheal
transection or deformity; however, some centers
prefer the use of a horizontal incision between
the third and fourth tracheal rings, regardless of
age (Woods et al. 2019). Suction should be avail-
able to minimize wound contamination and aspi-
ration, and to ensure a clear view.
Following further essential communication
with the anesthesiologist, the endotracheal tube is
partially withdrawn under direct vision and the
tracheostomy tube is placed. The tip of the tra-
cheostomy tube to be inserted should be lubri-
cated with a water-soluble surgical lubricant and
positioned over the incision, and poised for inser-
tion when the endotracheal is withdrawn. If, for
any reason, the tracheostomy tube does not fit
easily into the trachea, then the tracheostomy
tube should be removed, and the endotracheal
tube should be advanced beyond the tracheal
incision, so that ventilation will not be compro-
mised. This might occur if the tube selection was
L. Woods et al.
incorrect and so a smaller size should always be
available for insertion after it is certain that the
infant is well oxygenated.
As soon as the tracheostomy tube is in place,
the obturator should be removed, and the anes-
thesiologist should disconnect the ventilator hose
from the endotracheal tube and connect it to the
tracheostomy tube. Once that is done, the anes-
thesiologist should administer several deep
breaths to the patient to confirm that the tube is in
the proper place and that the infant can be venti-
lated satisfactorily. Tracheoscopy can also be
performed at this stage to confirm a satisfactory
position, at least two-to-three tracheal rings
above the carina, as well as assessing for the
presence of distal compression or obstruction.
Once adequate ventilation is confirmed, the
endotracheal tube can be removed completely.
The tube is secured with cotton ties. Trach
gauze, exudry, or mepilex dressing is placed
under the flanges or under the neckties.
Alternatives to securing the tube include velcro
ties without skin sutures; velcro ties without skin
sutures but consider suturing the Velcro to itself,
so that it cannot be opened; skin suture. The use
of skin sutures to secure the tube carries a risk of
accidental decannulation occurring without any-
one being aware. Finally, the stay sutures are
taped to the neck and chest and labeled.
EXIT procedures are usually performed under
general anesthetic to achieve uterine relaxation
and good uteroplacental perfusion to maximize
the time available for intervention (Kumar et al.
2019). While the fetus remains attached to the
umbilical cord and placenta, intubation by direct
laryngoscopy is initially attempted, followed by
surgical intervention if endotracheal intubation is
not achieved. Surgical intervention may include
rigid laryngotracheobronchoscopy, tracheos-
tomy, or partial excision of a mass. Subsequently,
the umbilical cord is cut and high doses of oxyto-
cin are administered (Novoa et  al. 2020).
Tracheostomy in this setting may need to be
revised once the airway is secure and the patient
is stabilized.
It should be noted that percutaneous tracheos-
tomy has been used in some centers for children;
however, due to safety concerns, its role has
30 Tracheostomy
mostly been limited to older children and adoles-
cents in centers that have experience in the tech-
nique. Evidence for percutaneous tracheostomy
is lacking and it is not considered routine in the
pediatric population.
30.7.2 Postoperative Management
In the immediate postoperative period, a chest
X-ray is performed in the recovery room or inten-
sive care setting to exclude a pneumothorax.
Initial care post-tracheostomy in a child usually
takes place in the intensive care setting at least
until the first tube change. Antibiotic prophylaxis
should be considered until the first tracheostomy
tube change in all patients.
Younger patients and those at risk of decan-
nulation, or patients who undergo tracheostomy
in cases with a critical airway, may require seda-
tion and chemical paralysis, whereas others may
require only postoperative analgesia. Sedation
requirements should be assessed daily. In sedated
patients, morphine and midazolam are often used
intravenously until enteral feeding is established,
at which point chlorohydrate and clonidine are
also used. This is to promote rapid tract healing
and to avoid movement which could precipitate
accidental decannulation or development of a
false passage. Where possible, enteral feeding
should begin and be tailored according to gastric
residual volume (Strychowsky et  al. 2016).
Patients are nursed with the head raised at 15–30
degrees and are given a proton-pump inhibitor
and laxatives.
In the early period postoperatively, from an
airway point of view, the ventilation should be
weaned as far as possible, while humidification
and regular suctioning at four hourly intervals
should take place as well as when needed. The
suction length should be measured so as not to
extend beyond the tip of the tube. A standardized
up-to-date information sheet should be placed
above the bed with specific data on the type of
tracheostomy tube, size of tracheostomy tube
(including cuff, inner diameter, outer diameter,
and length), cuff inflation, suctioning distance,
critical alert if applicable, and date of first trache-
379
ostomy change and subsequent tracheostomy
changes. Spare tubes, one of the same size and
one smaller, should be kept at the bedside. Where
cuffed tubes are used, cuff pressures should be
checked regularly.
At 4 hourly intervals, the skin should be
checked for signs of infection or breakdown. If
there is any concern, the gauze should be
removed, the site should be cleaned thoroughly
with saline or quarter strength hydrogen perox-
ide, and an antibiotic ointment can be applied to
the incision site. When moving the patient, pre-
cautions should be taken to ensure that the head
is kept midline. The ventilation tubing should be
checked to ensure that it is not placing tension on
the skin. The ties should be changed daily or if
they become wet, and the tightness should be
checked every nursing shift.
The first tube change is recommended to be
carried out by the airway surgeon and an inten-
sivist (Woods et  al. 2019). As an example of a
protocol for the first tube change, it can be carried
out at the bedside in the intensive care setting
with the neck in extension and after a bolus of
propofol (1–2  mg/kg), with enteral feeding
ceased four hours prior. Equipment should
include suction, a head light, and an exact replica
of the original tube placed. A full tracheostomy
set and an intubating introducer should be
available.
The first tracheostomy tube change is usually
performed by the surgeon to ensure that the stoma
has sufficiently matured. Further tube changes
can be undertaken by nursing staff and the parent
or caregiver education can be commenced.
Opinions on the timing of the first tracheostomy
tube change are variable, but recent guidelines
have stated that the timing of the first change
should be at the discretion of the otolaryngolo-
gist, with variation between 3 and 7  days, and
most performing the first change between days 5
and 7. Further recent research has suggested that
a tube change can safely be performed on the sec-
ond day postoperatively, facilitated by using mat-
uration sutures, and has the potential for several
advantages including shorter length of stay in
intensive care, reduced cost, lower requirement
for heavy sedation, less complications from with-
380
drawal of sedation, earlier wound assessment and
the reduction of potential wound complications,
earlier placement of fresh tracheostomy ties,
earlier start to caregiver training, and earlier
­ improvement in swallowing skills, and it may
patient mobility and rehabilitation (Woods et al.
2019). The stay sutures should be left untouched
for the period until the first tube change, after
which they can be cut.
Regarding COVID-19, tracheostomy patients
with unknown, suspected, or positive virus status
should be managed on a closed respiratory circuit
with in-line suction until the infection is cleared
or testing is performed and is negative (Chan
et al. 2020).
30.7.3 Home Instruction and Care
Before a patient is discharged home with a tra-
cheostomy in situ, multidisciplinary input is nec-
essary to ensure that the family and caregivers
have been appropriately trained in managing all
aspects of tracheostomy care and maintenance.
Typically, a tracheostomy kit is prepared to
accompany the patient, which will include extra
tubes of varying sizes, suction equipment, and a
suction machine, as well as items for stoma care.
Parents and caregivers should be trained in
and evaluated on their ability to change a tube. It
is often helpful to arrange for home nursing visits
until the family becomes familiar and comfort-
able with the new devices. This is especially true
when it comes time to the first scheduled trache-
ostomy change, if it is to be done at home.
Devices to humidify air are physiologically
important in tracheostomy patients. Heat-and-­
moisture exchangers (HMEs) are filter devices
for heat and moisture exchange. They are placed
on the hub of the tracheostomy tube and can help
speech quality as well as reduce mucus produc-
tion. Significant secretions may preclude the use
of these devices. Sterile saline drops placed into
the tracheostomy tube or saline nebulization can
be helpful if secretions become thick and difficult
to suction.
In children who are awake and responsive,
medically stable, tolerate cuff deflation, have a
patent upper airway, and are able to manage oral
L. Woods et al.
and tracheal secretions, consideration should be
given to a monitored trial of a speaking valve as
it allows spontaneous voice and expression,
also be helpful in the decannulation of patients
(Watters 2017).
In COVID-19 cases, if a closed circuit is
unavailable, the use of HME devices with an inte-
grated hydroscopic viral/bacterial filter has been
advised, if tolerated by the patient (Bann et  al.
2020). The use of filter HMEs is also recom-
mended for ambulatory tracheostomy patients, if
tolerated, to potentially reduce the risk of acquir-
ing COVID-19 (Bann et al. 2020).
30.7.4 Complications
Despite a relatively high in-hospital mortality
rate of children undergoing tracheostomy at
approximately 10%, only a minority of these
deaths are related to the tracheostomy itself. The
majority of deaths in tracheostomized pediatric
patients are related to medical comorbidities and
occur after the index hospitalization during which
the tracheostomy was performed (Funamura
et al. 2017). Most tracheostomy-related mortali-
ties are caused by tube obstruction, tube mis-
placement, and accidental decannulation.
The tracheostomy-related complication rate is
approximately 15–19% (Berry et  al. 2009).
Tracheostomy complications can be divided into
early and delayed events. Early complications
include pneumothorax, pneumomediastinum,
subcutaneous emphysema, hemorrhage, cricoid
injury, recurrent laryngeal nerve injury, pulmo-
nary edema, respiratory arrest, mucosal injury,
esophageal injury, creation of a false passage,
mucus plugging, and accidental decannulation.
Prevention of most of these complications is pos-
sible through careful and thorough management
perioperatively and intraoperatively.
Delayed complications include mucus plug-
ging, accidental decannulation, stomal granula-
tion, tracheocutaneous fistula, tracheal
granuloma, suprastomal collapse, subglottic ste-
nosis, hemorrhage (stomal, tracheal, and tracheo-
innominate fistula), tracheoesophageal fistula,
30 Tracheostomy
and swallow dysfunction. Stomal granulation can
be managed with more frequent dressing and
tube changes and close attention to wound care,
with the occasional need for silver nitrate cau-
tery. The best management for a tracheocutane-
ous fistula is not clear from the literature, with
many different approaches described and no sig-
nificant difference between them and so a deci-
sion on timing and technique of surgical closure
is best made depending on the surgeon and fam-
ily preference, as well as resource availability
(Cheng and Setabutr 2016; Osborn et al. 2013).
Suprastomal granuloma can usually be man-
aged endoscopically prior to decannulation.
Tracheoesophageal fistulation, suprastomal col-
lapse, and subglottic stenosis are complex airway
complications that may require significant open
airway reconstructive surgery. Management of
delayed bleeding will be dependent on the identi-
fication of the source. Tracheoinnominate fistula
is usually fatal, and any suspicion of impending
fistulation should be imaged and referred to a car-
diothoracic surgeon.
Complications can occur during tube change
and include a difficult tube change, creation of a
false passage, subcutaneous emphysema, pneu-
mothorax, and an inability to adequately and
safely secure the patient’s airway (Deutsch 1998).
The use of maturation sutures is often used as an
extra safety measure in pediatric tracheostomy
cases to ensure adequate tract formation and has
been shown to reduce complications including
accidental decannulation rates (Craig et al. 2005;
Park et al. 1999). Artificial tract maturation gives
the added benefit of facilitating a safe earlier tube
change.
It is worth noting that evidence suggests a
higher rate of complications from tube change
outside the ICU setting (Tabaee et al. 2007).
30.7.5 Decannulation
As soon as the child’s condition allows, the ulti-
mate goal of tracheostomy management should
be decannulation. Timing depends to a large
degree on the reason for the tracheostomy. If the
underlying airway obstruction resolves or is
381
treated, or if the airway size increases with the
growth of the child, decannulation can be consid-
ered. Assessment to ensure safety is crucial to
ensure there is low risk with decannulation, as
failure rates can be from 6.5 to 21.4% (Watters
2017).
Multiple decannulation protocols exist, but, at
a minimum, should include resolution of the
underlying indication for tracheostomy and the
child being off ventilation for at least 3 months
prior. The child should undergo flexible laryn-
goscopy to assess vocal cord movement and air-
way collapse and microlaryngoscopy with
bronchoscopy with the treatment of any airway
lesion that could preclude decannulation, such as
a granuloma. During this procedure, the trache-
ostomy should be removed to assess for airway
collapse or obstruction. Adenotonsillar hypertro-
phy may need to be treated if present. Functional
assessment, such as a sleep study with a tracheos-
tomy cap in situ, should be considered. In most
cases, the tube size should be changed down (no
lower than size 3  in infants) and capped with
observation in the hospital. In cases with narrow
airways, such as young infants, or with tracheo-
malacia, a capping trial may not be appropriate.
It has been shown that tracheocutaneous fis-
tula rates are higher in patients with a longer time
to decannulation (Ha et  al. 2017; Woods et  al.
2019). While time to decannulation is often
affected by multiple factors, such as patient fol-
low-­up and medical comorbidities, it is useful to
consider that earlier decannulation, where possi-
ble, will likely lead to higher rates of closure
without the need for surgery. Tracheocutaneous
fistula rates are higher in those with a longer time
to decannulation and further research is needed
to clarify the most appropriate approach to the
management of persistent tracheocutaneous
fistula.
30.8 Conclusion
Tracheostomy is being performed on children
with increasing frequency and is usually per-
formed surgically. The indications have changed
over the past 100 years and it is now most com-
382
monly performed in patients requiring prolonged
mechanical ventilation. Multidisciplinary preop-
erative assessment is valuable. There are some
areas of debate regarding the best surgical tech-
nique. Care should be taken to minimize early
complications related to poor technique.
Postoperative management should take place
in the intensive care setting. Sedation is fre-
quently necessary for children until the first tube
change, which can be performed early if stomal
maturation sutures have been used and the condi-
tion of the patient is appropriate. Delayed com-
plications may require the input of a specialist
airway surgeon. Clinical nurse specialists play an
important role in perioperative care. Each center
should have a protocol for decannulation. Ex
utero intrapartum treatment may necessitate tra-
cheostomy in a high-pressure setting and requires
significant planning. Guidelines have been devel-
oped regarding tracheostomy management dur-
ing the COVID-19 pandemic and should be
adhered to.
References
Bann DV, Patel VA, Saadi R, Goyal N, Gniady JP, McGinn
JD et  al (2020) Best practice recommendations for
pediatric otolaryngology during the COVID-19 pan-
demic. Otolaryngol Head Neck Surg 162(6):783–794
Bence CM, Wagner AJ (2019) Ex utero intrapartum
treatment (EXIT) procedures. Semin Pediatr Surg
28(4):150820
Berry JG, Graham DA, Graham RJ, Zhou J, Putney HL,
O’Brien JE et  al (2009) Predictors of clinical out-
comes and hospital resource use of children after tra-
cheotomy. Pediatrics 124(2):563–572
Carroll CM, Pahor A (2001) The history of tracheotomy. J
Ir Coll Physicians Surg 30(4):237–238
Chan JYK, Wong EWY, Lam W (2020) Practical aspects
of otolaryngologic clinical services during the 2019
novel coronavirus epidemic: an experience in Hong
Kong. JAMA Otolaryngol Head Neck Surg
Cheng J, Setabutr D (2016) Tracheocutaneous fistula
closure in children. Int J Pediatr Otorhinolaryngol
89:107–111
Craig MF, Bajaj Y, Hartley BE (2005) Maturation sutures
for the paediatric tracheostomy—an extra safety mea-
sure. J Laryngol Otol 119(12):985–987
Dal'Astra AP, Quirino AV, Caixeta JA, Avelino MA (2017)
Tracheostomy in childhood: review of the literature
on complications and mortality over the last three
decades. Braz J Otorhinolaryngol 83(2):207–214
L. Woods et al.
de Trey L, Niedermann E, Ghelfi D, Gerber A, Gysin C
(2013) Pediatric tracheotomy: a 30-year experience. J
Pediatr Surg 48(7):1470–1475
Deutsch ES (1998) Early tracheostomy tube change
in children. Arch Otolaryngol Head Neck Surg
124(11):1237–1238
Deutsch ES (2010) Tracheostomy: pediatric consider-
ations. Respir Care 55(8):1082–1090
Doherty C, Neal R, English C, Cooke J, Atkinson D,
Bates L et al (2018) Multidisciplinary guidelines for
the management of paediatric tracheostomy emergen-
cies. Anaesthesia
Edwards JD, Houtrow AJ, Lucas AR, Miller RL, Keens
TG, Panitch HB et  al (2016) Children and young
adults who received tracheostomies or were initiated
on long-term ventilation in PICUs. Pediatr Crit Care
Med 17(8):e324–e334
Funamura JL, Yuen S, Kawai K, Gergin O, Adil E, Rahbar
R et al (2017) Characterizing mortality in pediatric tra-
cheostomy patients. Laryngoscope 127(7):1701–1706
Gergin O, Adil EA, Kawai K, Watters K, Moritz E, Rahbar
R (2016) Indications of pediatric tracheostomy over
the last 30 years: has anything changed? Int J Pediatr
Otorhinolaryngol 87:144–147
Ha TA, Goyal M, Ongkasuwan J (2017) Duration of
tracheostomy dependence and development of tra-
cheocutaneous fistula in children. Laryngoscope
127(12):2709–2712
Holloway AJ, Spaeder MC, Basu S (2015) Association of
timing of tracheostomy on clinical outcomes in PICU
patients. Pediatr Crit Care Med 16(3):e52–e58
Jackson C (1909) Tracheotomy. Laryngoscope
19(4):285–290
Kumar K, Miron C, Singh SI (2019) Maternal anesthesia
for EXIT procedure: a systematic review of literature.
J Anaesthesiol Clin Pharmacol 35(1):19–24
Lee JH, Koo CH, Lee SY, Kim EH, Song IK, Kim HS et al
(2016) Effect of early vs. late tracheostomy on clini-
cal outcomes in critically ill pediatric patients. Acta
Anaesthesiol Scand 60(9):1281–1288
Mahadevan M, Barber C, Salkeld L, Douglas G, Mills N
(2007) Pediatric tracheotomy: 17 year review. Int J
Pediatr Otorhinolaryngol 71(12):1829–1835
Mahida JB, Asti L, Boss EF, Shah RK, Deans KJ,
Minneci PC et  al (2016) Tracheostomy placement
in children younger than 2 years: 30-day outcomes
using the national surgical quality improvement pro-
gram pediatric. JAMA Otolaryngol Head Neck Surg
142(3):241–246
Norris MC, Joseph J, Leighton BL (1989) Anesthesia for
perinatal surgery. Am J Perinatol 6(1):39–40
Novoa RH, Quintana W, Castillo-Urquiaga W, Ventura W
(2020) EXIT (ex utero intrapartum treatment) surgery
for the management of fetal airway obstruction: a sys-
tematic review of the literature. J Pediatr Surg
Osborn AJ, de Alarcon A, Hart CK, Cotton RT, Rutter
MJ (2013) Tracheocutaneous fistula closure in the
pediatric population: should secondary closure be
the standard of care? Otolaryngol Head Neck Surg
149(5):766–771
30 Tracheostomy
Ozmen S, Ozmen OA, Unal OF (2009) Pediatric trache-
otomies: a 37-year experience in 282 children. Int J
Pediatr Otorhinolaryngol 73(7):959–961
Park JY, Suskind DL, Prater D, Muntz HR, Lusk RP
(1999) Maturation of the pediatric tracheostomy
stoma: effect on complications. Ann Otol Rhinol
Laryngol 108(12):1115–1119
Pizza A, Picconi E, Piastra M, Genovese O, Biasucci DG,
Conti G (2017) Early versus late tracheostomy in pedi-
atric intensive care unit: does it matter? A 6-year expe-
rience. Minerva Anestesiol 83(8):836–843
Principi T, Morrison GC, Matsui DM, Speechley KN,
Seabrook JA, Singh RN et al (2008) Elective trache-
ostomy in mechanically ventilated children in Canada.
Intensive Care Med 34(8):1498–1502
Strychowsky JE, Albert D, Chan K, Cheng A, Daniel
SJ, De Alarcon A et al (2016) International Pediatric
Otolaryngology Group (IPOG) consensus recommen-
dations: Routine peri-operative pediatric tracheotomy
care. Int J Pediatr Otorhinolaryngol 86:250–255
383
Tabaee A, Lando T, Rickert S, Stewart MG, Kuhel WI
(2007) Practice patterns, safety, and rationale for tra-
cheostomy tube changes: a survey of otolaryngology
training programs. Laryngoscope 117(4):573–576
Watters KF (2017) Tracheostomy in infants and children.
Respir Care 62(6):799–825
Wetmore RF, Marsh RR, Thompson ME, Tom LW (1999)
Pediatric tracheostomy: a changing procedure? Ann
Otol Rhinol Laryngol 108(7 Pt 1):695–699
Wood D, McShane P, Davis P (2012) Tracheostomy in
children admitted to paediatric intensive care. Arch
Dis Child 97(10):866–869
Woods R, Geyer L, Mehanna R, Russell J (2019) Pediatric
tracheostomy first tube change: when is it safe? Int J
Pediatr Otorhinolaryngol 120:78–81
Zhu H, Das P, Roberson DW, Jang J, Skinner ML, Paine
M et  al (2015) Hospitalizations in children with
preexisting tracheostomy: a national perspective.
Laryngoscope 125(2):462–468
Part IV
Chest
 Chest Wall Deformities
Robert E. Kelly and Donald Nuss
31.1 Introduction
Congenital chest wall deformities can be grouped
anatomically. Depression of the sternum and
adjacent costal cartilages, pectus excavatum, is
most common in North America, Europe, and
Asia. Protrusion of the sternum, pectus carina-
tum, is more common in South America. Sternal
clefts and Poland’s syndrome, which is a spec-
trum of problems involving unilateral absence of
chest wall structures, are much less common.
Any of these may be present at birth, but pectus
excavatum and carinatum may develop or worsen
at the time of the pubertal growth spurt. In the last
20 years or so, there has been a dramatic increase
in interest in these problems, with associated
innovation and evolution of care to ever less inva-
sive methods.
31.2 Pectus Excavatum
31.2.1 Description
Pectus excavatum is a condition in which the
sternum and anterior chest wall are abnormally
R. E. Kelly (*) · D. Nuss
Children’s Hospital of The King’s Daughters,
Norfolk, VA, USA
Eastern Virginia Medical School, Norfolk, VA, USA
e-mail: Robert.Kelly@chkd.org
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_31
31
close to the spine. Most commonly, there is a
cup-shaped, focal depression of the lower ster-
num. The depression can also occur as a broadly
based, shallow, saucer-shaped depression. In
some patients, the depression extends from the
xiphoid to the clavicles, like a trench. The depres-
sion can be centered on the middle of the sternum
or be quite asymmetrical.
31.2.2 History
As an inherited, externally visible deformity, the
condition has been noted in antiquity in Egyptian
art (Bialas et al. 2015), and has been identified in
CT scans of Egyptian mummies (Kwiencinski
2016). The condition was first described quite
clearly by Bauhinus (Bauhinus 1600) in 1595
and one of Leonardo da Vinci’s drawings shows
the condition (Ashrafian 2013).
However, little could be done for the condition
until thoracic surgery became possible in the
early twentieth century. Initial attempts at correc-
tion centered on resection of costal cartilages and
sternal elevation, which was accomplished
through external fixation (Sauerbruch 1931) in
Germany. The condition became treated more
routinely following work begun in the 1940s by
Welch (1958) in Boston and Ravitch (Ravitch
1949) in Baltimore, in which sections of the cos-
tal cartilages were resected and the sternum was
elevated. Subsequently, a substernal bar was used
387
388
for support. In 1998, Nuss and colleagues
reported elevation of the sternum utilizing a sub-
sternal bar without cartilage resection (Nuss et al.
1998), which has revolutionized treatment.
31.2.3 Incidence and etiology
Pectus excavatum occurs with variable inci-
dence in different parts of the world. Its fre-
quency of occurrence is of the order of 1  in
1000. The cause of pectus excavatum remains
unclear. At the p­ resent time, differential growth
of costal cartilages and ribs, such that the ster-
num is propelled inwardly by disproportionate
growth, is the theory with the best support (Park
et  al. 2015a, b). About 40% of patients give a
family history of pectus excavatum, and investi-
gation has shown that autosomal dominant,
autosomal recessive, and X-linked inheritance
occur in different kindreds (Horth et al. 2012). It
is unclear why the majority of patients do not
have a family history. In most reports, about
80% of patients are male. Whether these find-
ings represents genetic or epigenetic phenome-
nona remains unclear. Between 5% and 15% of
patients have Marfan syndrome; about 15% of
patients have associated scoliosis.
31.2.4 Pathophysiology
Patients with pectus excavatum report symptoms
in at least two-thirds of cases. The symptoms
reported most frequently are shortness of breath
with exertion, easy fatigability with cardiovascu-
lar exercise, and chest pain. Shortness of breath
does not occur at rest, but only with significant
cardiac and respiratory exertion. Chest pain tends
to occur within the depression itself, and is gen-
erally a short-lived, lancinating pain without
clear precipitating or relieving factors, and does
not persist long enough to cause the patient to
take medication to treat it; chest pain often occurs
at rest.
Investigations into the functional effects of the
anatomical deformity have shown clear evidence
of diminution of cardiac function with exercise,
R. E. Kelly and D. Nuss
and of limitation of pulmonary capacity. Forced
vital capacity and FEV1 are reduced, on average,
in studies containing large numbers of patients
(Kelly et al. 2016). This results in a shift of the
normal Gaussian distribution of PFT values to be
shifted to the left. However, many individuals
have normal pulmonary function evaluation. The
best explanation for this effect is decreased
motion of the chest wall at the site of the depres-
sion, resulting in a less efficient “bellows” action
during respiration; hence breathing is less effi-
cient, and vigorous exercise leads to symptoms
of shortness of breath and fatigability (Redlinger
Jr et al. 2012) (Fig. 31.1).
Cardiac evaluation has shown mitral valve
prolapse (Shamberger et  al. 1987) in approxi-
mately 14% of patients in the series in Norfolk
Virginia; and about half of patients’ mitral valve
prolapse is relieved after surgical correction.
Multiple well-done reports demonstrate improve-
ment of cardiac function following pectus repair
Normal Distribution of PFT’s
100
Distribution of FVC in Pectus Patients
35%
80
30%
25%
20%
15%
10%
5%
0%
30-39 40-49 50-59 60-69 70-79 80-89 90-99 100-
109
110-
119
120-
129
130-
139
140
Percent Predicted
Fig. 31.1  Forced vital capacity in pectus excavatum
patients. For full details, see Kelly et al. 2016
31  Chest Wall Deformities
(Chao et  al. 2018, Danish reports). This is best
explained by the relief of right ventricular com-
pression, and consequent increase in stroke
volume.
31.2.5 Body Image Effects
Body image effects are important and bring many
patients for treatment. In boys, it is common that
the patient is unwilling to take his shirt off at the
beach or to for sports, such as swimming. In
many girls, the condition produces uneven breast
projection, particularly when the deepest point of
the depression is well off the midline. This causes
both psychologic distress and practical difficulty
in purchasing swimsuits and bras. Carefully per-
formed studies in several parts of the world have
shown that these symptoms are relieved follow-
ing surgical correction (Kelly 2018).
31.2.6 Diagnosis and Differential
Diagnosis
Patients found on examination to have a mild
defect are referred for an exercise and posture
program, and followed at intervals of a year or
more. Patients with pectus excavatum, which is
significant on exam, are considered for inter-
vention. All patients are considered for nonsur-
gical treatment. For those in whom nonsurgical
treatment is unlikely to be successful, objective
evaluation by CT scan, PFT’s, EKG, and echo-
cardiogram is carried out preoperation.
It is our practice to consider the patient’s prob-
lem from four points of view: Firstly, the anat-
omy of the chest wall; secondly, the effect on the
lungs; thirdly, the effect on the heart, and finally,
the effects on body image and self-perception.
While other forms of imaging can replicate the
indices necessary to quantify the defect, a CT
scan of the chest provides clear cross-sectional
anatomy which makes the severity of the anatom-
ical deformity evident to the patient, family, other
doctors, and any others. Current technology
makes the radiation dose quite low, and CT
389
Fig. 31.2  Haller index is A divided by B; Correction
index is C-B divided by C
images are much superior to MRI in evaluating
the chest wall. CT scanning allows easy objective
measurement of the Haller index and correction
index, which allow comparison between patients
(see Fig. 31.2). A Haller index of greater than 3.2
is considered severe, and a correction index
greater than 10% is considered severe (St. Peter
et al. 2012; Sujka and St. Peter 2018).
Pulmonary function testing (PFT) identifies
restrictive disease: forced vital capacity less than
80% predicted, with a normal FEV1/FVC ratio
and minimal improvement in FEV1 with alb-
uterol. Restrictive disease by these criteria are
present in perhaps 20% of pectus excavatum
patients. Pulmonary function testing will also
identify patients whose symptoms of shortness of
breath or fatigability are due to asthma or allergy,
which occur at the same rate in pectus patients as
in the general population. Asthma and allergy are
not treated by operation, of course, but rather
medically. In those patients who will require
operation, and have asthma, it is helpful to opti-
mize asthma management before a procedure.
Cardiac evaluation by electrocardiogram will
disclose right bundle branch block, present in a
significant fraction of patients in our series, or
other rhythm concerns. Echocardiogram identi-
fies mitral valve prolapse, present in 10–15% (in
persons this age, the rate is approximately 1%)
(Kelly et al. 2020).
We evaluate the patient’s body image con-
cerns by asking the patient and family. Frequently,
teenage boys will deny any concern, but parents
390
will note an unwillingness to take off a shirt and
slouched posture to hide the defect. Often, body
image concerns turn out to be a major driver of
the patient’s presentation for evaluation.
We perform metal allergy testing in all
female surgical patients, who have a higher inci-
dence (up to 20%) of metal allergy than male
patients, and in male patients with a personal or
family history of metal allergy (Obermeyer
et al. 2018a, b).
31.3 Management
31.3.1 Vacuum Bell
Treatment with the vacuum bell is considered in
all patients, and discussed with the patient and
family. The vacuum bell was developed in
Germany by an engineer, Eckardt Klobe. Its effi-
cacy has been evaluated in several centers. In
older patients and in those with the deeper defor-
mity, the vacuum bell has been much less suc-
cessful. However, the vacuum bell has an
extremely low complication rate, and hence it is
morally acceptable to use even if success is
unlikely in a given patient who wishes to use it.
However, the time, trouble, inconvenience, and
expense of treatment should be weighed against
the likelihood of success. We follow the protocol
developed by Haecker in Basel, Switzerland. The
Fig. 31.3  Vacuum bell in use
R. E. Kelly and D. Nuss
vacuum bell is applied initially for 20 min twice
a day and the duration of application is increased
to 2 h twice daily over a period of 10 weeks. The
amount of vacuum is increased by squeezing a
bulb like that on a sphygmomanometer, and is
increased steadily over the same interval. Patients
are followed at 3- to 6-month intervals. At our
hospital, good or excellent results were achieved
in only 37% of patients, but patients who were
11 years old or younger who had a depth of defect
of 1.5 cm or less, and had a flexible chest wall,
were three times as likely as others to achieve a
good or excellent result (Obermeyer et al. 2018a,
b) (Fig. 31.3 and Table 31.1).
Table 31.1 Experience with the Nuss procedure at
Children’s Hospital of the King’s Daughters, in Norfolk,
Virginia, USA, where the procedure was developed by Dr.
Donald Nuss
Minimally Invasive Pectus Excavatum Repair: Clinical
Experience as of 1987–2018
  • 3955 patients have been evaluated for P.E.
  • 2286 patients have had pectus repair.a
  • 1909 patients have had bar removal.
  • 2165 patients have had primary operation.
  • 1772 patients have had primary bar removal.
  • 131 patients have had re-do operations:
       ➢ 71 failed Nuss procedure
       ➢ 60 failed Ravitch procedure
Data collected through 12/31/2015
a
Patients with more than one pectusrepair at CHKD now
counted only once
31  Chest Wall Deformities
31.4 Nuss Procedure: Technique
(Fig. 31.4)
In patients who fail vacuum bell treatment, who
are likely to fail based on our data, or who desire
surgical treatment after discussion of the options,
surgical treatment is offered. For surgical treat-
ment, we favor the procedure developed at our
hospital by Dr. Donald Nuss. In this procedure, a
steel bar is placed beneath the patient’s sternum
and used to elevate the chest wall. The operation
is undertaken under general anesthesia, and the
technical aspects are quite important for success-
ful outcome. The reader is strongly advised to
consult our publications from the home of the
Nuss procedure, detailing surgical technique,
written in close communication with the retired
Dr. Nuss (Kelly 2018). Briefly, the patient is posi-
tioned on the operating table on a warmed gel
pad with all appropriate anesthetic monitoring
equipment. General endotracheal anesthesia is
induced. The chest is prepared with ChloraPrep
and the patient is administered intravenous
cefazolin. A 5  mm thoracoscopic introducer is
passed into the right hemithorax in approximately
the right sixth intercostal space in the anterior
axillary line. After confirming intra-thoracic
placement of the scope, the chest is insufflated
with CO2 to a pressure of 8 mmHg. With the tho-
racoscope, the configuration of the depression
Metal bar
Fig. 31.4  Diagram of Nuss procedure
391
and its relation to the heart are inspected. A fine
needle is inserted to the right of the xiphoid or to
the right of the sternum under vision with the
scope. A Rultract crane is attached to the operat-
ing table on the left side at the level of the
patient’s umbilicus. Under vision with the scope,
at the site demonstrated to be safe by the needle,
a 3 mm incision is made. Under vision with the
scope, a hook or L-shaped sternal elevator is
passed into the right hemithorax, swiveled
around, and applied to the lower body of the ster-
num. The sternum is then elevated with the winch
of the Rultract. A Lewin spine forceps can be
applied to the sternum for traction instead of the
Uemura or L-lifter, or the chest can be elevated
with a bone hook introduced adjacent to the
xiphoid and elevation carried out by an assistant.
The vacuum bell can also be used. Sternal eleva-
tion improves visualization and puts the soft tis-
sue between the pericardium and the Sternum
under stretch, facilitating dissection. In males,
transverse lateral thoracic incisions are made,
each about 2–3 cm long (see Fig. 31.5), at a site
corresponding to the deepest point of the pectus.
In female patients, the incision is placed where
the breast joins to the chest wall (Figs. 31.5, 31.6,
and 31.7).
Subcutaneous tunneling is carried out and a
Lorenz introducer, designed for this purpose is
brought through the intercostal space selected for
introduction of the bar into the chest (Fig. 31.7).
The site of entry to the chest needs to be medial
to the pectus ridge (the most anterior part of the
chest wall on either side). Then, with the sternum
elevated, a pawing motion is used beginning at
the sternum and proceeding back towards the
spine to make a plane in the soft tissue between
the sternum and the spine.
When this is carried out in the correct plane, a
foamy appearance of the soft tissue is seen and
there is no bleeding (Fig. 31.8).
The introducer is brought under vision with
the scope from the right hemithorax into the left
hemithorax. The 30 degree scope is swiveled
around to be sure that the internal mammary ves-
sel or the lung is not at the site chosen for bring-
ing the introducer out of the chest wall on the left.
After the introducer is pushed through the
392
Fig. 31.7  Small, large, and extra large introducers
­utilizing the patient’s CT scan. The bar is tied to
Fig. 31.5  Showing location of camera, incision site just
cephalad, view of hook inside chest, and chest wall ele-
vated, facilitating vision with cardiac compression evident
on preop CT scan
Fig. 31.6  Location of skin incision in female patient
selected intercostal space, it is brought out the
left lateral thoracic incision. An umbilical tape is
tied to the end of the introducer which is guided
under vision with the thoracoscope from the left
to the right.
A steel bar is bent to conform to that patient’s
chest shape. This can be done at the operating
table, or it can be bent at the factory by the manu-
R. E. Kelly and D. Nuss
facturer. In metal allergy, a titanium bar is used;
this must be pre-shaped at the manufacturer
the end of the umbilical tape and guided from the
right to the left under vision with the scope. The
bar is turned over with instruments designed for
the purpose. The Rultract Crane is released and
inspection of the chest determines whether a sec-
ond bar will be necessary. If a second bar is nec-
essary, it is placed one interspace cephalad or
caudad to the first bar. In female patients, the
incision is extended; in male patients a second
incision may be necessary, or it may be possible
to introduce the second bar through the first inci-
sion (Fig. 31.9).
The bar is stabilized with a tongue-in-groove
stabilizer on the patient’s left side held in place
with figure-of-eight FiberWire or sternal wire
(Fig. 31.10).
Next, the bar is secured to the underlying rib
by wrapping it with multiple thicknesses of #1
polydioxanone (PDS) suture (Fig.  31.11). The
ligature is introduced under vision with the scope
with an Endo Close needle and grasped with an
orthopedic tendon passer from the other side of
the rib. This is accomplished at at least two dif-
ferent rib crossings. Then, through each hole and
the bar in each hole in the stabilizer, heavy 0
Vicryl was used to secure the bar to the underly-
ing musculature. At the conclusion of this pro-
cess, it should not be possible to move the bar
with one’s hands. The Rultract crane is let down
under vision with the scope and the hook or
L-shaped lifter is taken out under vision with the
31  Chest Wall Deformities 393

Fig. 31.8  A posteriorly directed pawing motion allows introduction into an avascular plane for passing in front of the
pericardium

Fig. 31.9  Correct configuration of the bar: neither a

Fig. 31.10  Tongue in groove stabilizer, held in place
semicircle nor a table top with vertical edges
with figure of eight wrap of FiberWire or sternal wire

Fig. 31.11  The bar is

wrapped to the
underlying ribs with
Pericostal Suture
multiple thicknesses of
heavy PDS suture to
prevent rotation of the
bar

Fixation to TWO separate ribs

Decreases degrees of rotational freedom
394 R. E. Kelly and D. Nuss

a b

Fig. 31.12 (a) Patient positioned on the operating table.
Arms abducted about 75 degrees at the shoulder, elbows
slightly flexed, rolled up foam beneath the wrists. Betadine
in the depression demonstrates its position and symmetry.
The intercostal spaces are marked, and the deepest part of
the depression is marked with a large dot. Lateral inci-
sions will be placed to allow bar to support the area
marked with the dot. (b) Postop after correction, showing
position of lateral incisions

scope to ensure there is no bleeding from that

site. The chest is reinspected with the thoraco-
scope to be sure there is no bleeding from any site
of perforation of the chest wall. The wounds are
closed in layers with absorbable suture. Then, the
insufflation tubing is cut and placed to underwa-
ter seal. The anesthesiologist repeatedly inflates
the lungs with positive pressure. When the bub-
bling has stopped, indicating that all the insuf-
flated CO2 has been evacuated, positive pressure
is held in the thoracoscopic introducer is removed
Fig. 31.13 Flattening the bar with Zimmer Biomet
while the up reposition fibular of 8 muscle pillar
instruments designed for the purpose (Zimmer Biomet,
suture is tied. Usually, no chest tube is necessary. Inc., Warsaw, Indiana). A variety of other instruments are
A chest x-ray is exposed on the operating table in available and can be used as well
both AP and lateral views. The patient is then
awakened, extubated, and taken to the recovery
area where a gentle emergence from anesthesia is
allowed. Postoperative pain relief can be obtained
by a number of methods (Fig. 31.12). Excellent
results have been reported by centers around the
world with this method (Kelly and Daniel, 2018).
Bars are removed 2–3  years after insertion.
Both incisions are opened, and any bony over-
growth around the bar is removed with osteotome
and rongeurs. The bar is flattened with instruments
designed for the purpose, and removed with trac-
tion along the length of the bar. With this mature
Fig. 31.14  After the bar has been flattened at both ends
technique, we have been able to achieve good and all bony overgrowth removed with osteotome and
results and few complications in a large number of rongeurs, the bar can be removed with lateral traction via
patients (Kelly et al. 2020) (Figs. 31.13 and 31.14). a bone hook
31  Chest Wall Deformities 395

31.4.1 Open Operation cles which have been divided and reattached ini-
tially. A multicenter study of pectus excavatum
An open operation for pectus excavatum is showed similar outcomes and similar complica-
required for some patients with carinatum on tion rates for the 2 procedures.
one side of the chest, and excavatum on the
other. Also, it is required for chondromanubrial
carinatum, which has an upper protrusion and 31.4.2 Pectus Carinatum
lower depression; often such patients are referred
as pectus excavatum. An open operation is car- 31.4.2.1 Description
ried out by a method that resects as little of the The name pectus carinatum, coined years ago,
costal cartilages as possible. The patient is posi- suggests the similarity of the sternal protrusion to
tioned supine on the operating table with arms the protrusion of the keel of a boat (Fig. 31.15).
tucked at the sides. A transverse inframammary Pectus carinatum occurs in two major mor-
or, occasionally, a vertical midline incision phologies: upper, or chondromanubrial carina-
approximately 6 or 7 cm long is made. The skin tum, is also referred to as pectus arcuatum,
and subcutaneous tissue are elevated off the pec- Currarino-Silverman deformity, and horns of
toralis major. The pectoralis major is detached steer deformity. It results from angulation and
from its origins. The rectus is detached inferi- fusion of the manubrium and body of the sternum
orly. At each costal cartilage, incision is made in or of sternum bradycardia inferior to the sterno-
the perichondrium, which is elevated. A small manubrial joint. Lower, or chondrogladiolar pec-
section, usually between 1 and 3 cm, is removed. tus carinatum is far more common, and results
Only those cartilages necessary to allow a desir- from anterior angulation of the lower portion of a
able position of the sternum are resected. The straight sternum and lower costal cartilages
sternum is elevated anteriorly. We generally (Fig. 31.16).
place a steel or titanium bar posterior to the ster-
num for additional support, though this is not
done at all centers. Fracture of the anterior ster-
nal table may be required, especially if the ster-
num itself is curved, and is accomplished either
with an osteotome or reciprocating saw. Titanium
plates may be used to stabilize the sternum after
sternal resection. The substernal bar is secured to
the ribs with heavy PDS. The sternum and rectus
are then reattached utilizing heavy absorbable
suture after closing the perichondrium at each
cartilage. A closed suction drain is placed poste-
rior to the sternum and anterior to the sternum.
The patient is awakened, extubated and recov-
ered in the same gentle manner as for Nuss pro-
cedure. Postoperative care is similar, though Fig. 31.15  Chondrogladiolar (lower) pectus carinatum,
patients should not sit up using the rectus mus- the most common morphology
396
Fig. 31.16 Chondromanubrial pectus carinatum, or
Currarino Silverman syndrome, photos and CT scan. This
variant is much less frequently seen. CT scan shows
31.5 History
31.5.1 Incidence and Etiology
Pectus carinatum is most frequently reported to
be about 1/10 as common as pectus excavatum in
the United States. However, with the advent of
nonsurgical treatment for pectus carinatum, it
R. E. Kelly and D. Nuss
fusion of the sterno-manubrial joint, short sternum, and
angulation of the sternum. The angulation can also occur
between the sternebrae of the body of the sternum
may soon be found that the incidence is higher. In
other places, the incidence may be different: in a
17-year report describing more than 7800 patients
from Buenos Aires, Argentina, 55% had pectus
carinatum. The cause of pectus carinatum is not
clear. There is a similar frequency of family his-
tory to patients with pectus excavatum at about
40%. A small percentage of patients have associ-
31  Chest Wall Deformities
ated Marfan syndrome or Ehlers-Danlos syn-
drome. Scoliosis accompanies a small fraction of
patients.
31.5.2 Clinical Features
31.5.2.1 Symptoms
Symptoms associated with pectus carinatum
occur in a surprising fraction of patients when
they are sought. Symptoms generally occur with
exercise, and are similar to those of excavatum
patients: fatigability and chest pain. The
­explanation for fatigability may lie with ineffi-
ciency of the respiratory Bellows action, but this
remains to be seen. Chest pain particularly occurs
on compression, for example when the patient
lies prone. In our recent series, 46% of patients
had some symptom on questioning. This is
important, because typically, pediatricians do not
ask, and hence believe this to be asymptomatic.
31.5.2.2 Cardiac and Pulmonary
Effects
Cardiac and pulmonary effects of pectus carina-
tum have not been intensively studied. There is
no loss of intrathoracic volume, and there is no
cardiac compression. Effects of the condition on
the respiratory bellows have not been intensively
studied. Some authors have suggested that flat-
tening of the diaphragm, similar to what happens
in the barrel chest of emphysema, might be
responsible for symptoms of shortness of breath.
31.5.2.3 Body Image Effects
The body image effects of pectus carinatum are
every bit as powerful as those of pectus carina-
tum. Patients typically adopt a slouched posture
to minimize the protrusion of the chest wall.
Investigations have disclosed deficits in body
image and self-esteem which have been improved
on correction (ref Paulson).
31.5.3 Evaluation
Evaluation of pectus carinatum patients should
include evaluation of the anatomy by chest x-ray
397
and, if indicated by symptoms or chest x-ray
findings, three-dimensional imaging such as CT
scan or MRI.  Pulmonary function testing and
echocardiogram should be indicated as clinically
necessary. In patients with a habitus consistent
with Marfan, echocardiogram should be per-
formed to evaluate the aortic root.
31.5.4 Treatment
As with pectus excavatum, pectus carinatum can
be treated surgically or nonsurgically. We favor a
staged approach to management, with initial
brace treatment, and surgical care for those who
fail or can be expected to fail brace treatment.
The surgical options include the Abramson pro-
cedure, labeled in Germany as the reverse Nuss
procedure, in which a metal bar is used to pull the
sternum posteriorly. The open operation is also
very effective. The sandwich procedure intro-
duced by Park is another option in countries
where the implants are available.
31.5.5 External Brace Treatment
External brace treatment for pectus carinatum is
based on the principles of Nicholas Andrey who is
considered the father of orthopedics, and the
effects of this treatment can be explained by Julius
Wolf’s law of bone remodeling. Force applied
therapeutically to deformed bones and cartilage
can produce a gradual remodeling, which is cor-
rective and beneficial. In 1979, Haje of Brasilia,
Brazil published the first report on successful
treatment of pectus carinatum with an orthotic
referred to as a dynamic chest compressor. When
this approach was introduced elsewhere, nonop-
erative management gained adherents in different
parts of the world. By the year 2000, in the United
States, Canada, South Korea, South America as
well as in Europe, the treatment was well-estab-
lished (refs). In Norfolk, Virginia, we have uti-
lized the pressure-controlled brace approach
introduced by Fraire and Martinez-Ferro
(Martinez-Ferro 2008) since 2009. In this
approach, the amount of force applied to the chest
398
Docking
Holes
Fig. 31.17  Argentine brace for pectus carinatum
can be measured and adjusted to keep the brace
comfortable. The Argentine brace is easy to con-
ceal, which has improved adherence to brace
treatment compared to prior braces. Patients are
asked to wear the brace at least 8 h daily, but as
many hours as possible, removing the brace only
for sports and bathing. The brace is fitted by phys-
ical therapists, who also instruct patients in use of
exercises to develop posture and the chest mus-
cles. Patients are followed at 3–6 month intervals,
and treatment lasts 1–2 years. We began using this
brace in 2009, and have reported our favorable
results with its use (Kelly et al. 2021). In the last
10  years to 2018, we have treated 370 patients
with a brace. Although almost 1/3 of patients
dropped out of treatment, the brace was success-
ful in a majority of those who completed treat-
ment; that has been the experience in other centers
around the world as well (Fig. 31.17).
31.5.6 Reverse Nuss Procedure
of Abramson
Horatio Abramson of Buenos Aires realized that
if the sternum could be pushed out to treat pectus
excavatum, it could be pulled back posteriorly for
pectus carinatum. In his operation, a steel bar is
bent to fit the desired configuration of the chest
(Abramson et  al. 2009). Stabilizers similar to
those used to prevent the rotation of an excava-
tum bar are used to wrap wire or cable around the
rib subpericostally, and the compressive bar is
R. E. Kelly and D. Nuss
slid into these tongue-in-groove stabilizers and
held in place with a screw. This operation has
been performed in several centers around the
world, with at least 2 centers reporting more than
100 cases. The operation is clearly efficacious.
However, at the present time there are a signifi-
cant number of equipment related difficulties, at
least in the United States. We have employed this
operation 23 times since introducing it in 2008.
While clearly effective, we have required revi-
sion operations in 1 patient (46%) and early bar
removal in 6 (26%) for wire or cable breaks, or
when the cable cuts into the ribs (Fig. 31.18).
31.5.7 Sandwich Technique
Park, of Seoul, South Korea, utilizes a technique
for carinatum and mixed excavatum/carinatum
side to side (not pectus arcuatum), which utilizes
one bar anterior to the sternum and one bar poste-
riorly, which are linked together to compress the
chest wall between them. In a series of more than
50 patients, he reports good success and few
complications (Park and Kim 2016) (Fig. 31.19).
31.5.8 Open Operation
The open operation for pectus carinatum should
be employed in all patients with the chondroma-
nubrial (also called Currarino-Silverman, pectus
arcuatum, or horns of steer) variant (Currarino
and Silverman, 1958). The stiff, angulated ster-
num will not respond to external compression by
brace. The open operation is similar to that
employed for excavatum except, of course the
sternum needs to be brought posteriorly. Often,
wedge resection of a portion of the sternum is
necessary. It may be advantageous to use sternal
plates and screws to stabilize the sternum follow-
ing wedge osteotomy. In pectus arcuatum, both
stabilization of the upper sternum, at the site of
the wedge osteotomy, and placement of a bar
under the lower sternum may be necessary. For
chondrogladiolar or lower carinatum, it may not
be necessary to provide stabilization other than by
use of the perichondrium (Figs. 31.20 and 31.21).
31  Chest Wall Deformities 399

a b

c d

Fig. 31.18  Abramson operation for pectus carinatum. (a) Diagram of procedure. (b) Patient preop. (c) Marking the
patient, locating the bar at the point of greatest prominence, and locating the stabilizers laterally. (d) Postoperation

Fig. 31.19 Sandwich
technique: one bar is
inside the chest, and one
bar externally;
compressing them
together corrects the
deformity
400 R. E. Kelly and D. Nuss

a b

c d

Fig. 31.20  Chondromanubrial pectus carinatum (Curra- ing perichondrium. (d) Costal cartilage resection. (e) One
rino-Silverman syndrome). (a) Preop, showing upper ster- type of plate and screw apparatus to stabilize sternal oste-
nal protrusion with lower sternal depression puddling otomy (SternaLock Blu, Zimmer Biomet, Warsaw, Indi-
Betadine. (b) Elevating pectoralis major muscle. (c) Incis- ana, USA)
31  Chest Wall Deformities 401

a b

c d

e f

Fig. 31.21  Chondrogladiolar pectus carinatum managed (d) Transverse sternal osteotomy. (e) Immediately postop.
by open operation. (a) Preop. (b) Preop CT scan, with (f) 2 months postop
Haller index <2.0 (=1.5). (c) Resected costal cartilages.

31.6 Uncommon Chest Wall
Conditions
31.6.1 Poland’s Syndrome
Poland’s syndrome was described by an English
medical student, Alfred Poland, in 1841 (Poland
1841), though many features had previously been
described in France and Germany. The syndrome
has several features, and few patients express all
the anomalies; about 70% are male. Absence of
the pectoralis major, pectoralis minor, and/or ser-
ratus anterior are commonly thought of as central
to the definition. In addition, athelia or amastia,
nipple deformities, and upper extremity hand
abnormalities including syndactyly are also
402
Fig. 31.22  Poland’s syndrome, showing absence of pectoral muscles on left
found. The patient’s right side is more commonly
affected. Hypoplasia or aplasia of the chest wall
structures can occur; and pectus excavatum or
carinatum. Surgical correction is reserved for
functional chest wall depression or aplasia, and is
seldom required. Autologous rib grafts have been
used for correction of aplasia with success
(Seyfer et  al. 1988; Shamberger et  al. 1989;
Shamberger 1998) (Fig. 31.22).
31.6.2 Sternal Cleft
Cleft sternum is a rare malformation (<1% of all
chest wall malformations). It is due to failure of
sternal fusion in early embryonic development.
Sternal clefts can be classified as complete, supe-
rior, or inferior (Samarrai et al. 1985).
Superior clefts are either U-shaped or
V-shaped, and are most frequently isolated anom-
alies. The heart is normal, and normally located.
Surgical correction should be organized elec-
tively soon after birth when possible. In the neo-
natal period, the sternal edges can be approximated
more easily because of the flexibility of the chest.
After 1  year of age, primary repair is difficult,
and more extensive operation with use of pros-
thetic materials or autologous rib or cartilage
grafts (Knox et al. 1994).
R. E. Kelly and D. Nuss
At CHKD, we have treated a neonate with ster-
nal cleft who developed pectus excavatum subse-
quently. The Nuss procedure was performed at age
7 years successfully (Cunningham et al. 2017).
Complete and inferior clefts can be associated
with ectopia cordis or omphalocele. These com-
plex patients should be managed at centers with
experience and expertise in these problems
(Groner 2003; Engum 2008).
31.6.3 Jeune’s Syndrome
Jeune’s syndrome, a congenital condition, and
acquired Jeune’s syndrome are conditions in
which the chest cage is too small (Jeune et  al.
1954; Haller et al. 1996). Various operations have
been tried to correct the inadequate chest cage,
but most are not successful. In particular, the
acquired Jeune’s syndrome which has followed
too early and too extensive open operations for
pectus excavatum, is not successfully corrected
by the Nuss procedure, and requires an open
approach. There is hope that with biologic
implants both the skeleton and soft tissues of the
chest cage can be enlarged, but at present the
condition is difficult to treat except symptomati-
cally (Campbell et  al. 2003; Phillips and van
Aalst 2008).
31  Chest Wall Deformities 403

31.7 Conclusion Jeune M, Carron R, Beraud C et  al (1954)

Polychondrodystrophie avec blocage thoracique
d’evolution fatale. Pediatrie 9:390–392
The most common chest wall deformities, pectus Kelly RE (2018) Modification and further development of
excavatum and pectus carinatum, can be success- the original Nuss procedure: blessing or curse? Eur J
fully treated by a variety of methods. Evaluation Pediatr Surg 28:304–319
Kelly RE, Daniel A (2018) Outcomes, quality of life, and
and management can proceed in a stepwise fash- long-term results after pectus repair from around the
ion, utilizing radiography and functional testing. globe. Semin Pediatr Surg 27:170–174
Children with rare chest wall deformities should Kelly RE, Obermeyer RJ, Nuss D (2016) Diminished
be thoroughly evaluated and managed by teams pulmonary function in pectus excavatum: from
denying the problem to finding the mechanism. Ann
with expertise in their care. Cardiothorac Surg 5:466–475
Kelly RE, Obermeyer RJ, Goretsky MJ et  al (2020)
Recent modification of the Nuss procedure: the pur-
suit of safety during the minimally invasive repair of
References pectus excavatum. Ann Surg. https://doi.org/10.1097/
SLA.0000000000003877
Abramson H, D’Agostino JD, Wuscovi S (2009) A 5-year Kelly RE, Obermeyer RJ, Goretsky MJ et al (2021) Ten-
experience with a minimally invasive technique for year experience with staged management of pectus
pectus carinatum repair. J Pediatr Surg 44:118–124 carinatum: Results and lessons learned. J Pediatr Surg
Ashrafian H (2013) Leonardo de Vinci and the first por- 56:1835−1840.
trayal of pectus excavatum. Thorax 68:1031 Knox L, Tuggle D, Knott-Craig CJ (1994) Repair of con-
Bauhinus J (1600) Observationum Medicariam. Liber II, genital sternal clefts in adolescence and infancy. J
Observ 264, Francfurti 1600.507. Pediatr Surg 29:1513–1516
Bialas AJ, Kaczmarski J, Kozak J, Kempinska-­ Kwiencinski J (2016) Pectus excavatum in mummies
Miroslawska B (2015) Pectus excavatum in relief from from ancient Egypt. J Interact Cardiovasc Thorac Surg
Ancient Egypt (dating back to circa 2400 BC). Interact 23:993–995
CardioVasc Thorac Surg 20:556–557 Martinez-Ferro M, Fraire C, Bernard S (2008) Dynamic
Campbell RM, Smith MD, Mayes T et al (2003) The char- compression system for the correction of pectus cari-
acteristics of thoracic insufficiency syndrome associ- natum. Semin Pediatr Surg 17:194–200
ated with fused ribs and congenital scoliosis. J Bone Nuss D, Kelly RE Jr, Croitoru DP et al (1998) A 10-year
Joint Surg 85:399–408 review of a minimally invasive technique for the
Chao CJ, Jaroszewski D, Gotway M et al (2018) Effects correction of pectus excavatum. J Pediatr Surg
of pectus excavatum repair on right and left ventricular 33:545–552
strain. Ann Thorac Surg 105:294–301 Obermeyer RJ, Gaffar S, Kelly RE et al (2018a) Selective
Cunningham ME, Nuss D, Kelly RE (2017) Successful versus routine patch metal allergy testing to select bar
correction of pectus excavatum using the Nuss material for the Nuss procedure in 932 patients over 10
procedure after neonatal sternal cleft repair. J years. J Pediatr Surg 53:260–264
Pediatr Surg Case Rep. https://doi.org/10.1016/ Obermeyer RJ, Cohen NS, Kelly RE et  al (2018b)
jepsc.2017.08.009 Nonoperative management of pectus excavatum with
Currarino G, Silverman FN (1958) Premature oblitera- vacuum bell therapy: a single center study. J Pediatr
tion of the sternal sutures and pigeon breast deformity. Surg 53:1221–1225
Radiology 70:532–540 Park HJ, Kim KS (2016) The sandwich technique for
Engum SA (2008) Embryology, sternal clefts, ectopia repair of pectus carinatum and excavatum/carina-
cordis, and Cantrell’s pentalogy. Semin Pediatr Surg tum complex. Ann Cardiothorac Surg 5(5):434–439.
17:154–160 https://doi.org/10.21037/acs.2016.08.04
Groner JI (2003) Ectopia cordis and sternal defects. Park CH, Kim TH, Haam SJ et al (2015a) Rib overgrowth
In: Zeigler MM, Azizkhan RG, Weber TR (eds) may be a contributing factor for pectus excavatum:
Operative pediatric surgery. McGraw-Hill, New York, evaluation of prepubertal patients younger than 10
pp 279–293 years old. J Pediatr Surg 50:1945–1948
Haller JA, Colombani PM, Humphries CT et  al (1996) Park CH, Kim TH, Haam SJ et  al (2015b) Asymmetric
Chest wall constriction after too extensive and too pectus excavatum is associated with overgrowth of
early operations for pectus excavatum. Ann Thorac ribs rather than cartilage. Thorac Cardiovasc Surg
Surg 61:1618–1625 5:427–432
Horth L, Stacey M, Kelly RE Jr et al (2012) Advancing Phillips JD, van Aalst JA (2008) Jeune’s syndrome
our understanding of the inheritance and transmission (asphyxiating thoracic dystrophy) congenital and
of pectus excavatum – inheritance of pectus exavatum. acquired. Semin Pediatr Surg 17:167–172
J Pediatr Genetics 2012:161–173
404
Poland A (1841) Deficiency of the pectoralis muscles.
Guys Hosp Rep 6:191–193
Ravitch MM (1949) The operative treatment of pectus
excavatum. Ann Surg 129:429–444
Redlinger RE Jr, Wootton A, Kelly RE et  al (2012)
Optoelectronic plethysmography demonstrates abro-
gation of regional chest wall motion dysfunction in
patients with pectus excavatum after Nuss repair. J
Pediatr Surg 47:160–164
Samarrai AR, Charmockley HA, Attr AA (1985) Complete
cleft sternum: classification and surgical repair. Int
Surg 70:71–73
Sauerbruch F (1931) Operative Beseitigung der
Angeborenen Trichterbrust. Dtsch Z Chir 234:760
Seyfer AE, Icochea R, Graber GM (1988) Poland’s anom-
aly: natural history and long-term results of chest wall
reconstruction in 33 patients. Ann Surg 208:776–782
Shamberger RC (1998) Congenital chest wall deformi-
ties. In: Grosfeld JL, O’Neill JA Jr, Fonkalsrud EW,
R. E. Kelly and D. Nuss
Coran AG (eds) Pediatric surgery, 5th edn. Elsevier,
Philadelphia, pp 787–817
Shamberger RC, Welch KJ, Sanders SP (1987) Mitral
valve prolapse associated with pectus excavatum. J
Pediatr 111:404–407
Shamberger RC, Welch KJ, Upton J III (1989) Surgical
treatment of thoracic deformity in Poland’s syndrome.
J Pediatr Surg 24:760–765
St Peter SD, Weesner KA, Weissend EE et al (2012)
Epidural vs. patient-controlled analgesia for postoper-
ative pain after pectus excavatum repair: a prospective,
randomized trial. J Pediatr Surg 47:148–153
Sujka JA, St. Peter SD (2018) Quantification of pectus
excavatum: anatomic indices. Semin Pediatr Surg
27:122–126
Welch KJ (1958) Satisfactory surgical correction of pec-
tus excavatum deformity in childhood. J Thorac Surg
36:697–713
 Breast Disorders in Children
and Adolescents
Steffi Mayer, Jan-Hendrik Gosemann,
Benno M. Ure, and Martin L. Metzelder
32.1 Introduction
During the fourth to fifth week of gestation, the
ectodermal milk line or mammary ridges emerge
bilaterally on both sides of the ventral body wall
from the axilla to the groin. Subsequently, the
proximal and distal parts of the milk lines disap-
pear, while the pectoral mammary anlage com-
prises the mammary primordium. Epithelial
tissue forms buds that divide into primitive breast
alveoli. In the last trimester, the primitive mam-
mary ducts are formed by canalization of the epi-
thelial branches. Finally, breast parenchyma
differentiates into lobules and the fibrous stroma
as well as fat tissue develop. The nipple-alveolar
complex (NAC) forms late in gestation, and a
pigmented areola is first seen at 20–24 weeks.
At birth, breast enlargement occurs in both
genders due to relatively high levels of placental
estrogens in the perinatal period, followed  by a
surge in hypothalamic-pituitary-gonadal axis
activity causing a mini puberty within the first
S. Mayer (*) · J.-H. Gosemann
Department of Pediatric Surgery, University of
Leipzig, Leipzig, Germany
e-mail: steffi.mayer@medizin.uni-leipzig.de
B. M. Ure
Hannover Medical School, Hannover, Germany
M. L. Metzelder
Department of Pediatric Surgery, University Clinic of
Surgery, Medical University of Vienna, Vienna,
Austria
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_32
32
months of life. When estrogen levels drop after
birth  and prolactin and oxytocin secretion  is
induced, mammary tissue may produce colos-
trum, the so-called witch’s milk.
With puberty, ovarian follicles produce estro-
gens that induce ductal development and site-­
specific adipose deposition in the breast. The
initiation of breast growth and maturation (the-
larche) starts at the age of 8–13  years, 2  years
before the menarche. Thelarche usually starts
unilaterally with a firm disklike area and is not an
indication for a biopsy. Breast development takes
place as described by Tanner in five stages.
The adult breast is typically located between
the second and the sixth rib, as well as the sternal
edge and the midaxillary line fixed between the
superficial and deep pectoral fascia. The nipple is
found over the fourth intercostal space with an
oval-shaped NAC in 91% of a mean diameter of
2.3 cm and a 2.7–2.0 horizontal-to-vertical-width
ratio. The internal mammary and lateral thoracic
vessels mainly supply the breast tissue. Sensory
innervation comes from T3 to T5 nerve roots.
Diseases of the breast vary widely in infants
and adolescence. Some degree of breast asymme-
try is normal, with the left breast commonly
being larger than the right one. But even this
nature-given asymmetry may hamper the
child’s well-being bearing the wish for surgical
correction. Breast disorders include congenital
anomalies, infections, trauma, endocrinological
disorders, as well as benign and malignant
405
406 S. Mayer et al.

tumors. Moreover, surgical interventions such as
biopsies of the prepubertal breast, thoracotomy,
chest tube as well as central venous catheter
placement, or abscess drainage, may irreversibly
cause iatrogenic  disruption of breast develop-
ment. Also, burn injuries can impede breast
development profoundly.
symmetry. It may be accomplished with tissue
expansion and subsequent implant placement
with respect to possible blood supply aberran-
cies. For asymmetric breasts, usually bilateral
procedures with a reduction of the larger and
enlargement of the smaller breast are performed.

32.2.2 Breast Atrophy

32.2 Congenital
and Developmental Breast atrophy occurs in normally configurated
Anomalies breasts. Bilateral atrophy can be the result of eat-
ing disorders or hormonal imbalances with high
32.2.1 Hypoplastic Anomalies androgens and low estrogen levels. In particular,
if bilateral atrophy is accompanied by abnormal
Hypoplastic anomalies arise from an abnormal pubertal development, an endocrinological eval-
involution of the mammary ridge with varying uation with regard to ovarian dysfunction, hypo-
presentation. Athelia describes the absence of the thyroidism, or androgen-producing tumors is
NAC, amazia is the absence of breast tissue, and required. Unilateral atrophy can be the result of
amastia (aplasia) is the deficiency of breast tissue trauma or surgical procedures such as chest tube
and NAC. Hypoplastic anomalies are rare and are as well as central venous catheter placement,
often associated with other anomalies and syn- biopsies, or abscess drainage as well as after
dromes including Poland’s syndrome, choanal-­ radiotherapy, severe burns, mononucleosis and in
atresia-­
athelia syndrome, scalp-ear-nipple scleroderma. The atrophy occurs by scarring,
syndrome, Al-Awadi/Rass-Rothschild syndrome, tethering of the breast to the chest wall with con-
Mayer-Rokitansky-Kuster-Hauser syndrome, tour deformity and restricted growth, or direct
dermoid cysts, anomalies of the plate and upper damage to the breast bud, preferentially in pre-
extremities, as well as ectodermal dysplasia. mature babies, in whom it may be difficult to
Poland’s syndrome occurs in 1:20,000–30,000 identify the NAC. Thus, thoracic surgical proce-
live births and is characterized by the absence of dures, especially in young female infants, should
the pectoralis muscles as well as abnormalities of always take the location and development of the
the breast, chest wall, and, occasionally, the ipsi- breast into consideration.
lateral upper extremity. It preferentially occurs
right sided and is found more often in boys (3:1).
Poland’s syndrome has also been linked to renal 32.2.3 Polythelia and Polymastia
diseases as well as thoracic tumors or leukemia.
Several strategies of surgical breast reconstruction Accessory nipples or NAC are termed polythelia
in Poland’s syndrome at the end of puberty have and occur along the milk line from the axilla to
been suggested, including placement of a mam- the pubis. They are most frequently located cau-
mary prosthesis and flap techniques. dal to a normally developed breast. The condition
Bilateral hypoplasia of the breast is usually a affects up to 5% of the population. A possible
familial condition. Significant hypoplasia may association of polythelia with congenital malfor-
also be related to connective tissue disorders, mations has been suggested, especially for renal
mitral valve prolapse, or genitourinary tract anomalies such as renal duplication, obstruction,
abnormalities. Surgical reconstruction of breast agenesis, and malignancy, and should be ruled
and/or nipple is best delayed until breast develop- out by a renal ultrasound. Surgery for polythelia
ment has been completed to achieve optimal can be performed for cosmetic reasons.
32  Breast Disorders in Children and Adolescents
Polymastia, which describes true accessory
breast tissue, is typically found in the lower
axilla and behaves like normal breast tissue. It
can cause discomfort in puberty, menstrual
periods, pregnancy, and lactation and is also
susceptible for similar diseases including breast
cancer. Thus, excision is indicated to prevent
discomfort and for cosmesis. The operation is
preferentially performed prior to the onset of
puberty to avoid possible glandular develop-
ment. However, high rates of postoperative
complications such as seroma, infection,
incomplete excision, nerval damages, painful
scarring or deformity need to be discussed with
the patient before surgery.
Ectopic breast tissue located outside the milk
line is extremely rare and has been found both in
the perineum and in the face.
32.2.4 Macromastia and Breast
Hypertrophy
Macromastia describes an excessive enlarge-
ment of the breast. It can be the result of virginal
hypertrophy, tumors, exogenous or endogenous
conditions leading to increased levels of estro-
gen and progesterone, and other rare conditions
(Table 32.1).
Table 32.1  Macromastia in female children and adoles-
cents: differential diagnosis and underlying disorders
Virginal hypertrophy
Breast tumors
Fibroadenoma
Hamartoma
Phyllodes tumor
Carcinoma
Sarcoma
Lymphoma
Metastasis of other tumors
Hormonally active tumors of other organs
Ovarian, adrenal, and pituitary gland tumors
Exogenous hormones
Estrogen, testosterone, corticosteroids, gonadotropins
Drugs
D-penicillamine, marijuana
407
Virginal hypertrophy is a usually bilateral
rapid and disproportional enlargement of the
breast due to a hypersensitivity of the fatty and
fibrous elements of mammalian tissue to gonadal
hormones in adolescents. Serum levels of estra-
diol are usually normal. There is no association
with obesity. An occasional association with
Hashimoto thyroiditis and rheumatoid arthritis
suggests an autoimmune cause of juvenile breast
hypertrophy. Typically, female patients suffer
from painful, large, and tender breasts associated
with skin alterations like striae, thin skin, or
dilated veins. The growth of the breast can be
excessive, leading to tissue necrosis and skin rup-
ture. Antiestrogenic medical treatment with bro-
mocriptine, tamoxifen, medroxyprogesterone
acetate, or dydrogesterone can reduce growth.
However, mammoplasty or mastectomy are often
indicated and have been reported to improve pre-
operative pulmonary function and complaints of
shortness of breath. Surgery should be postponed,
if tolerable, until breast size is static for 1 year.
Hormonal stimulation of breast tissue with
consecutive macromastia can also occur due to
hormonally active tumors of the ovary and the
adrenal or pituitary gland. Besides, exogenous
hormones such as corticosteroids, gonadotropins,
estrogen, and testosterone or other exogenous
factors, including marijuana and D-penicillamine,
can cause macromastia.
32.2.5 Gynecomastia
Gynecomastia is referred to an often bilateral but
asymmetric enlargement of glandular breast tis-
sue, resulting in a felt or seen enlargement of the
male breast. It can be observed in up to 70% of
boys during puberty and is familiar in 50% of
cases. Gynecomastia is characterized by fibro-
blastic stromal as well as ductal proliferation due
to an imbalance between estrogens and andro-
gens or an increased sensitivity to normal male
estrogen concentrations. It can be classified
according to Nydick et al. defining stage 1 as lim-
ited to the subareolar area and stage 5 resembling
a female breast (Fig. 32.1).
408
Fig. 32.1  Gynecomastia stage 5 resembling a female
breast in a 16-year-old boy
Initially, a detailed history and clinical exami-
nation including the testes is performed.
Gynecomastia combined with hypogonadism
may indicate Klinefelter’s syndrome. Testicular
tumors, malnutrition, liver diseases, hyperthy-
roidism, androgen insensitivity syndrome, sec-
ondary hypogonadism, and defects in testicular
enzymes are other known reasons for pseudogy-
necomastia. Moreover, a large number of medi-
cations like corticosteroids or cimetidine,
substances like marijuana, alcohol, heroin, meth-
adone, or amphetamines, as well as anabolic ste-
roids and herbal remedies can cause reversible
gynecomastia/macromastia. These issues need to
be addressed in the medical history.
In cases without evidence of any of the known
conditions associated with gynecomastia, partic-
ularly without evidence of a tumor, a follow-up
every 6  months has been suggested. In patients
with persistence of the condition for more than
1 year, endocrinological and oncologic workup is
proposed. The workup should include determina-
tion of serum levels of human chorionic gonado-
tropin, luteinizing hormone, testosterone, and
estradiol, as well as imaging studies in cases of
suspected testicular or adrenal tumors.
True pubertal gynecomastia is usually a self-­
limited condition with spontaneous regression in
75–95% of patients typically within 1–2  years.
However, fibrosis and hyalinization with regres-
sion of epithelial proliferation and persistence of
the enlarged fibrotic and inactive breast tissue
can also occur. Moreover, psychosocial distress
S. Mayer et al.
despite psychological support, pain, or tender-
ness may necessitate a treatment. Medical treat-
ment with estrogen receptor modulators,
antiestrogens, and aromatase inhibitors has been
tested especially in the early period of pubertal
gynecomastia before fibrous remodeling, but is
limited by potential side effects. Surgical treat-
ment includes central glandular resection and/or
liposuction. Postoperative complications are
common (20%) and include hematoma, seroma,
and scarring as well as technical errors including
asymmetry and contour irregularities.
32.2.6 Mastitis and Abscess
Mastitis in the newborn period mostly affects
term or near-term infants, females more often
than males. Staphylococcus aureus represents the
causative organism in nearly all cases, while,
occasionally, infections with Escherichia coli,
Salmonella, Shigella, and Klebsiella have been
reported. The neonatal breast usually presents
with typical signs of infection like induration,
swelling, occasional fluctuation, and excretion of
pus. An appropriate antibiotic treatment should
be initiated. Small abscesses can be needle aspi-
rated under ultrasonographic guidance. For larger
abscesses with relevant fluctuance, incision and
drainage are required. Of note, an asymmetric
reduction in breast size has been reported from a
small series of adolescents who had undergone
former drainage of breast abscesses. Therefore,
the periareolar incision should be limited and
placed radially to minimize iatrogenic lesions to
the mammary ducts.
Breast abscesses in adolescent females are
rare and may be attributed to nonpuerperal masti-
tis or infection of an epidermal cyst.
32.2.7 Nipple Discharge
Nipple discharge involves any fluid that seeps
out of the nipple. Galactorrhea represents milky
discharge and is a normal phenomenon in new-
borns. The neonatal breast responds to fetal pro-
lactin, which peaks at birth. Galactorrhea
32  Breast Disorders in Children and Adolescents
beyond the neonatal period, not related to preg-
nancy or breastfeeding, can be caused by pitu-
itary adenomas, neurologic disorders, or
hypothyroidism. Chronic stimulation of the
nipple, viral infections, and burns can lead to
neurogenic lactation. Prolactinoma with failure
of sexual maturation represents the most fre-
quent hypothalamic cause in boys and may also
occur in girls. Drugs such as estrogens, opiates,
catecholamine-depleting agents, and metoclo-
pramide can also induce galactorrhea. Besides
history and physical examination, diagnostic
studies should therefore include a pregnancy
test, prolactin levels, renal and thyroid function
tests, and, if indicated, magnetic resonance
imaging of the brain.
Other nipple discharges may include pus, cyst
liquid, or blood. Benign dilatations of the ducts
can lead to bacterial overgrowth with the forma-
tion of abscesses and purulent or bloody discharge,
requiring adequate antibiotic treatment. As there is
a high rate of spontaneous resolution of ductal
ectasia, surgical excision is only indicated if cystic
dilations of ducts persist. Bloody nipple discharge
can be caused by chronic irritation in athletic ado-
lescents. Bloody discharge may also be caused by
intraductal papillomas, which show ductal cells in
the effluence by exfoliative cytology and should
be excised. However, in adolescents and young
women, bloody nipple discharge remains a diag-
nostic dilemma, in particular, when malignant
sources have to be ruled out. Ultrasound examina-
tion may help to differentiate the underlying cause.
Central duct resection may also become necessary
in these patients.
32.3 Breast Masses
32.3.1 Prepubertal Masses
Occasionally, breast development is asynchro-
nous at the onset of puberty. In these cases,
unilateral growth of one breast may appear as
a palpable mass under the areola weeks to
months before growth of the other. This repre-
sents a physiological phenomenon. Thus, a
409
biopsy is not indicated and should not be per-
formed liberally to prevent injury to the devel-
oping breast.
32.3.2 Adolescent Masses
32.3.2.1 Fibroadenomas
Fibroadenomas represent the most common
breast masses in children and adolescents with
an overall incidence of 2% in this population.
They are a hyperplastic, rather than a neoplastic
process with a wide range of differential diag-
noses (Table  32.2). Fibroadenomas arise from
intralobular fibrous tissue, which may also dif-
ferentiate in phyllodes tumors. Juvenile fibroad-
enomas can occur before the onset or at early
puberty and mimic juvenile hypertrophy (giant
fibroadenomas). Adult fibroadenomas occur in
adolescence and are multiple in up to 25% of
patients.
Fibroadenomas typically present as firm,
mobile, painless, and easily palpable nodules,
often located in the upper outer quadrant. Physical
examination includes the breast, skin, and nipple
discharge as well as axillary lymph nodes.
Ultrasound examination reveals a well-­
circumscribed hypoechoic oval mass that is usu-
ally greater in width than in height. The diameter
usually ranges between 1 and 3 cm.
Table 32.2  Unilateral breast masses in female children
and adolescents: differential diagnosis and underlying
disorders
Physiologic unilateral premature thelarche
Inflammatory conditions
Mastitis, abscess
Fibrosis
Necrosis
Breast tumors
Fibroadenoma
Hemangioma, lymphangioma
Cyst
Papilloma
Lipoma, neurofibroma
Phyllodes tumor
Carcinoma
Metastasis
410
A small proportion of fibroadenomas grow
rapidly. However, the majority shows an initial
period of growth with subsequent stabilization
and spontaneous remission in up to 40% of the
patients. The risk of malignancy in fibroadeno-
mas is less than 0.3%, and even lower in children
and adolescence. Therefore, observation of pre-
sumed fibroadenomas is suggested for at least
one complete menstrual cycle to prevent unnec-
essary iatrogenic injury to the developing breast.
Fibroadenomas that are asymptomatic, not rap-
idly growing, and without cosmetic alterations,
may be managed conservatively or removed, if
the patient does not feel comfortable with obser-
vation. Surgical intervention is warranted for
fibroadenomas greater than 5 cm, growing, pain-
ful, deforming, multiple, bilateral or ultrasono-
graphically suspect for malignancy.
Giant fibroadenomas, which account for up to
2% of all fibroadenomas, are larger than 5 cm or
weigh more than 500 g. They are the most com-
mon cause for unilateral macromastia and can be
associated with rapid growth, skin ulcerations,
and venous engorgement. Moreover, giant fibro-
adenomas may not be distinguished from phyl-
lodes/malignant tumors by mammography or
ultrasonography. Thus, resection is warranted in
these patients.
32.3.2.2 Phyllodes Tumors
Phyllodes tumors are fibroepithelial tumors, his-
tologically classified from benign to malignant.
Phyllodes tumors can grow rapidly and recur but
rarely metastasize. Although these tumors mostly
occur in adults, they have also been reported in
prepuberty and adolescence. Clinically, phyl-
lodes tumors cannot be differentiated from fibro-
adenomas; ultrasonographically, they impose
with heterogeneous echo pattern and lobulations.
Occasionally, bloody nipple discharge is
observed, and fine-needle biopsy may not estab-
lish the diagnosis in all patients. Therefore,
biopsy through excision is recommended.
Benign phyllodes tumors should be excised
with a 1–2  cm margin of normal breast tissue.
Mastectomy is recommended in patients with
malignant phyllodes tumors. However, wide
resection or lumpectomy of clinically palpable
S. Mayer et al.
masses may also be an option. In cases of inad-
equate resection or chest wall infiltration, re-­
excision and radiation therapy have been
suggested. In patients with local recurrence,
which occurs in 20%, mastectomy is indicated.
Phyllodes tumors can metastasize in lymph
nodes, soft tissue, bone, pancreas, lung, and
nerve system. The recurrence rate in benign
tumors greater than 5 cm in diameter has been
reported to be as high as 39% compared to 10%
in smaller ones. Malignant transformation has
been observed in 20% of benign tumors with
recurrence. However, the 5-year survival rate in
adults with benign phyllodes tumors is 96% and
66% in malignant tumors. The prognosis in ado-
lescents has been suggested to be favorable. The
role of adjuvant radiotherapy for malignant
phyllodes tumor in children is still under debate.
32.3.2.3 Malignant Tumors
Primary carcinomas of the breast are extremely
rare in children. However, the condition has been
described in infants as young as 6 years of age.
Primary breast cancer in female adolescents
include all histologic types seen in adults includ-
ing invasive intraductal, invasive lobular, and sig-
net ring tumors. Adolescents and young adult
women (15–39  years) are more likely to have
familial cancer predisposition genes (e.g.,
BRCA1 or BRCA2), larger breast tumors, unfa-
vorable biological characteristics, distant meta-
static disease at diagnosis, and an adverse
outcome. Juvenile secretory carcinomas are char-
acterized by a lower grade malignant behavior
with a better prognosis. These tumors have a cap-
sule and may appear cystic on ultrasonography.
The treatment of primary carcinomas of the
breast in children and adolescents is performed
according to the national protocols for adult pri-
mary breast cancer and according to histology,
stage, and hormone receptors. Breast-conserving
surgery may be feasible. Sentinel lymph node
sampling is advocated as 20–30% of patients
have axillary metastases.
Rhabdomyosarcoma, lymphoblastic non-­
Hodgkin’s lymphoma of the breast, and other
breast tumors like liposarcoma, fibrosarcoma, or
osteosarcoma are extremely rare in children.
32  Breast Disorders in Children and Adolescents 411

Metastases in the breast have been reported for De Silva NK, Brandt ML (2006) Disorders of the breast
retinoblastoma, osteosarcoma, neuroblastoma, in children and adolescents, part 1: disorders of growth
and infections of the breast. J Pediatr Adolesc Gynecol
leukemia, lymphoma, and rhabdomyosarcoma. 19:345–349
Dixon JM, Mansel RE (1994) ABC of breast diseases.
Congenital problems and aberrations of normal breast
32.4 Conclusion development and involution. BMJ 309:797–800
Fruhstorfer BH, Malata CM (2003) A systematic approach
to the surgical treatment of gynaecomastia. Br J Plast
Although functional deficits induced by breast Surg 56:237–246
deformities that require surgical repair in young Gao Y, Saksena MA, Brachtel EF et  al (2015) How to
infants are rare, breast disorders can afflict both approach breast lesions in children and adolescents.
Eur J Radiol 84:1350–1364
patients and families, particularly during puberty Guss CE, Divasta AD (2017) Adolescent gynecomastia.
and adolescence, causing social embarrassment, Pediatr Endocrinol 14:371–377
depression, peer rejection, as well as low self-­ Holcomb GW, Murphy JP, Ostlie DJ (2014) Ashcraft’s
confidence and sexual dysfunction. Surgical cor- pediatric surgery. Springer
Kulkarni D, Dixon JM (2012) Congenital abnormalities
rection can improve self-esteem and health-related of the breast. Womens Health 8:75–86
quality of life in these patients. However, surgical Latham K, Fernandez S, Iteld L et  al (2006a) Pediatric
interventions may result in malfunctioning of breast deformity. J Craniofac Surg 17:454–67
lactation and breastfeeding in later life. Cosmetic Latham K, Fernandez S, Iteld L et  al (2006b) Pediatric
breast deformity. J Craniofac Surg 17:454–467
repairs should therefore be delayed until breast Lee M, Soltanian HAT (2015) Breast fibroadenomas in
maturation and development is complete to adolescents: current perspectives. Adolesc Health
achieve the maximum possible symmetry. Of Med Ther 6:159–163
note, vascular malformations like hemangiomas Lee EJ, Chang YW, Oh JH, Hwang J, Hong SS, Kim
HJ (2018) Breast lesions in children and adoles-
and lymphangiomas may also impede breast cents: diagnosis and management. Korean J Radiol
developing during rapid growth with subsequent 19:978–999
breast deformity. Therefore, therapeutic options Lemaine V, Cayci C, Simmons PS et  al (2013)
including propranolol therapy for hemangiomas Gynecomastia in adolescent males. Semin Plast Surg
27:56–61
or (early) surgery have to be discussed with the Lewis EJ, Crutchfield CE, Prawer SE (1997) Accessory
families. nipples and associated conditions. Pediatr Dermatol
The pediatric surgeon usually accompanies 14:333–334
families and patients during childhood and ado- Merlob P (2003) Congenital malformations and develop-
mental changes of the breast: a neonatological view. J
lescence until potential surgical repair. Moreover, Pediatr Endocrinol Metab 16:471–485
the contour of the breast changes during life and Moir CR, Johnson CH (2008) Poland’s syndrome. Semin
might require additional surgical procedures later Pediatr Surg 17:161–166
in life. Successful transition of these patients to Murphy BL, Glasgow AE, Ubl DS et al (2018) Surgical
treatment of adolescent breast disorders: institutional
adult plastic surgical treatment is therefore experience and national trends. J Pediatr Adolesc
mandatory. Gynecol 31:299–303
Nordt CA, DiVasta AD (2008) Gynecomastia in adoles-
cents. Curr Opin Pediatr 20:375–382
Oprić S, Oprić D, Gugić D et al (2012) Phyllodes tumors
Bibliography and fibroadenoma common beginning and different
ending. Coll Antropol 36:235–241
Beer GM, Budi S, Seifert B et al (2001) Configuration and Peña KS, Rosenfeld JA (2001) Evaluation and treatment
localization of the nipple-areola complex in men. Plast of galactorrhea. Am Fam Physician 63:1763–1770
Reconstr Surg 108:1947–1952 Pryor LS, Lehman JA, Workman MC (2009) Disorders
Brown J, Schwartz RA (2003) Supernumerary nipples: an of the female breast in the pediatric age group. Plast
overview. Cutis 71:344–346 Reconstr Surg 124:50e–60e
Caouette-Laberge L, Borsuk D (2013) Congenital anoma- Richards MK, Goldin AB, Beierle EA et al (2017) Breast
lies of the breast. Semin Plast Surg 27:36–41 malignancies in children: presentation, management,
De Silva NK (2018) Breast development and disorders and survival. Ann Surg Oncol 24:1482–1491
in the adolescent female. Best Pract Res Clin Obstet Rogers DA, Lobe TE, Rao BN et al (1994) Breast malig-
Gynaecol 48:40–50 nancy in children. J Pediatr Surg 29:48–51
412
Sadove AM, van Aalst JA (2005a) Congenital and
acquired  pediatric breast anomalies: a review of 20
years’ experience. Plast Reconstr Surg 115:1039–50
Sadove AM, van Aalst JA (2005b) Congenital and
acquired  pediatric breast anomalies: a review of 20
years’ experience. Plast Reconstr Surg 115:1039–1050
Sonmez K, Turkyilmay Z, Karabulut R et  al (2006)
Surgical breast lesions in adolescent patients and a
review of the literature. Acta Chir Belg 106:400–404
S. Mayer et al.
Templeman C, Hertweck SP (2000) Breast disorders in
the pediatric and adolescent patient. Obstet Gynecol
Clin N Am 27:19–34
Vargas HI, Romero L, Chlebowski RT (2002) Management
of bloody nipple discharge. Curr Treat Options in
Oncol 3:157–161
 Congenital Airway Malformations
Patricio Varela and Richard Azizkhan
33.1 Introduction
As an initial step in evaluating an infant with
respiratory compromise, a clinician must thor-
oughly review the history of the child’s airway
symptoms. This review may provide information
that helps in identifying the underlying etiology,
which may determine or have an important
impact on management strategy. Close attention
should be paid to circumstances that trigger the
onset of symptoms, the duration of symptoms,
and symptom progression over time. Questioning
parents about a history of dysphagia or feeding
problems, the nature of their child’s cry, and the
possibility of foreign body aspiration can also
yield useful information. Additionally, any previ-
ous history of endotracheal intubation, trauma, or
cardiopulmonary abnormalities should be care-
fully reviewed.
Airway obstruction may range from subtle to
severe. Less severe airway compromise fre-
quently manifests in subtle symptoms such as
P. Varela (*)
Dr Luis Calvo Mackenna Children’s Hospital,
Division of Pediatric Surgery, Pediatric Airway and
Chest Wall Malformation Center, University of Chile,
Santiago, Chile
Division of Pediatric Surgery, Clinica Alemana
Medical Center, Santiago, Chile
R. Azizkhan
Cincinnati Children’s Hospital Medical Center,
Cincinnati, OH, USA
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_33
33
irritability, whereas more severe obstruction is
likely to manifest in severe suprasternal and
intercostal retractions, tachypnea, lethargy, and
cyanosis. Stridor, a harsh sound caused by turbu-
lent airflow through a partial obstruction of the
airway, is the most important symptom of upper
airway obstruction. Depending on the location of
the obstruction in the upper airway, this symptom
can be present during either the expiratory or
inspiratory phase of the respiratory cycle or dur-
ing both of these phases. The characteristics of
stridor as well as its relationship to the respira-
tory cycle are generally helpful in establishing a
differential diagnosis and in setting the priorities
for diagnostic evaluation.
33.2 Diagnostic Evaluation
Conducting a complete and thorough endoscopic
evaluation is the most critical part of airway
assessment. This evaluation generally incorpo-
rates both flexible and rigid bronchoscopy. For
the evaluation of certain types of lesions (e.g.,
tracheomalacia), airway assessment is performed
with the patient awake or lightly sedated and
spontaneously breathing. Clinicians should be
aware of the fact that 17% of patients have a sec-
ond airway lesion. Evaluation of the entire air-
way is thus essential. As up to 45% of children
with congenital airway obstruction also have
other significant nonairway anomalies, these
413
414
patients require meticulous and complete overall
investigation.
Complementary imaging studies are useful in
both diagnosis and patient management.
Computed tomography (CT) and magnetic reso-
nance (MR) imaging studies provide rapid and
precise methods of measuring the extent and
length of airway narrowing or displacement.
Additionally, they assist in identifying associated
mediastinal and pulmonary anomalies. In par-
ticular, MR angiography is useful in assessing
the relationship of mediastinal great vessel
­anomalies (e.g., vascular rings, pulmonary artery
slings) to the airway. Newer computer software
allows for three-dimensional image reconstruc-
tion and is helpful in planning surgical proce-
dures. Echocardiography is primarily used to
assess the presence of intracardiac defects and
can detect the majority of associated mediastinal
vascular anomalies. Contrast swallow studies are
valuable in assessing esophageal motility, aspira-
tion, and some mediastinal lesions that impact
the airway. Fiber-optic endoscopic evaluation of
swallowing (FEES) is performed to evaluate
structural and functional disorders of swallowing
and to identify functional problems of the larynx,
pharynx, epiglottis, and proximal esophagus.
33.3 Congenital Laryngeal
Anomalies
33.3.1 Laryngomalacia
33.3.1.1 Pathogenesis
Laryngomalacia is characterized by laxity of
both the glottic and supraglottic tissues, causing
the epiglottis, arytenoids, and aryepiglottic folds
to collapse and partially obstruct during inspira-
tion. This malformation is the most common con-
genital laryngeal anomaly. It accounts for
60–75% of laryngeal problems in the neonate
and is also the most common cause of stridor in
the neonate. Stridor caused by laryngomalacia
usually is evident soon after birth or within the
first few days of life. It is generally mild but can
be exacerbated by feeding, crying, and lying in a
supine position. Fifty percent of children with
P. Varela and R. Azizkhan
laryngomalacia experience a worsening of stridor
during the first 6  months of life. Children with
severe laryngomalacia may have apnea, cyanosis,
severe retractions, and failure to thrive. Cor pul-
monale has been reported in cases that are
extremely severe (Backer et al. 1997; Hysinger
and Panitch 2016; Ngerncham et al. 2015).
33.3.1.2 Diagnosis
Diagnosis is confirmed by flexible transnasal
fiber-optic laryngoscopy, which reveals short ary-
epiglottic folds, with prolapse of the cuneiform
cartilages. Collapse of the supraglottic structures
is seen on inspiration, and inflammation indica-
tive of reflux laryngitis is also frequently seen
(Fig.  33.1). Although symptoms spontaneously
resolve by the age of 1 year, the infant with severe
laryngomalacia may die of asphyxiation.
The need for surgical intervention and the
type of surgery required are primarily based on
the severity of symptoms. For severe symptoms,
supraglottoplasty (also termed epiglottoplasty) is
the operative procedure of choice. Redundant
mucosa of the aryepiglottic folds is excised and
one or both cuneiform cartilages also may be
removed. If the aryepiglottic folds alone are
divided, postoperative intubation is generally not
required. Reflux management is useful in helping
to minimize laryngeal edema. Patients are
observed overnight in the intensive care unit.
Fig. 33.1  Endoscopic view of laryngomalacia in an
infant showing partial collapse of the supraglottic struc-
tures during inspiration
33  Congenital Airway Malformations
Occasionally, the infant’s obstructive symptoms
continue despite an adequate appearance of the
larynx after surgery. Such cases are sometimes
related to underlying neurologic problems and
have a high likelihood of requiring tracheotomy
placement (Austin and Ali 2003; Azizkhan 2005;
Rutter et al. 2003).
33.3.2 Subglottic Stenosis
Subglottic stenosis (SGS) involves a narrowing
of the subglottic lumen. This malformation may
be either congenital or acquired, with the latter
more commonly seen and generally a sequela of
prolonged intubation during the neonatal period.
33.3.2.1 Pathogenesis
It is thought to be caused by the failure of the
laryngeal lumen to recanalize during embryogen-
esis. Incomplete recanalization results in various
degrees of stenosis that range from mild to severe;
levels of severity are graded according to the
Myer-Cotton grading system (Table  33.1). Mild
SGS (no obstruction to 50% obstruction) may
manifest in recurrent upper respiratory infections,
which frequently are diagnosed as croup. More
severe cases (71–99% obstruction) may present
with acute airway compromise and require endo-
tracheal intubation or tracheostomy placement at
the time of delivery. Many infants with severe
obstruction may, however, remain asymptomatic
for weeks or months. When stridor is present, it is
initially inspiratory. As the severity of obstruction
increases, stridor becomes biphasic (Herrera et al.
2007; Ngerncham et al. 2015).
33.3.2.2 Diagnosis
In the full-term neonate, SGS is defined as a lumen
of 4 mm or less in diameter at the level of the cri-
Table 33.1  Myer-Cotton grading system for subglottic
stenosis level of airway obstruction
Classification From To
Grade I No obstruction 50% obstruction
Grade II 51% obstruction 70% obstruction
Grade III 71% obstruction 99% obstruction
Grade IV No detectable lumen
415
Fig. 33.2  Very high grade of congenital subglottic steno-
sis in a symptomatic neonate
coid; in the premature infant, it is considered pres-
ent when this lumen measures 3  mm or less.
Congenital SGS (Fig. 33.2) may arise as an isolated
anomaly or may be associated with other congenital
head and neck lesions and syndromes such as a
small larynx in a patient with Down syndrome.
Radiologic evaluation of the nonintubated air-
way can provide information regarding the location
and extent of the stenosis. Useful studies include a
chest X-ray, inspiratory and expiratory lateral soft
tissue neck films, and fluoroscopy to demonstrate
the dynamics of the trachea and larynx. High-
kilovoltage airway films are particularly important
as they identify the characteristic steeple-like con-
figuration seen in patients with SGS as well as pos-
sible tracheal stenosis. Flexible and rigid
endoscopies are the gold standard of evaluating the
airway. Flexible endoscopy documents the dynam-
ics of the hypopharyngeal and laryngeal airways,
whereas rigid endoscopy provides a clear image of
the entire laryngotracheobronchial airway.
In patients with mild to moderate disease,
congenital SGS improves with age. Less than
50% of these patients require tracheotomy place-
ment to maintain their airway. Children with sig-
nificant airway obstruction are best managed
with open airway reconstruction. Costal cartilage
grafts are placed through either the anterior or
posterior lamina of the cricoid cartilage or both.
416
Stenting and placement of a temporary tracheos-
tomy may be necessary. Recently, superior results
for the management of severe SGS have been
obtained by performing cricotracheal resection.
Successful management depends on the presence
of comorbidities such as gastroesophageal reflux
(GER), eosinophilic esophagitis, and low-grade
tracheal infection.
33.3.3 Vocal Cord Paralysis
Vocal cord paralysis is the third most common
cause of neonatal stridor. This condition can be
either congenital or acquired and can occur either
unilaterally or bilaterally. Unilateral paralysis is
usually an acquired condition caused by damage
to the recurrent laryngeal nerve, whereas bilateral
vocal cord paralysis is usually evident at birth.
33.3.3.1 Pathogenesis
Bilateral paralysis is generally idiopathic, it is fre-
quently seen with central nervous system problems
such as hydrocephalus and Chiari malformation of
the brainstem. Most children with bilateral paralysis
present with significant airway compromise, though
they do not aspirate (Ngerncham et al. 2015).
33.3.3.2 Diagnosis
Diagnosis is made by awake flexible laryngoscopy.
Subsequent investigation for the underlying cause
should then be carried out. Stabilization can be
achieved with intubation, continuous positive air-
way pressure (CPAP), or high-flow nasal cannula as
an alternative temporizing measure. Most infants
I II IIIa
Fig. 33.3  Posterior Laryngeal Cleft Classification
P. Varela and R. Azizkhan
(90%) affected bilaterally require tracheotomy
placement. Up to 50% of children with congenital
idiopathic bilateral vocal cord paralysis experience
spontaneous resolution of their paralysis by the age
of 1 year. As such, decannulation is almost always
delayed to allow time for this to occur.
A number of surgical procedures have been
used, in the past, for congenital bilateral paraly-
sis, including laser cordotomy and partial or
complete arytenoidectomy. Other procedures
have also been proposed: vocal process lateral-
ization (open or endoscopically guided), endo-
scopic posterior graft placement, or anterior
posterior cricoid split. Each of these options
aims to achieving an adequate decannulated air-
way while maintaining voice and preventing
aspiration.
33.3.4 Posterior Laryngeal Cleft
33.3.4.1 Pathogenesis
Posterior laryngeal cleft is a rare congenital mal-
formation that results from embryologic failure
of the laryngotracheal groove to fuse.
33.3.4.2 Classification
This malformation comprises six anatomic sub-
types that differ with respect to involvement of
the larynx and trachea (Fig. 33.3). Other associa-
tions, many of which affect the airway, are com-
mon. Such anomalies include tracheomalacia
(always present in varying levels of severity), tra-
cheoesophageal fistula (20%), laryngomalacia,
vocal cord paralysis, SGS, and innominate artery
THORACIC
INLET
IIIb IVa IVb
33  Congenital Airway Malformations
compression (Bennett and Holinger 2003; Davies
and Cywes 1978; Eriksen et al. 1990; Evans
1985; Grillo et al. 2002; Kluth et al. 1987; Rahbar
et al. 2006, 2009 ; Sandu and Monnier 2006).
Associated conditions that do not involve the
airway include GER, which is present in most
children, cleft lip and palate, congenital heart
defects, and hypospadias. The most common
associated syndrome is Opitz-Frias syndrome,
which is characterized by hypertelorism, anorec-
tal malformations, genital urinary anomalies, and
laryngeal clefting.
33.3.4.3 Diagnosis
Diagnosis can be extremely challenging and elu-
sive, as presenting symptoms vary greatly and are
not specifically diagnostic. Symptoms are often
subtle and may mimic those of other disorders
(e.g., GER). Some patients present early with
feeding problems, choking, chronic coughing,
wheezing, cyanotic spells, and apnea. There may
be associated stridor either due to redundant
mucosa on the edge of the cleft or a small cricoid
ring. Severe tracheomalacia may also signifi-
cantly compromise the airway, especially in chil-
dren with an associated tracheoesophageal
fistula. Although contrast swallow studies may
demonstrate aspiration, definitive diagnosis
requires rigid laryngotracheal bronchoscopy,
with the interarytenoid area being specifically
probed to determine if a posterior laryngeal cleft
is present (Fig. 33.4).
Fig. 33.4  Type III laryngotracheal cleft. View under 0
degrees rigid endoscopy
417
In children who are symptomatic and do not
have other more severe anomalies, repair of the
posterior laryngeal cleft should be carried out as
soon as possible to prevent chronic microaspira-
tion with long-term pulmonary sequelae. Prior to
repair, consideration should be given to whether
the infant requires tracheotomy placement, gas-
trostomy tube placement, and Nissen fundopli-
cation. Most type I and some type II clefts are
amenable to endoscopic surgical repair, whereas
clefts that extend into the cervical or thoracic
trachea require open repair. A transtracheal
approach is advised as it provides unparalleled
exposure of the cleft while protecting the recur-
rent laryngeal nerves. A two-layer closure is rec-
ommended, with the option of performing an
interposition graft (perichondrium or perios-
teum) if warranted. Type IV clefts are often asso-
ciated with multiple congenital anomalies. These
long clefts are exceedingly difficult to repair and
are prone to anastomotic breakdown. Success
rates for cleft repair vary significantly (50–90%)
depending on both the severity of the cleft and
the presence of coexisting congenital anomalies
and comorbidities.
33.3.5 Laryngeal Atresia
33.3.5.1 Congenital High Airway
Obstruction Syndrome
(CHAOS)
CHAOS is a life-threatening, prenatally diag-
nosed condition caused by complete or near-­
complete obstruction of the fetal airway. This
obstruction may be due to laryngeal atresia
(Fig. 33.5a) or tracheal agenesis. Atresia is some-
times an isolated anomaly but is often seen asso-
ciated with a spectrum of other anomalies,
including hydrocephalus malformation of the
aqueduct of Sylvius, bronchotracheal fistula,
esophageal atresia, tracheoesophageal fistula,
syndactyly, and genitourinary, vertebral, and car-
diac anomalies (Lim et al. 2003). Prenatal ultra-
sound findings indicative of CHAOS include
bilaterally enlarged echogenic lungs, dilated air-
ways, and flattened or everted diaphragms with
associated fetal ascites and nonimmune hydrops
418
a
b threatening anomalies are ruled out, careful endo-
Fig. 33.5 (a) Laryngeal atresia. No lumen in the subglot-
tic space. (b) Fetal ultrasonography demonstrating find-
ings consistent with the diagnosis of congenital high
airway obstruction: enlarged echogenic lungs, dilated air-
way (white arrow), flattened or everted diaphragms, and
fetal ascites (white star) and hydrops. Fetal liver and intes-
tines are marked with a black arrow (courtesy of Timothy
Crombleholme MD, Cincinnati, Ohio)
(Fig. 33.5b). A fetus identified with these sono-
graphic features is at significant risk of intrauter-
ine death and faces a high mortality rate during
the rest of pregnancy to delivery.
Diagnosis in the middle of the second trimes-
ter generally correlates with a poor outcome. A
P. Varela and R. Azizkhan
fetus presenting in the third trimester with
CHAOS, in the absence of associated anomalies
or hydrops, is likely to have incomplete obstruc-
tion. As such, this fetus is more likely to do well
until delivery by the ex utero intrapartum tech-
nique (EXIT) procedure. This procedure main-
tains placental circulation to the fetus, while
securing the airway at the time of delivery.
Instrumentation of the airway, including trache-
ostomy, may be accomplished at this time.
For the newborn diagnosed with CHAOS,
securing and maintaining the airway are of utmost
importance. Once the infant’s cardiorespiratory
status is stable and other critical or potentially life-
scopic evaluation of the airway precedes elective
laryngotracheal reconstruction. Nevertheless, opti-
mal timing for reconstruction has not yet been
determined. Although an adequate airway can be
constructed, adequate speech may not be feasible.
33.4 Anomalies of the Trachea
and Bronchi
33.4.1 Tracheal Agenesis
Tracheal agenesis is a rare developmental abnor-
mality that is almost always incompatible with
life. Severe respiratory distress is present at birth,
with the neonate attempting ventilation through
bronchoesophageal communications. Although
temporary ventilation may be possible with intu-
bation of the esophagus, this generally cannot be
sustained and results in neonatal demise. When a
bronchoesophageal communication is not pres-
ent, the fetus will have CHAOS (Lim et al. 2003;
Ngerncham et al. 2015; Tazuke et al. 2015).
33.4.2 Tracheal Webs and Stenosis
A broad spectrum of rare tracheal anomalies is clas-
sified as tracheal stenosis. Affected segments differ
in the degree and extent of stenosis, ranging from
gossamer thin webs to more severe long segments
of stenosis that may involve the entire airway.
33  Congenital Airway Malformations
33.4.2.1 Tracheal Webs
Tracheal webs are rare and involve an intraluminal
soft tissue stenosis of the trachea. These webs may
be membranous or composed of thick, inelastic
tissue. Symptoms include biphasic s­ tridor or expi-
ratory wheezing, with severity dependent on the
degree of tracheal narrowing. Thin webs can be
readily managed by hydrostatic balloon dilatation
alone. Children with a web greater than 1 cm in
length, or in whom the airway cartilage is sus-
pected to be deficient or structurally abnormal, are
best managed by segmental tracheal resection
(Backer et al. 1997; Ngerncham et al. 2015).
33.4.2.2 Cartilaginous Ring Aplasia
Cartilaginous ring aplasia is an exceedingly rare
anomaly in which a short segment of the trachea
lacks cartilage. This creates a region that is both
malacic and stenotic. The remainder of the tra-
chea is normal and children generally do not have
other congenital anomalies. Segmental resection
of the trachea is usually curative (Ngerncham et
al. 2015; Puri and Höllwarth 2006).
33.4.2.3 Tracheal Cartilaginous
Sleeve
Tracheal cartilaginous sleeve is an anomaly in
which there are no discrete cartilaginous rings but
rather a fused cartilaginous cylinder, with or with-
out a membranous portion. This anomaly is often
associated with craniosynostosis syndromes such
as Apert, Pfeiffer, Crouzon, and Goldenhar. Patients
present during the neonatal period with respiratory
illness or in early infancy with acute respiratory
symptoms, which may include biphasic stridor with
respiratory distress, cough, and recurrent respira-
tory infections. Also, tracheal rigidity may cause
Fig. 33.6 (a, b)
Congenital tracheal a b
stenosis. (a) Endoscopic
view demonstrating
complete tracheal rings.
(b) Histology of tracheal
segment demonstrating
virtually complete
tracheal cartilaginous
ring
419
difficulty in clearing secretions. Bronchoscopy
shows a smooth anterior tracheal wall without the
normal appearance of tracheal rings. The mem-
branous posterior tracheal wall may be normal,
reduced, or absent. CT and MR imaging may assist
in delineating the extent of the lesion. Most of the
patients require resection and repair.
33.4.2.4 Complete Tracheal Rings
Complete tracheal rings are an anomaly in which
the trachea alone or both the trachea and bronchi are
narrowed. In more than 50% of infants, a segmental
stenosis is found. In such patients, the tracheal car-
tilage is abnormally shaped and forms complete
rings (Figs. 33.6a, b). The clinical manifestations of
complete tracheal rings vary from life-threatening
respiratory distress at birth to subtle symptoms of
airway compromise in older children. Many infants
present with worsening of respiratory function over
the first few months of life. Symptoms include stri-
dor, retractions, cough, and alterations of cry.
Atypical and persistent wheezing and rhonchi and
sudden death can also occur. Over 80% of children
with complete tracheal rings have other congenital
anomalies, the most common of which are esopha-
geal, cardiac, skeletal, and genitourinary. Fifty per-
cent of children with complete tracheal rings also
have a left pulmonary artery sling or vascular ring
(Backer et al. 2001; Bando et al. 1996; Benjamin
and Inglis 1989; Cotton 2003; DeMarcantonio et al.
2017; Fiore et al. 2005; Furman et al. 1999;
Gorostidi et al. 2016; Gustafson et al. 2000; Hewitt
et al. 2016; Hysinger 2018; Rutter 2006; Speggiorin
et al. 2012).
In some patients, placement of an endotracheal
tube may exacerbate respiratory distress by caus-
ing acute swelling and inflammation. Partially
420
obstructing tracheal lesions may also become life-
threatening following the onset of a respiratory
infection. In an infant or child with an abnormal
trachea, the cross-sectional area of airway can be
decreased by one-third to one-half of its normal
diameter with as little as 1  mm of edema. This
accounts for the rapid progression of symptoms in
some children who have acute inflammatory con-
ditions superimposed on existent tracheal narrow-
ing (Backer et al. 2001; Bando et al. 1996;
Benjamin and Inglis 1989; Cotton 2003;
DeMarcantonio et al. 2017; Fiore et al. 2005;
Furman et al. 1999; Gorostidi et al. 2016;
Gustafson et al. 2000; Hewitt et al. 2016; Hysinger
2018; Rutter 2006; Speggiorin et al. 2012).
33.4.2.5 Diagnosis
Expeditious diagnostic evaluation to define aber-
rant and normal tracheobronchial anatomy is
required. Although an initial high-kilovolt airway
film may show tracheal narrowing, the precise
location and extent of the narrowing is best
achieved by endoscopic techniques. CT scans pro-
vide a rapid and precise method of measuring the
extent and length of airway narrowing or displace-
ment. Visualization of the anatomic relationship
between the airways and surrounding structures
can be enhanced with intravascular contrast. Newer
computer software allows for three-dimensional
image reconstruction and is helpful in planning
Fig. 33.7  CT scan with three-dimensional reconstruction
to demonstrate anatomy of the trachea in a patient with
congenital tracheal stenosis involving the distal trachea.
This patient had an aberrant tracheal right bronchus
P. Varela and R. Azizkhan
surgical procedures (Fig.  33.7). MR imaging is
useful in assessing the relationship of the mediasti-
nal great vessels to the airway. Echocardiography
is primarily used to assess the presence of intracar-
diac defects and can detect the majority of associ-
ated mediastinal vascular anomalies.
Most children with complete tracheal rings
require tracheal reconstruction. If a pulmonary
artery sling or vascular ring is present, repair of
such an anomaly should be undertaken concurrent
with the tracheal repair. Segmental tracheal resec-
tion with end-to-end anastomosis is considered
the treatment of choice for short-segment tracheal
stenosis. Slide tracheoplasty is currently the pro-
cedure of choice for long segments of tracheal
involvement, having replaced patch tracheoplasty
(Fig. 33.8). This approach yields significantly less
morbidity than other tracheal reconstruction tech-
niques and is applicable to virtually all anatomic
variants of complete tracheal rings. Slide tracheo-
plasty uses only autologous tracheal tissue and is
performed by transecting the trachea into two
equal segments. The anterior wall of the upper
half of the trachea and the posterior wall of the
lower trachea are incised. These segments are
then slid over each other and anastomosed with
5–0 monofilament and absorbable sutures.
Following surgery, the airway has four times the
cross-sectional area and one-half the length of its
previous dimension. Airflow is increased 16-fold
with this method of airway reconstruction.
Postoperatively, endotracheal intubation is
required for 1 day to several weeks, though most
patients are extubated within 48 h. To minimize
the risk of damage to the newly reconstructed air-
way, unnecessary movements of the endotracheal
tube or unplanned extubation must be avoided.
Nasotracheal intubation is used preferentially
because the endotracheal tube can be stabilized
in position more securely. Patients require con-
tinuous monitoring, careful pulmonary toilet, and
endoscopic removal of any obstructing granula-
tion tissue. Just prior to and to ensure a safe extu-
bation, the integrity and patency of the
reconstructed airway are assessed by flexible
fiber-optic endoscopy through the endotracheal
tube. Airway configuration following slide tra-
cheoplasty may resemble figure of 8 trachea
(Fig. 33.9) but is not associated with an obstruc-
33  Congenital Airway Malformations
Fig. 33.8  Slide tracheoplasty procedure: the trachea is
transversely divided at the midpoint of the tracheal steno-
sis. After proximal and distal tracheal mobilization, the
anterior portion of the cephalic trachea segment and the
a b
Fig. 33.9  Figure of 8 trachea following repair of com-
plete tracheal rings: (a) preoperative endoscopy docu-
ments complete tracheal rings; (b) postoperative
endoscopy at 6  weeks demonstrates the figure of 8 tra-
tive airway. In most cases, the trachea remodels
to a normal oval shape within 1  year of recon-
struction. Long-term survival following this pro-
cedure is currently 80–90% in main airway
centers. Mortality is usually associated with
severe comorbidities such as cardiac disease
rather than airway complications.
421
posterior portion of the caudal tracheal segment are
incised. The two tracheal segments are then overlapped
and obliquely sutured together (*courtesy Gaston Bellia
Md, Buenos Aires, Argentina)
c
chea, and the patient had no airway symptoms; (c) endos-
copy at 2  years demonstrates significant remodeling of
airway
33.4.3 Tracheal Diverticulum
and Tracheal Bronchus
Tracheal diverticulum and tracheal bronchus are
relatively common abnormalities of tracheal bud-
ding that occur during the third and fourth gesta-
tional weeks, when the trachea bifurcates and
422
differentiates. Tracheal diverticula resemble a
bronchus, though they originate from the trachea
and end blindly or communicate with a rudimen-
tary lung. Tracheal bronchi most often affect the
right upper lobe bronchus and may connect to an
isolated intrathoracic lung segment or the apical
segment of an upper lobe. Both anomalies are
often seen in children with other congenital
defects. Diagnosis is established by airway
endoscopy (Ngerncham et al. 2015). Most chil-
dren are asymptomatic and do not require treat-
ment. Those who are symptomatic experience
symptoms such as pneumonia and respiratory
distress during the neonatal period. Additionally,
they have an associated stenosis of the bronchus
or other lung anomalies. Resection of involved
lobe and bronchus in these patients is generally
curative.
33.4.4 Tracheomalacia
and Bronchomalacia
33.4.4.1 Pathogenesis
Tracheomalacia and bronchomalacia are condi-
tions in which the structural integrity of the tra-
chea or bronchi is diminished and the cartilaginous
rings of the airway lack the necessary rigidity to
prevent airway collapse during expiration.
Malacia may occur in localized segments or dif-
fusely throughout the airway. Tracheomalacia is
the most common congenital tracheal anomaly.
It may be idiopathic or associated with a num-
ber of conditions, including esophageal atresia or
tracheoesophageal fistula, aberrant innominate
artery, mediastinal masses, prolonged intubation
a trachea b c
esophagus
Fig. 33.10 (a–d) Types of tracheal collapse (a) normal anatomy, (b) anterior collapse, (c) posterior intrusion, (d) com-
bined (*courtesy Diana Romero MD, Bogotá, Colombia)
P. Varela and R. Azizkhan
for interstitial lung disease, or bronchopulmo-
nary dysplasia. Premature neonates with bron-
chopulmonary dysplasia or children with chronic
indwelling cuffed endotracheal or tracheostomy
tubes are at particular risk for developing com-
bined severe tracheal and bronchial malacia.
Several types of tracheal collapse have been
described (Fig. 33.10a–d).
Presenting symptoms vary and depend on the
severity, duration, and location of the malacia.
Most children are either asymptomatic or mini-
mally symptomatic and most cases involve poste-
rior malacia of the trachealis, with associated
broad tracheal rings. Other associated abnormali-
ties include laryngeal clefts and tracheoesopha-
geal fistulae. Presenting symptoms may include a
honking cough, stridor, wheezing, respiratory
distress when agitated, and cyanosis. Some chil-
dren are misdiagnosed with allergic asthma and
unsuccessfully treated with bronchodilators.
Diagnosis is best established by rigid or flexible
bronchoscopy, with the patient breathing sponta-
neously; this demonstrates dynamic distortion
and compression of the trachea. Children who are
minimally symptomatic are watched expectantly.
Their symptoms often resolve by age 3. Children
who experience a worsening of symptoms require
more intensive medical or surgical intervention.
Respiratory monitoring with nasal CPAP may be
beneficial in some patients.
Segmental tracheal involvement is managed
with endoscopic or open aortopexy, with thymec-
tomy and anterior suspension of the ascending
arch of the aorta to the posterior periosteum of
the sternum. More diffuse malacia may require
tracheotomy placement with positive pressure
d
33  Congenital Airway Malformations
ventilation over a long period of time. The place-
ment of intratracheal stents (Fig. 33.11) is used
selectively in patients with severely problematic
tracheomalacia or bronchomalacia that is unre-
sponsive to nonoperative therapy or not suitable
for surgical treatment (Gerber and Holinger
2003; Grillo 1994; Hysinger and Panitch 2016;
Monnier 2011; Serio et al. 2014; Shieh et al.
2017; Sztano et al. 2016; Torre et al. 2012; Valerie
et al. 2005; Varela et al. 2018; Vinograd et al.
1987).
Major complications associated with this
approach can occur, including stent collapse, dis-
lodgement, or rarely, stent erosion into the great
vessels. Additionally, stent removal can cause
tracheal tearing or major hemorrhage.
Fig. 33.11  Balloon-expandable metallic stent placed in
the trachea with a severe tracheomalacia associated with a
type III esophageal atresia
a b
trachea
esophagus
Fig. 33.12 (a–c) Posterior tracheopexy. Esophagus is
mobilized to the right and the posterior tracheal wall is
stitched to the spine. (a) Normal anatomy, (b) severe pos-
423
A posterior tracheopexy is proposed for severe
tracheomalacia. During the procedure, the esoph-
agus is mobilized to the right, and the posterior
wall of the trachea is attached to the spine with
multiples stitches under endoscopic visualization
(Fig. 33.12a–c).
Regarding severe bronchomalacia, a three-
dimensional external airway splint (Fig. 33.13a, b)
has been proposed to reinforce the bronchial
wall, avoiding collapse due to the aorta or pulmo-
nary artery.
33.4.5 Esophageal Lung
Isolated bronchial connection between the esopha-
gus and the airway is extremely rare and occurs
more frequently in females (2:1) (Fig.  33.14).
Associated cardiac, genitourinary, vertebral, and
diaphragmatic anomalies are common. This mal-
formation is thought to develop from a supernu-
merary lung bud arising from the esophagus. Most
commonly, a lower lobe is aerated by this ectopic
bronchus, but an entire main bronchus and lung
may be involved. As in pulmonary sequestration
anomalies, the pulmonary vasculature in this
anomaly may be abnormal, with the arterial supply
coming off the aorta and venous drainage going
into either the systemic or pulmonary veins.
Because of inadequate bronchial drainage,
children usually have recurrent pulmonary infec-
c
trachea
esophagus
terior tracheal collapse, (c) posterior tracheopexy (cour-
tesy Diana Romero MD, Colombia)
424 P. Varela and R. Azizkhan

Fig. 33.13 (a) Severe left bronchial collapse (courtesy Gaston Bellia MD, Buenos Aires, Argentina). (b) 3D airway
splint is placed in a severe left bronchial collapse (Courtesy Gaston Bellia MD, Buenos Aires, Argentina)

on the affected segment of the lung, collapse,

consolidation, cavitation, and cyst formation
within the pulmonary parenchyma are commonly
seen. The diagnosis is confirmed by contrast
studies of the esophagus, though occasionally,
false-negative results occur. Excision of the
abnormal lung and closure of the bronchoesopha-
geal fistula are the treatments of choice. Prognosis
depends on early diagnosis and treatment and the
severity of associated anomalies.

Fig. 33.14  Esophageal lung. A pulmonary lobe (red 33.4.6 Tracheobronchial Biliary
arrow) is connected to the esophagus (yellow arrow) Fistula

tion; however, occasionally esophageal bronchus Congenital tracheobronchial biliary fistulae are
is not discovered until adolescence or adulthood. extremely rare and may arise from the distal tra-
Although radiographic findings differ depending chea or either mainstem bronchus. All children
33  Congenital Airway Malformations
with this anomaly have significant respiratory
problems but the cardinal symptom is bile-stained
sputum. The diagnosis is established either by
bronchoscopy or endoscopic retrograde cholan-
giopancreatography (ERCP). Surgical division of
the fistulous tract is the only effective therapy for
this malformation.
33.4.7 Subglottic Hemangioma
33.4.7.1 Pathogenesis
Hemangiomas of infancy are benign congenital
vascular tumors. These tumors are characterized
by vascular endothelium that undergoes a phase
of growth followed by slow, spontaneous involu-
tion that occurs over several years and is gener-
ally complete by the first decade of life.
These tumors most commonly present cutane-
ously but can occur in any anatomic site. No
symptoms occur during the first weeks of life.
They usually become evident between 2 and
4  months of age. Progressive resolution occurs
between 12 and 18  months with a threefold
female preponderance.
Almost all lesions that occur within the tra-
cheobronchial tree are in the subglottis. The natu-
ral history of subglottic hemangiomas (SGH)
a b
Fig. 33.15 (a) Endoscopic view of a subglottic hemangioma in a patient with multiple cutaneous infantile hemangio-
mas (b)
425
generally mirrors that of cutaneous lesions, and
more than 50% of patients with a subglottic hem-
angioma also have cutaneous hemangiomas. As
the hemangioma undergoes proliferation, pro-
gressive deterioration of the airway usually
occurs. Presenting symptoms include biphasic
stridor with retractions. The degree of obstruc-
tion varies and can be exacerbated by certain
positions or crying, both of which increase
venous pressure and lead to vascular engorge-
ment. When airway obstruction is severe, apnea,
cyanosis, and “dying spells” may occur.
Diagnosis is based on medical history and find-
ings on airway endoscopy. Lesions are typically
asymmetric and may be covered by a normal
smooth mucosa (Fig. 33.15). Because of the risk
of hemorrhage, biopsy is not advised.
Most patients require treatment and many
treatment modalities are often combined. These
include medical treatments, endoscopic resec-
tion, and open surgical procedures. Depending
on both the severity of the obstruction and the
expertise of involved clinicians, early symptoms
are managed with intralesional or systemic ste-
roids. The use of systemic steroids is not recom-
mended for longer than 3 weeks if the symptoms
don not improve. The use of propranolol seems to
be an effective medical treatment that allows
426
complete and fast resolution with evident
improvement of respiratory symptoms. Propanol
is our first treatment approach in most of the
SGH. Some surgeons advocate endoscopic treat-
ment with laser fulguration or the use of microde-
brider. Surgical options consider translaryngeal
resection, whereas others place a tracheotomy
below the lesion, with the expectation of removal
following involution of the hemangioma.
33.4.8 Bronchogenic Cyst
Bronchogenic cysts result from abnormal budding
of the bronchial tree in which a portion of the lung
bud develops independently. The cyst walls fre-
quently contain cartilage and are lined with ciliated
columnar epithelium. These lesions tend to enlarge,
thus causing airway obstruction. Infants with bron-
chogenic cysts most commonly present with respi-
ratory distress. They also may have cough, chest
pain, or wheezing. Although a plain chest X-ray
may suggest the presence of a bronchogenic cyst, a
CT scan and barium esophagram are useful in con-
firming this diagnosis. Thoracoscopic resection is
curative (Fig. 33.16a, b).
33.4.9 Bronchial Atresia
Localized bronchial atresia is a rare anomaly in
which the atretic bronchus obstructs the flow of
secretions and air from the distal lung to the main
a b
Fig. 33.16  Bronchogenic cyst: (a) right hilar lesion seen on CT scan, (b) operative photograph demonstrating the
bronchogenic cyst in situ
P. Varela and R. Azizkhan
tracheobronchial tree. This condition may simu-
late lobar emphysema or a mediastinal mass. At
birth, the obstructed lung retains fluid, but eventu-
ally the affected lobe or segment becomes hyper-
aerated as air enters through the pores of Kohn.
Patients accumulate secretions proximal to the
atresia and a mucocele forms. Emphysema of the
segment may cause compression of the normal
lung tissue and may be associated with wheezing
and stridor. Plain chest radiographs often demon-
strate a hilar mass with radiating solid channels
surrounded by hyperaerated lung. A CT chest scan
may indicate a cystic central mucocele and can
help differentiate bronchial atresia from a bron-
chogenic cyst or lobar emphysema. Although chil-
dren with bronchial atresia may be asymptomatic
for long intervals, they are at risk of serious pul-
monary infection when entrapped secretions
become infected. When this anomaly is identified,
resection is both indicated and curative (Azizkhan
2005; Puri and Höllwarth 2006).
33.4.10 Bronchial Agenesis
Congenital absence of a bronchus occurs more
commonly than tracheal agenesis, and in contrast
to tracheal agenesis, this anomaly is compatible
with life. Several possible anatomic variants have
been described, including lobar, bronchial, and
parenchymal agenesis. Specifically, there may be
complete agenesis of the lung and its bronchus
and blood supply; aplasia, in which there is a
33  Congenital Airway Malformations
rudimentary bronchus and absent lung; or hypo-
plasia, in which there is a rudimentary bronchus
and hypoplastic lung. As is common in children
with airway malformations, children with bron-
chial agenesis also may have other congenital
anomalies of the skeletal, cardiovascular, gastro-
intestinal, and genitourinary systems. Diagnosis
is confirmed by chest radiographs and airway
endoscopy. The majority of patients do not
require surgical intervention. Nevertheless, these
patients are important to identify as bronchial or
lobar atresia may mimic other airway anomalies
such as bronchial stenosis, extraluminal airway
obstruction by tumors or masses, or complete
intraluminal obstruction in which there is no aer-
ation distal to the obstruction.
33.4.11 Bronchial Stenosis
Congenital bronchial stenosis is extremely rare,
with reported cases caused by compressive vas-
cular, cardiac, and congenital cystic lesions or
soft tissue cartilaginous stenoses. Symptoms and
treatment depend on the anatomic location of the
lesion and its severity. In contrast, acquired
bronchial stenosis is more common and is a
major cause of morbidity and mortality in infants
who require prolonged intubation and respira-
tory support. Most such cases can be managed
endoscopically.
33.5 Conclusions
Congenital airway malformations include a wide
number of anomalies with a broad spectrum of
symptoms. Treatments of all these conditions are
usually performed by an airway team, working in
collaboration with critical and intensive care,
neonatologists, radiologists, and anesthetists.
References
Austin J, Ali T (2003) Tracheomalacia and bronchoma-
lacia in children: pathophysiology, assessment, treat-
ment and anaesthesia management. Paediatr Anaesth
13:3–11
427
Azizkhan RG (2005) Subglottic airway. In: Oldham
KT, Colombani PM, Foglia RP, Skinner MA (eds)
Principles and practice of pediatric surgery, Chapter
59, vol 2nd. Lippincott Williams & Wilkins,
Philadelphia, PA
Backer CL, Mavroudis C, Dunham ME, Holinger LD
(1997) Reoperation after pericardial patch tracheo-
plasty. J Pediatr Surg 32:1108–1111. discussion
1111–1102
Backer CL, Mavroudis C, Gerber ME, Holinger LD (2001)
Tracheal surgery in children: an 18-year review of four
techniques. Eur J Cardiothorac Surg 19:777–784
Bando K, Turrentine MW, Sun K, Sharp TG, Matt B,
Karmazyn B et al (1996) Anterior pericardial tracheo-
plasty for congenital tracheal stenosis: intermediate to
long-term outcomes. Ann Thorac Surg 62:981–989
Benjamin B, Inglis A (1989) Minor congenital laryngeal
clefts: diagnosis and classification. Ann Otol Rhinol
Laryngol 98:417–420
Bennett EC, Holinger LD (2003) Congenital malforma-
tions of the trachea and bronchi. In: Bluestone CD,
Stool SE, Alper CM, Arjmand EM et al (eds) Pediatric
otolaryngology, Chapter 84, vol 2nd, 4th edn. WB
Saunders, Philadelphia, PA
Cotton RT (2003) Posterior laryngeal cleft. In: Ziegler
MM, Azizkhan RG, Weber TR (eds) Operative pediat-
ric surgery, Chapter 26. McGraw Hill, New York
Davies MR, Cywes S (1978) The flaccid trachea and tra-
cheoesophageal congenital anomalies. J Pediatr Surg
13:363–367
DeMarcantonio MA, Hart CK, Yang CJ, Tabangin M,
Rutter MJ, Bryant R et al (2017) Slide tracheoplasty
outcomes in children with congenital pulmonary mal-
formations. Laryngoscope 127:1283–1287
Eriksen C, Zwillenberg D, Robinson N (1990) Diagnosis
and management of cleft larynx. Literature review and
case report. Ann Otol Rhinol Laryngol 99:703–708
Evans JN (1985) Management of the cleft larynx and
tracheoesophageal clefts. Ann Otol Rhinol Laryngol
94:627–630
Fiore AC, Brown JW, Weber TR, Turrentine MW (2005)
Surgical treatment of pulmonary artery sling and tra-
cheal stenosis. Ann Thorac Surg 79:38–46
Furman RH, Backer CL, Dunham ME, Donaldson J,
Mavroudis C, Holinger LD (1999) The use of bal-
loon-expandable metallic stents in the treatment of
pediatric tracheomalacia and bronchomalacia. Arch
Otolaryngol Head Neck Surg 125:203–207
Gerber ME, Holinger LD (2003) Congenital laryn-
geal anomalies. In: Bluestone CD, Stool SE, Alper
CM, Arjmand EM et al (eds) Pediatric otolaryngol-
ogy, Chapter 83, vol 2nd, 4th edn. WB Saunders,
Philadelphia, PA
Gorostidi F, Reinhard A, Monnier P, Sandu K (2016)
External bioresorbable airway rigidification to treat
refractory localized tracheomalacia. Laryngoscope
126:2605–2610
Grillo HC (1994) Slide tracheoplasty for long-segment
congenital tracheal stenosis. Ann Thorac Surg.
58:613–619
428
Grillo HC, Wright CD, Vlahakes GJ, MacGillivray TE
(2002) Management of congenital tracheal steno-
sis by means of slide tracheoplasty or resection and
reconstruction, with long-term follow-up of growth
after slide tracheoplasty. J Thorac Cardiovasc Surg
123:145–152
Gustafson LM, Hartley BE, Liu JH et al (2000) Single-
stage laryngotracheal reconstruction in children: a
review of 200 cases. Otolaryngol Head Neck Surg
123:430–434
Herrera P, Caldarone C, Forte V, Campisi P, Holtby H,
Chait P et al (2007) The current state of congenital
tracheal stenosis. Pediatr Surg Int 23:1033–1044
Hewitt RJ, Butler CR, Maughan EF, Elliott MJ (2016)
Congenital tracheobronchial stenosis. Semin Pediatr
Surg 25:144–149
Hysinger EB (2018) Laryngomalacia, Tracheomalacia
and Bronchomalacia. Curr Probl Pediatr Adolesc
Health Care. 48(4):113–118
Hysinger EB, Panitch HB (2016) Paediatric tracheomala-
cia. Paediatr Respir Rev 17:9–15
Kluth D, Steding G, Seidl W (1987) The embryology of
foregut malformations. J Pediatr Surg 22:389–393
Lim FY, Crombleholme TM, Hedrick HL, Flake AW et
al (2003) Congenital high airway obstruction syn-
drome: natural history and management. J Pediatr
Surg 38:940–945
Monnier P (2011) Pediatric airway surgery. Springer, New
York. https://doi.org/10.1007/978-3-642-13535-4
Ngerncham M, Lee EY, Zurakowski D, Tracy DA,
Jennings R (2015) Tracheobronchomalacia in pedi-
atric patients with esophageal atresia: comparison of
diagnostic laryngoscopy/bronchoscopy and dynamic
airway multidetector computed tomography. J Pediatr
Surg 50:402–407
Puri P, Höllwarth ME (eds) (2006) Pediatric surgery.
Springer, Berlin
Rahbar R, Rouillon I, Roger G, Lin A, Nuss RC,
Denoyelle F et al (2006) The presentation and man-
agement of laryngeal cleft: a 10-year experience. Arch
Otolaryngol Head Neck Surg 132:1335–1341
Rahbar R, Chen JL, Rosen RL, Lowry KC, Simon DM,
Perez JA et al (2009) Endoscopic repair of laryngeal
cleft type I and type II: when and why? Laryngoscope
119:1797–1802
P. Varela and R. Azizkhan
Rutter MJ (2006) Evaluation and management of upper
airway disorders in children. Sem Pediatr Surg
15:116–123
Rutter MJ, Cotton RT, Azizkhan RG, Manning PB (2003)
Slide tracheoplasty for the management of complete
tracheal rings. J Pediatr Surg 38:928–934
Sandu K, Monnier P (2006) Endoscopic laryngotra-
cheal cleft repair without tracheotomy or intubation.
Laryngoscope 116:630–634
Serio P, Fainardi V, Leone R, Baggi R, Grisotto L, Biggeri
A et al (2014) Tracheobronchial obstruction: follow-
up study of 100 children treated with airway stenting.
Eur J Cardiothorac Surg 45:e100–e109
Shieh HF, Smithers CJ, Hamilton TE, Zurakowski D,
Rhein LM, Manfredi MA et al (2017) Posterior tra-
cheopexy for severe tracheomalacia. J Pediatr Surg
52:951–955
Speggiorin S, Torre M, Roebuck DJ, McLaren CA, Elliott
MJ (2012) A new morphologic classification of con-
genital tracheobronchial stenosis. Ann Thorac Surg
93:958–961
Sztano B, Kiss G, Marai K, Racz G, Szegesdi I, Racz K et al
(2016) Biodegradable airway stents in infants - potential
life-threatening pitfalls. Int J Pediatr Otorhinolaryngol
91:86–89
Tazuke Y, Okuyama H, Uehara S, Ueno T, Nara K,
Yamanaka H et al (2015) Long-term outcomes of
four patients with tracheal agenesis who underwent
airway and esophageal reconstruction. J Pediatr Surg
50:2009–2011
Torre M, Carlucci M, Speggiorin S, Elliott MJ (2012)
Aortopexy for the treatment of tracheomalacia in chil-
dren: review of the literature. Ital J Pediatr 38:62
Valerie EP, Durrant AC, Forte V, Wales P, Chait P, Kim
PC (2005) A decade of using intraluminal tracheal/
bronchial stents in the management of tracheomalacia
and/or bronchomalacia: is it better than aortopexy? J
Pediatr Surg 40:904–907
Varela P, Torre M, Schweiger C, Nakamura H (2018)
Congenital tracheal malformations. Pediatr Surg
Int 34(7):701–713. https://doi.org/10.1007/
s00383-018-4291-8
Vinograd I, Filler RM, Bahoric A (1987) Long-term func-
tional results of prosthetic airway splinting in tracheo-
malacia and bronchomalacia. J Pediatr Surg 22:38–41
 Mediastinal Masses in Children
Maria Molina and Israel Fernandez-Pineda
34.1 Introduction
The mediastinum is the body compartment lim-
ited by the thoracic inlet superiorly, the sternum
anteriorly, the spine dorsally, the diaphragm infe-
riorly and both pleural cavities laterally. In the
paediatric population, identification of a medias-
tinal mass includes a wide variety of congenital
and neoplastic lesions. Mediastinal tumours are
very rare and represent a small amount of all the
cancers in the paediatric population. Mediastinal
masses are usually classified depending on their
location in the mediastinum. Thus, three different
compartments can be identified: anterior, middle
and posterior mediastinum. Approximately 40%
of mediastinal masses are benign, and 60% are
malignant. Nearly 40% of all of them occur in
children <2 years of age.
Special mention to first edition chapter author,
Dr. S.J. Schochat, for his work in previous papers and his
permission to use materials of the previous edition.
M. Molina · I. Fernandez-Pineda (*)
Department of Pediatric Surgery, Virgen del Rocio
Children’s Hospital, Sevilla, Spain
e-mail: israel.fernandez.sspa@juntadeandalucia.es
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_34
34
34.2 Diagnosis
There are two factors to be taken into account
when evaluating a child with a mediastinal mass:
its location within the mediastinum and the
patient’s age.
Regarding the location of the mass, it is impor-
tant to differentiate the three compartments we
mentioned earlier: anterior, middle and posterior
mediastinum (Fig.  34.1). Anterior mediastinum
is the space found between the sternum (in front),
the trachea, great vessels and pericardium (poste-
riorly) and the pleurae and lungs on each side. It
includes the thymus, anterior mediastinal lymph
nodes and, in some patients, the substernal exten-
sion of the thyroid and parathyroid, although that
is very rare. Masses within this compartment are
usually teratomas, thymic abnormalities such as
Anterior Posterior
mediastinum mediastinum
Middle
mediastinum
Fig. 34.1  Compartments of the mediastinum
429
430
ectopic thymus/hyperplasia/cysts/tumours, lym-
phatic malformations and lymphomas. Middle
mediastinum contains the trachea, mainstem
bronchi, the heart, and the great vessels as well as
the hilar lymph nodes. Masses in this area include
lymphomas, bronchogenic cysts, granulomatous
infections within the mediastinal lymph nodes
and congenital vascular lesions such as double
aortic arch. The posterior component is located
between the posterior pericardium and the verte-
bral column, and it accommodates the aorta, the
thoracic oesophagus and the sympathetic nerve
chains. Posterior mediastinal lesions include the
tumours of neurogenic origin, foregut cysts, the
undifferentiated sarcomas and in some cases
extralobar sequestrations. A list of the mediasti-
nal masses, which can be found within each com-
partment, can be seen in Table 34.1.
Secondly, a patient’s age can help surgeons to
diagnose the mediastinal mass due to a tendency
of certain lesions to appear in certain groups of
age. For example, in children under 2  years of
age, the most common mediastinal mass within
the posterior mediastinum is the neuroblastoma,
in the anterior mediastinum is thymic hyperplasia
and in the middle mediastinum is the b­ ronchogenic
cyst. In children older than 2 years, lymphomas
are the most common mediastinal masses identi-
fied. Hodgkin’s disease is more commonly found
in 13-year-old patients, whereas the mean age of
Table 34.1  List of the mediastinal masses which can be
found within each compartment
Compartment Benign Malignant
Anterior Teratoma Non-Hodgkin
Thymic hyperplasia lymphoma
Thymic cysts (rare) Hodgkin
lymphoma
Thymic tumours
(rare)
Malignant germ
cell tumours
Endocrine (rare)
Middle Lymphatic Non-Hodgkin
malformation lymphoma
Foregut cysts and Hodgkin
duplications lymphoma
Carcinoma (rare)
Posterior Ganglioneuroma Neuroblastoma
Foregut cysts (rare)
M. Molina and I. Fernandez-Pineda
children who present with non-Hodgkin’s lym-
phoma is 11 years old.
They can present with a variety of symptoms,
from completely asymptomatic and noted inci-
dentally in an x-ray performed for another reason
unrelated to the mediastinal mass, to acute respi-
ratory distress, shortness of breath, stridor, cough,
cervical adenopathy or even superior vena cava
syndrome and Horner syndrome, both of them
being rare. It is more common for patients under
2 years old to present with signs of airway com-
pression due to the smaller, softer, more pliable
tracheobronchial tree. Also, their mediastinum is
free so large mediastinal masses can cause a sig-
nificant shift of the mediastinum with compro-
mise of the contralateral hemithorax. Older
children may present with fever, cough and short-
ness of breath. Horner syndrome is associated
with neurogenic tumours in the posterior medias-
tinum. Neurological symptoms may appear in
cases of dumbbell tumours involving the spinal
cord. Also, secretion of catecholamines, alpha-­
fetoprotein (AFP) or gonadotropins may induce
changes and signs that help uncover the tumour.
A number of complementary tests can be per-
formed when evaluating a child with a mediasti-
nal mass. They include:
• Imaging techniques
All imaging complementary tests are thought
to help with identifying the characteristics and ori-
gin of the mass within the mediastinum; delineat-
ing its extension; and providing a differential
diagnosis. The first diagnostic technique to be per-
formed continues to be the chest x-ray, in both
posteroanterior and lateral projections, as it helps
to localize the position of the mass within the
mediastinum. Also, there are some abnormalities
that can lead to certain diagnosis. For example,
vertebral anomalies could be associated with neu-
roenteric variety of enterogenous cysts. Also, cal-
cifications, if present in the posterior mediastinum,
would suggest the presence of a neuroblastoma,
whereas if present in the anterior mediastinum
could be the sign of presence of teratomas. If
enterogenous or bronchogenic cysts are suspected,
a barium swallow study should be performed. In
34  Mediastinal Masses in Children 431

infants, chest ultrasound could be performed and 34.3 Anterior Mediastinum

help diagnose a thymic lesion such as hyperplasia.
Echocardiography should be performed to delin- The most common anterior mediastinal mass is
eate the heart and great vessels if lesions of these
thymic hyperplasia (Fig.  34.2), with the typical
structures are suspected. As a computed tomogra- ‘sail sign’ on routine chest x-ray as the principal
phy (CT) scan, along with angio-CT, offers differ-
tool for diagnosis. If no clear diagnosis can be
ent multiplanar reconstructions, it should be reached with a chest x-ray, then an US can be
performed when preparing for tumour resection performed, as it usually differentiates thymic
and when no clear diagnosis has been reached hyperplasia from other mediastinal masses. If
with other techniques. Magnetic resonance imag- necessary, CT or MRI can be performed.
ing (MRI) provides a better view of the tumour’s Treatment consists of observation and sometimes
anatomy and its r­elations with the mediastinum, corticosteroids. Rare cases are associated with
especially in masses of vascular origin. Also, MRI
compression of vital structures which indicates
should be considered in cases of posterior medias-
surgical resection. Thymic cysts and malignant
tinal masses in order to identify the possible intra-
tumours are extremely rare in children and only
spinous extension of the tumour (the so-called 20 thymomas have been reported in childhood.
dumbbell tumours). Other imaging tests, such as Benign teratomas are the most frequent ante-
fluorodeoxyglucose-­ positron emission tomogra- rior mediastinal neoplasm seen in children under
phy (18FDG-PET), combined with CT or MRI, or 2  years old. They can be diagnosed prenatally,
metaiodobenzylguanidine (MIBG), would help but their diagnosis is usually after birth due to
with diagnosis if lymphoma or neuroblastoma are respiratory distress or cardiac insufficiency. They
suspected, respectively. are usually encapsulated, containing cystic and
solid areas with calcifications and fat tissue. No
• Bone marrow and lymph node biopsy invasion of adjacent structures is normally seen.
Surgical resection through a posterolateral inci-
Lymph node biopsy is especially important in sion is usually the preferred approach.
children with middle mediastinum masses with Malignant germ cell tumours, with endoder-
suspected lymphoma. Bone marrow aspiration mal sinus or yolk sac component, are usually
and biopsy should also be performed if non-­ seen in older children and adolescents. They may
Hodgkin’s lymphoma is suspected. act as functioning tumours and thus can induce
precocious puberty. As they usually elevate AFP,
• Skin test: complement fixation both this and HCG levels should be obtained at
diagnosis, even in the follow-up as a therapy-­
This is especially important in children with response marker. In cases where there is a large
middle mediastinum masses to rule out granulo-
matous infections.
Anterior mediastinum
• Serum markers (alpha-fetoprotein or AFP,
human chorionic gonadotropin or HCG)
< 2 years > 2 years
Benign teratoma Malignant germ
This is especially important in children with Thymic hyperplasia cell tumor
Cystic hygroma Thymoma
anterior mediastinum masses with suspected Lymphoma
malignant germ cell tumours.

• Urinary catecholamines

This is especially important in children with

posterior mediastinum masses with suspected
neuroblastoma, both at diagnosis and follow-up. Fig. 34.2  Anterior mediastinum
432
mass, primary radical resection may not be indi-
cated. Their therapy includes a multidisciplinary
team with paediatric oncologists, surgeons and
radiation therapists. A biopsy rather than partial
resection is indicated, followed by chemotherapy
and delayed primary excision.
Lymphatic malformations, when appearing in
the anterior mediastinum, are usually associated
with a cervical or axillary extension, which
makes the diagnosis more obvious. US would
help with the characterization of the mass, along
with MRI. The main risk with the lymphatic mal-
formations in the anterior mediastinum is when
their size is so big that may compromise the air-
way, especially in newborns. Ex utero intrapar-
tum treatment (EXIT procedure) involves partial
delivery of the foetus with the foetal-placental
circulation maintained, so that the foetal airway
obstruction can be managed with placental sup-
port. This may include direct laryngoscopy-­
bronchoscopy intubation, tracheostomy or
surgical intervention of the mass (which is rare).
When approaching them, it is important to bear
in mind that they are benign, they do not involute
and that radical surgery may compromise impor-
tant structures. Thus, in cases of being symptom-
atic sirolimus and sclerosis is a preferred
approach, the latter only being feasible if macro-
cystic type is seen.
Regarding vascular tumours, most hemangio-
mas within the anterior mediastinum are in conti-
nuity with cervicofacial components. They tend to
regress over time but in some cases antiangiogenic
treatment with propranolol associated with cortico-
steroids could be necessary if airway involvement
is seen. Large kaposiform hemangioendotheliomas
could be dangerous due to the association with
Kasabach-Merritt phenomenon, in which platelet
trapping increases risk of haemorrhage. Treatment
with sirolimus and/or vincristine should be started
with close monitoring and adjusting.
34.4 Middle Mediastinum
Bronchogenic cysts are the most frequent middle
mediastinum masses in children under 2  years
old (Fig. 34.3). As they are located in the subcari-
M. Molina and I. Fernandez-Pineda
Middle mediastinum
< 2 years > 2 years
Bronchogenic cyst Lymphoma
Granuloma
Fig. 34.3  Middle mediastinum
nal region, they are usually associated with an
expiratory stridor due to a partial obstruction of
the lower trachea during expiration. Their diag-
nosis commonly needs a barium swallow study,
which shows a displacement of the oesophagus.
Surgical resection through a posterolateral thora-
cotomy is the preferred approach, and in some
cases a small portion of the cyst may be left
attached to the membranous trachea if needed.
Lymphomas, both Hodgkin and non-­
Hodgkin’s, are the most frequent tumours involv-
ing the middle mediastinum, and they are also the
most common mediastinal masses encountered in
children over 2  years old. If suspected, lymph
node biopsy and bone marrow aspiration and
biopsy should be performed as the initial diagnos-
tic work-up. Lymph node biopsy should be made
from a cervical, supraclavicular or other periph-
eral lymphadenopathy if present, or if not, through
mediastinoscopy, mediastinotomy or CT-guided
needle biopsy approach. In some cases, large
mediastinal masses can cause respiratory distress
and the patient may need a course of steroids or
even local radiotherapy before biopsy has been
done to reduce the risk of respiratory decompen-
sation in the anaesthesia induction, although it is
not ideal, as it can make the subsequent diagnosis
difficult. Samples should be placed in saline so
that immunologic surface marker studies can be
performed; hence, a proper classification is made,
and adequate therapy is started.
Sometimes, the origin of the mediastinal mass
in the middle mediastinum comes from a granu-
lomatous infection, which enlarges the paratra-
34  Mediastinal Masses in Children
cheal, subcarinal and hilar lymph nodes. In these
cases, skin tests and complement fixation titres
should give the diagnosis. In places where
­histoplasmosis is endemic, such as the Midwest
of the USA, the azygous node can be abnormally
enlarged and diagnosis may need a mediastinos-
copy, mediastinotomy or, rarely, a thoracotomy.
34.5 Posterior Mediastinum
The most common posterior mediastinum mass is
the mediastinal neuroblastoma (Fig.  34.4). It is
also the most common mass found in newborns.
These neuroblastomas behave differently than the
abdominal neuroblastoma. They are usually local-
ized and have more favourable biologic features,
with favourable prognosis after resection. If meta-
static disease is found on diagnosis, prognosis
continues to be discouraging. They can present as
asymptomatic masses, in which the mass is found
on an x-ray performed for another complaint, or
they can present with a variety of symptoms.
Children can have neurological symptoms and
can appear in cases of dumbbell extension of the
neuroblastoma due to spinal cord compression,
such as sudden paraplegia. Respiratory distress
due to external compression of the trachea can
also be found. Moreover, high thoracic or cervi-
cal neuroblastomas can present with Horner
syndrome (unilateral ptosis, myosis, anhydro-
sis) and, in cases of large masses within the
mediastinum, superior vena cava syndrome can
appear. Paraneoplastic symptoms are uncommon.
Posterior mediastinum
< 2 years > 2 years
Neuroblastoma Ganglioneuroma
Enterogenous cyst Sarcoma
Fig. 34.4  Posterior mediastinum
433
Treatment of these neuroblastomas is total exci-
sion, if at all possible. In cases of large masses,
multidisciplinary teams should assess patient’s
history. Radical chest wall resection is not indi-
cated, and biopsy followed by chemotherapy and
delayed primary excision is the appropriate
approach. For biopsy, percutaneous core needle
biopsy or even thoracoscopic approach tends to
be enough. Laminectomy could be necessary in
cases of spinal compression.
Other possible diagnosis when a posterior
mediastinum mass is found is enterogenous cysts.
They are very rare, and they can be found both in
the middle and posterior mediastinum. Their
symptoms may vary, from dysphagia if attached
to the oesophagus to peptic ulceration and bleed-
ing if containing gastric mucosa or even perfora-
tion. Large cysts may have abdominal extensions
and communicate with an intestinal duplication.
Neurenteric cysts are a rare variant that communi-
cates with the meninges through an intraspinous
component. Characteristically, they associate
with deformities of the lower cervical and upper
thoracic spine, which can be seen on chest x-ray.
Patients present with large mediastinal masses,
respiratory distress and in rare cases neurological
symptoms. In these cases, MRI should be per-
formed before surgery in order to have a better
view of the intraspinous cystic component.
Treatment requires resection through a thoracot-
omy (or in the later years thoracoscopy), excision
of the cyst making sure no secreting mucosa is
being left on the field and closing any communi-
cation with the airway. Special attention should be
paid to the ligation of the communication between
the thoracic and infraspinous component, to pre-
vent spinal fluid leak and meningitis.
34.6 Anaesthetic Management
of Children
with a Mediastinal Mass
The characteristics of the paediatric airway and
mediastinum give an increased risk of respiratory
compromise with airway collapse on the induc-
tion of general anaesthesia in children with large
mediastinal masses.
434
Small children and infants have a small com-
pressible airway, associated with significant air-
way resistance with even the smallest degree of
narrowing. Also, their mediastinum is not fixed,
which provokes mediastinal shifts more easily
with compression of the tracheobronchial tree,
superior vena cava or right ventricular outflow
tract. Cardiac output may also be diminished due
to pressure on the great vessels. Induction of
anaesthesia is associated with a decrease in the
functional residual capacity, decrease in lung
capacity and increase in lung retractile force.
These alterations are extenuated with the addi-
tion of paralysis. Narrowing of the trachea will
also occur when the patient changes from sponta-
neous to positive-pressure ventilation.
All of these factors may lead to respiratory
decompensation if not prevented.
The most important factor in preventing these
complications is anticipation of a possible airway
difficulty. Special care should be taken when eval-
uating the radiological findings both anatomical
relations between the mass and the mediastinal
structures and the mediastinal disposition, espe-
cially with CT examination. Some known risk
factors for anaesthetic complications are a 50%
reduction of predicted tracheal cross-­ sectional
area on the preoperative CT scan, mainstem bron-
chus and great vessel compression, orthopnoea
and upper body oedema. Some patients may need
to undergo preoperative radiation therapy, corti-
costeroids or chemotherapy prior to even the
biopsy is performed to reduce the risk of cardiore-
spiratory complications. Once the preoperative
evaluation is finished, the anaesthetic of choice
depends on the procedure to be performed (e.g.
needle biopsy versus thoracotomy). In some
cases, such as incisional or needle biopsies, local
anaesthesia could be enough, especially with
older children. Other cases with benign lesions
may benefit from one-lung anaesthesia with
placement of the endotracheal tube beyond the
obstruction has been found to be helpful, and
occasionally, ventilation through a rigidbroncho-
scope is necessary. Cardiopulmonary bypass and
extracorporeal membrane oxygenation is not usu-
ally necessary for these patients.
In summary, a high index of suspicion, metic-
ulous preoperative evaluation and a multidisci-
M. Molina and I. Fernandez-Pineda
plinary action plan decided upon by the surgeons,
anaesthesiologist, radiation therapist, haematol-
ogy/oncologist and pathologist can usually advert
the potential catastrophe associated with general
anaesthesia in children with critical mediastinal
masses.
34.7 Operative Technique
for Removal of Mediastinal
Neuroblastoma
Surgical approach of a mediastinal neuroblas-
toma can be via a posterolateral thoracotomy and
via video-assisted thoracic surgery (VATS).
VATS results in shorter operative times, less post-­
operative pain, shorter hospital stay and excellent
cosmetic results, although it requires skilful sur-
geons and it precludes palpation of the lesions.
Surgical resection is facilitated by single-lung
ventilation and can be done regardless of the sur-
gical approach selected.
The lung is retracted medially to reveal the
tumour covered with pleura and arising from the
sympathetic trunk; an assessment of the lymph
node involvement in the tumour vicinity is per-
formed. The pleura is incised around the tumour
approximately 1  cm from it and the fascia and
pleura mobilized toward the tumour. A plane of
dissection can usually be developed superficial to
the endothoracic fascia. The tumour is now mobi-
lized from the ribs by sharp dissection, and inter-
costal vessels entering the tumour will need
division. If the tumour extends far enough anteri-
orly, the azygos vein on the right side will need to
be ligated. Care is taken to avoid damage to the
first thoracic nerve passing laterally across the
first rib to join the brachial plexus. The superior
intercostal artery normally descends between the
first nerve and the sympathetic trunk. Other inter-
costal nerves may be sacrificed if they are inti-
mate with the tumour.
Dissection of the anterior part of the tumour
may be more difficult for its association with the
main structures in the superior mediastinum. The
most common relation is with the oesophagus
and the vagus nerve, thus an orogastric tube may
help with the dissection and help prevent any
injury. In cases of large masses, trachea may also
34  Mediastinal Masses in Children
be involved. Moreover, in tumours protruding
on the left side, special care should be taken if
closely associated is found with the thoracic
duct and the arch of the aorta with subclavian
and carotid branches. The thoracic duct should
also be identified and protected in lower
tumours on the right. It should now be possible
to dissect the tumour off the vertebral bodies,
along with any extension into the intravertebral
foramen. Titanium clips may prove useful to
control haemorrhage in small vessels, and these
will not interfere with subsequent CT scanning.
They may also be used as markers if all tumour
is not excised and radiation therapy is being
considered.
Any suspiciously involved lymph nodes
locally should be taken for biopsy (staging).
The chest is closed after leaving a chest drain.
34.8 Conclusion
Differential diagnosis in mediastinal masses is
based on the anatomical mediastinum compart-
ment they are located (anterior, middle or poste-
rior) and the patient’s age. Although mediastinal
tumours represent a small percentage of all pae-
diatric cancer, more than half of all mediastinal
masses are malignant. Knowledge of the most
frequent masses found in each compartment is
mandatory for an optimal management of these
patients. The understanding of the paediatric air-
way and mediastinum anatomy and their vulner-
ability under general anaesthesia is the most
important factor for preventing any respiratory
decompensation when dealing with paediatric
patients with mediastinal masses.
Bibliography
Anghelescu DL, Burgoyne LL, Liu T et  al (2007)
Clinical and diagnostic imaging findings predict
anesthetic complications in children presenting with
malignant mediastinal masses. Paediatr Anaesth
17(11):1090–1098
435
Cass DL, Olutoye OO, Cassady CI, Zamora IJ, Ivey RT,
Ayres NA, Olutoye OA, Lee TC (2013) EXIT-to-­
resection for fetuses with large lung masses and per-
sistent mediastinal compression near birth. J Pediatr
Surg 48:138–144
Grosfeld JL, Skinner MA, Rescorla FJ et  al (1994)
Mediastinal tumors in children: experience with 196
cases. Ann Surg Oncol 1:121–127
Hammer GB (2004) Anaesthetic management for the child
with a mediastinal mass. Pediatr Anesth 14:95–97
Johnson SM (2017) Mediastinal tumors. In: Mattei P (ed)
Fundamentals of pediatric surgery. Springer, Cham,
pp 359–365
Kang CH, Kim YT, Jeon SH, Sung SW, Kim JH (2007)
Surgical treatment of malignant mediastinal neuro-
genic tumors in children. Eur J Cardiothorac Surg
31(4):725–730
Lee EY (2009) Evaluation of non-vascular mediasti-
nal masses in infants and children: an evidence-­
based practical approach. Pediatr Radiol 39(Suppl
2):S184–S190
Malek MM, Mollen KP, Kane TD, Shah SR, Irwin C
(2010) Thoracic neuroblastoma: a retrospective
review of our institutional experience with comparison
of the thoracoscopic and open approaches to resection.
J Pediatr Surg 45:1622–1626
Patel R, Lim RP, Saric M, Nayar A, Babb J, Ettel M, Axel
L, Srichai MB (2016) Diagnostic performance of car-
diac magnetic resonance imaging and echocardiogra-
phy in evaluation of cardiac and paracardiac masses.
Am J Cardiol 117(1):135–140
Perger L, Lee EY, Shamberger RC (2008) Management of
children and adolescents with a critical airway due to
compression by an anterior mediastinal mass. J Pediatr
Surg 43(911):1990–1997
Saenz NC (1999) Posterior mediastinal neurogenic tumors
in infants and children. Semin Pediatr Surg 8:78–84
Shochat SJ (2009) Mediastinal masses in the children. In:
Puri P, Höllwarth M (eds) Pediatric surgery. Springer,
Heidelberg
Tovar JA, Martinez L (2019) Mediastinal tumors. In:
Holcomb GW III, Murphy JP, Peter SD (eds) Holcomb
and Ashcraft’s pediatric surgery, 7th edn. Elsevier,
New York, pp 403–421
Warnke RA, Link MP (1983) Identification and signifi-
cance of cell markers in leukemia and lymphoma.
Annu Rev Med 34:117–131
Williams HJ, Alton HM (2003) Imaging of paediatric
mediastinal abnormalities. Pediatr Respir Rev 4:55–66
Wright CD (2009) Mediastinal tumors and cysts in the
pediatric population. Thorac Surg Clin 19(1):47–
61. vi
Yaris N, Nas U, Cobanoglu U et  al (2006) Brief report:
thymic carcinoma in children. Pediatr Blood Cancer
47:224–227
 Pleural Effusion and Empyema
35
Michael Singh and Dakshesh Parikh

35.1 Introduction 35.2 Etiology and Pathogenesis

The volume and composition of pleural liquid is The etiology and pathogenesis of pleural effusion
governed by a number of interacting mechanisms can be subdivided into fetal congenital and
that contribute to maintain the conditions for an effi- acquired (Table 35.1).
cient mechanical coupling between the chest wall
and lung. In normal physiological conditions, a thin Table 35.1  Etiology and pathogenesis
layer of pleural fluid is maintained between the
Antenatal and congenital pleural effusion
pleural surfaces. The composition of pleural fluid is    Fetal hydrothorax
essentially that of a plasma filtrate. Pleural fluid    Congenital chylothorax
accumulating in excess amounts results in an effu-    Thoracic lymphangiomatosis
Acquired pleural effusion
sion and its nature may vary. It may contain blood,
Hemothorax
chyle, lymph, transudate, exudates, or pus. The dis-    Chylothorax
tinction between a transudate and an exudate is     Thoracic duct damage
mainly based on the different protein and cellular     Post-cardiac surgery in the absence of
thoracic duct damage
content. In general, transudates have a protein con-
   Hydrothorax
tent less than 30 g/L, glucose content >2.2 mMol/L,     Iatrogenic
and a low total cell count. In contrast, the exudate       Central venous line malposition
contains higher protein content >30 g/L, low glu-       Transdiaphragmatic hydrothorax
        Peritoneal dialysis
cose <2.2 mMol/L, and high cellular content mainly
        Ventriculoperitoneal shunt
leukocytes in infective effusion. Lymph has a pro-    None iatrogenic
tein content that is intermediate between transudate     Transdiaphragmatic, e.g., hepatic ascites,
and exudate and has a variable number of lympho- peritonitis
    Congestive heart disease
cytes. Chyle is lymph draining from the intestinal
    Nephrotic syndrome
tract. It is rich in lymphocytes and contains chylo-     Malignancy (non-Hodgkin lymphoma)
microns depending on diet containing long-chain Pleural exudate and empyema
triglycerides. Pleural empyema means the accumu-    Traumatic or iatrogenic
   Nontraumatic
lation of pus in the pleural cavity (Light 2006).
    Hydatid disease
    Bacterial infections
M. Singh · D. Parikh (*)
Department of Paediatric Surgery, Birmingham
Children’s Hospital, Birmingham, UK
e-mail: michael.singh2@nhs.net

© Springer Nature Switzerland AG 2023 437

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_35
438
35.2.1 Fetal and Congenital Pleural
Effusion
The presence of antenatal pleural effusions or fetal
hydrothorax is estimated to be 1 in 10,000–15,000
pregnancies. Pleural effusion in the postnatal
period may be congenital or acquired and can be as
a result of intrauterine, perinatal, or postnatal infec-
tions. The main causes of intrauterine infections
are cytomegalovirus, toxoplasma, rubella, and ade-
novirus (Klam et  al. 2005; Picone et  al. 2004;
Harigopal and Shaw 2009). Pleural effusions are
associated with chromosomal and congenital
anomalies. Fetal hydrothorax may be secondary to
a mediastinal tumor, congenital cystic adenoma-
toid lung malformation, pulmonary sequestration,
infection, or chromosomal anomaly or ABO blood
group incompatibility. Primary/idiopathic pleural
effusion has been called congenital primary chylo-
thorax although no chylomicrons can be detected
antenatally or immediately postnatally. The term
“antenatal chylothorax” is considered controversial
for pleural effusions containing a high lymphocyte
count in fetal pleural aspirate. Congenital lym-
phatic abnormalities include lymphangiectasia,
intrathoracic lymphatic malformations, genetic
syndromes (Noonan syndrome), congenital heart
disease, and mediastinal tumors (Klam et al. 2005).
Many large antenatal pleural effusions are man-
aged by the fetal ­intervention using pleuro-amni-
otic shunting to improve the outcome of the
resultant pulmonary hypoplasia (Klam et al. 2005;
Picone et al. 2004; Harigopal and Shaw 2009).
35.2.2 Acquired/Pathological Pleural
Effusion
35.2.2.1 Hemothorax
A hemothorax is usually traumatic in origin in
children and adolescents and mostly associated
with a pneumothorax. In many cases, it is iatro-
genic in nature, related to the insertion of central
venous lines and intercostal drains.
M. Singh and D. Parikh
35.2.2.2 Chylothorax
Chylothorax following thoracic operations
with an accidental damage or disruption of tho-
racic duct or accessory lymphatics is well-
known. Increased central venous pressure,
exceeding the pressure in the thoracic duct, can
result in leakage of the chyle into the pleural
space. The thrombosed superior vena cava fol-
lowing prolonged central venous lines, or any
other reason for thrombosis of the superior
vena cava, can result in chylothorax. The inci-
dence is on the rise from 1 to nearly 5%. The
increased complexity of the procedures and
possibly the earlier introduction of feeding
after surgery are held responsible for the
increased incidence. Procedures that predis-
pose to increased systemic venous pressure are
particularly at risk of causing chylothorax.
Chylothorax following the repair of congenital
diaphragmatic hernia is also commonly
encountered (Lasko and Langer 2009; Chan
et al. 2006).
35.2.2.3 Hydrothorax
Iatrogenic Hydrothorax
Hydrothorax in children can result from iatro-
genic or pathological causes.
Several case reports of pleural effusion in
children related to central venous catheter
insertion have been described in the immedi-
ate postoperative period or as a later complica-
tion. Hydrothorax, especially on the right, may
be caused by transdiaphragmatic fluid migra-
tion from the excessive fluid accumulation
from the abdominal cavity, e.g., in the context
of peritoneal dialysis or ventriculoperitoneal
shunting or other causes of ascites. The main
mechanism seems to be transport through open
communications between the abdomen and the
chest. Also, urine may enter the chest either
retroperitoneally or through the diaphragm in
case of urinary ascites (Lasko and Langer
2009; Chan et al. 2006).
35  Pleural Effusion and Empyema
Pathological/Secondary Hydrothorax
Pathological causes of hydrothorax include con-
gestive heart disease, nephrotic syndrome, and
malignancy, especially non-Hodgkin lymphoma,
and other metastatic diseases (Lasko and Langer
2009; Chan et al. 2006).
35.2.2.4 Pleural Exudate
and Empyema
Pleural exudate and empyema may be secondary
to trauma to the chest, esophageal foreign body
perforation, or surgical causes, such as leaking
bronchial stump and esophageal anastomotic
leak. Also, infradiaphragmatic pathology may be
responsible, e.g., retained gallstones after chole-
cystectomy, pancreatitis, and hollow viscus per-
foration, e.g., appendiceal perforation with
peritonitis (Parikh 2009).
Parapneumonic effusion, meaning effusion
between both leaves of the pleura due to pneu-
monia, once a rare complication of bacterial
pneumonia in children, has become increas-
ingly common worldwide. Vaccination with a
heptavalent pneumococcal polysaccharide con-
jugate vaccine has not decreased the relative
overall incidence of Streptococcus pneumoniae
infections but the relative incidence of the sub-
types has changed. Streptococcus pneumoniae
continues to be the main causative organism for
pediatric empyema. In contrast, the incidence
of Haemophilus influenzae empyema has
decreased. Epstein-Barr virus has been impli-
cated in the pathogenesis of pleural effusion. In
developing countries, Staphylococcus aureus is
an important causative infectious agent. In
areas endemic for tuberculosis, the incidence of
tuberculous effusion and secondary bacterial
empyema is high. Echinococcus granulosus is
another causative agent for empyema in
endemic areas (Parikh 2009; Byington et  al.
2002).
439
35.3 Pathology
The pleural fluid depends on the underlying
pathology. It can be bloody, as in the case of hemo-
thorax, tuberculosis, and malignancy, or milky as
in the case of chylothorax. Chylous fluid looks
relatively clear in the absence of enteral feeding.
Clear fluid is seen when there is leakage of dialysis
fluid or cerebrospinal fluid in the case of a ven-
triculoperitoneal drain. The pleural space may be
occupied with lymphatic cysts in lymphangioma-
tosis. In case of an exudate, the fluid looks cloudy.
In the event of an empyema, classically, three
stages are distinguished. The development of
these pathological stages is usually overlapping
and is often described as continuum. The timing
of development of pathology is empirical and
varies depending on the virulence of the organ-
ism, resistance/immunity of the host, and inter-
vention instituted such as antibiotics or inadequate
drainage (Parikh 2009).
1. Exudative phase: thin pleural fluid with rela-
tively few cells, lasting 24–72 h.
2. Fibrinopurulent phase: accumulation of
fibrinous material and loss of lung mobility,
lasting 7–10 days.
3. Organizing phase: formation of a fibrous
pleural peal.
35.4 Clinical Features
Pleural effusion interferes with lung expansion
and can cause mediastinal shift and respiratory
failure. When present antenatally, it can cause
mediastinal shift, fetal heart failure, hydrops, and
pulmonary hypoplasia. In the absence of infection,
respiratory symptoms such as tachypnea, dyspnea,
and orthopnea prevail. There is dullness of the
chest on the affected side with diminished breath
440 M. Singh and D. Parikh

sounds. The dullness may vary depending on the 35.5 Imaging

most dependent position of the fluid accumulation.
The heart sounds may be displaced to the other 35.5.1 Plain Chest X-Ray in AP
side. Depending on the underlying condition, the Position
effusion may be bilateral, e.g., in congestive heart
failure. The presence of peripheral edema may Blunting of the costo-phrenic and cardio-phrenic
suggest congestive heart failure, cirrhosis, or angles is an early sign of pleural effusion
nephrotic syndrome. Lymphadenopathy may point (Fig.  35.1). More extensive effusions cause a
in the direction of a malignant lymphoma. widening of the interlobar spaces. In the absence
Painless ascites may be the result of hepatic dis- of organization, the fluid shifts depending on the
ease, peritoneal dialysis, or ventriculoperitoneal position of the patient. The heart may be dis-
shunting. In the event of inflammation, general placed to the contralateral side (Figs. 35.2a, b and
signs, such as fever, tachycardia, and lethargy, are 35.3a). When the effusion becomes organized,
present in combination with severe local pain, exag- the fluid collection or collections become static.
gerating during respiration. The pain is often referred
to the back and the s­ houlder. The patient lies on the
affected side with curvature of the spine towards the
affected side. Respiration is fast and shallow. Local
signs differ depending on the stage of the disease. In
early pleurisy, there may be a pleural friction rub.
Breath sounds are diminished, especially in the
lower part of the chest when the patient is in a head
up position. Dullness of the chest may vary depend-
ing on the most dependent position of the fluid but
becomes fixed when loculation occurs. There may
be bronchial breathing unless the effusion is very
extensive. Heart sounds may be displaced to the
contralateral side. Abdominal pain and tenderness
may be present in pleural empyema and referred
pain and tenderness may mimic an intra-abdominal Fig. 35.1  Cardiac failure in a patient with Duchenne AP
etiology (Parikh 2009). chest X-ray

a b

Fig. 35.2  Congenital chylothorax (a) AP chest X-ray. the right. (b) After drainage, the mediastinal shift and
There is pleural effusion on the left. The lung is collapsed effusion have resolved
and the heart and the mediastinum are displaced toward
35  Pleural Effusion and Empyema
a b
c d
Fig. 35.3 (a) AP chest X-ray showing a parapneumonic
effusion with displacement of the mediastinum to the
right. (b) Ultrasound shows a collapsed lung with a fluid
loculation below the lung. (c) CT scan showing turbid
There is no need for a routine lateral X-ray
(Parikh 2009).
35.5.2 Ultrasound
Ultrasound examination gives an idea regarding
the amount of the pleural effusion and the degree
of organization (Fig.  35.3b). When organized,
there is septation and no shifting of the fluid with
changing patient position. Ultrasound is also
helpful during the process of identification and
actual puncture of a fluid collection to be drained
(Balfour-Lynn et al. 2005).
441
fluid in the pleural space around the lung and in between
the lung lobes. There is pleural thickening and air suggest-
ing pulmonary necrosis. Posteriorly, a chest drain is seen.
(d) Complete resolution after 4 months
35.5.3 CT and MR Scanning
CT or MRI scanning of the chest is performed to
obtain a more detailed and complete information
prior to intervention (Fig. 35.3c). Collections of
fluid or pus are identified. Thickening of the
pleura points in the direction of inflammation.
Hypodense areas or air pockets within the paren-
chyma suggest pulmonary necrosis and lung
abscess. A pneumothorax suggests pulmonary
necrosis with bronchopleural fistula. Underlying
congenital or malignant pathology may be identi-
fied (Parikh 2009).
442
35.6 Diagnosis and Differential
Diagnosis
From the history and clinical examination, a good
idea of the pathology can be obtained in the major-
ity of patients. The presence or absence of pain
allows a rough separation of pleural transudate and
exudate. Imaging provides information about the
localization of the fluid, the degree of organiza-
tion, and of potential underlying conditions. Blood
tests are helpful in differentiating between infec-
tious and noninfectious causes and may provide
clues for malignant disease, e.g., malignant lym-
phoma. Blood cultures are essential when parap-
neumonic disease is suspected (Parikh 2009).
Analysis of the pleural fluid is essential in the
treatment of pleural effusion and empyema. It
allows for differentiation between transudate,
exudate, and chyle and for bacteriological exami-
nation. Several biological markers have been
used to differentiate a transudate from an exudate
in children but protein content in pleural fluid
above 30 g/L performs as well as any other indi-
vidual test. If examined in patients under diuretic
therapy, however, a serum minus pleural fluid
albumin greater than 12 g/L rules out an exuda-
tive origin. Adenosine deaminase is raised in
tuberculosis but may also be raised in pleural
malignancies, lymphoproliferative diseases, and
rheumatoid disease. Pleural biopsy, especially
targeted biopsies, obtained during thoracoscopy,
improves the diagnostic yield in tuberculosis and
other rare causes.
Chylothorax is diagnosed when the fluid
retrieved contains more than 1.1 mmol/L triglyc-
erides with oral fat intake and has a total cell
count equal or above 1000 cells per μL. Cytology
may be of help in diagnosing malignant pleural
effusion in children, especially when associated
with lymphoma or leukemia. Bacterial analysis
of pleural fluid and blood should be performed
but the yield is rather low and reported to be
about 40–60%.
35.7 Management
The management of pleural effusion depends on
the underlying pathology.
M. Singh and D. Parikh
35.7.1 Fetal Hydrothorax
Fetal hydrothorax may resolve spontaneously
as the pregnancy progresses. Expectant man-
agement of stable and resolving effusions is
therefore appropriate. In contrast, in patients
developing hydramnios or hydrops, thoraco-­
amniotic shunting should be offered in the
absence of underlying conditions, such as chro-
mosomal anomalies (Picone et al. 2004).
35.7.2 Congenital Chylothorax
Congenital chylothorax should be treated conser-
vatively with mechanical ventilation, pleural fluid
evacuation, and total parenteral nutrition and sub-
sequent introduction of medium-chain triglycer-
ide (MCT) feeds for 6–12  weeks. Octreotide
treatment has been suggested, but its efficacy
remains difficult to prove especially against the
background that most cases of c­ ongenital hydro-
thorax disappear during conservative treatment
(Lasko and Langer 2009; Chan et al. 2006).
35.7.3 Pleural Effusion in the Context
of Lymphangiomatosis
Treatment is largely symptomatic and by no
means evidence-based. Several treatment modal-
ities have been suggested, e.g., feeding with
medium-chain triglycerides, octreotide, inter-
feron alpha, irradiation, surgical resection, liga-
tion of the thoracic duct, pleuroperitoneal
shunting, and pleurodesis, either surgically or
with talc or other sclerosing agents. Long-term
therapy with sirolimus is an emerging therapy,
which can stabilize the lymphatic fluid leakage
(Adams et al. 2016).
35.7.4 Chylothorax After Thoracic/
Cardiac Surgery
The treatment is basically conservative unless the
thoracic duct was surgically damaged, in which
case ligation of the leaking duct should be done
(Chan et al. 2006).
35  Pleural Effusion and Empyema
35.7.5 Pleural Effusion Related
to Central Venous Catheters
Withdrawal of the catheter and evacuation of the
fluid is all that is required.
35.7.6 Iatrogenic Hydrothorax
In the case of peritoneal dialysis or ventriculo-
peritoneal shunting, therapy may be directed at
closing potential diaphragmatic holes and divert-
ing the VP shunting to ventriculo-arterial shunt
and peritoneal dialysis to hemodialysis for a
period. If not feasible, pleurodesis may be an
option.
35.7.7 Non-iatrogenic Hydrothorax
In non-iatrogenic hydrothorax, treatment should
be directed at the underlying pathology. Puncture
or drainage may increase patient comfort dramat-
ically, but rapid aspiration and re-expansion of
the lung may lead to pulmonary edema. If treat-
ment of the underlying disease is not possible,
e.g., incurable malignant disease, pleurodesis
may be an option.
35.7.8 Empyema
Underlying causes should of course be treated.
For the treatment of the empyema itself, there are
consensus guidelines from the American
Association for Thoracic Surgery (Shen et  al.
2017). Treatment options include:
1. Antibiotics
2. Thoracocentesis
3. Tube thoracostomy
4. Intrapleural fibrinolytics
5. Debridement of empyema
(a) Thoracoscopically
(b) Mini-thoracotomy
6. Decortication: thoracotomy
443
There is no discussion on the prompt initiation
of antibiotic treatment. The type of antibiotics
administered should depend on age, underlying
medical condition, and suspected etiological
agent. Of course, the antibiotic regimen should
be adjusted according to culture results (Parikh
2009; Byington et  al. 2002; Balfour-Lynn et  al.
2005; Godfrey et al. 2019).
Repeat thoracocentesis in pediatric patients is
rarely needed, except after worsening of the dis-
ease severity (Haggie et al. 2020). When there is
pleural infection, a drain should be inserted for
patient comfort; this should be done under gen-
eral anesthesia. Moreover, general anesthesia
offers the opportunity for a thoracoscopy allow-
ing for inspection of the whole pleural cavity,
aspiration of all fluid, and controlled chest tube
placement. There is no evidence that large drains
are better than smaller drains. Drains should not
be inserted posteriorly as the intercostal spaces
are narrower there (Parikh 2009). Real-time
ultrasound-guided pigtail catheter chest drain
under radiological control is recommended
(Lewis et al. 2018). There is high-grade evidence
to support the utility of fibrinolytic therapy and
chest drainage, in the management of parapneu-
monic effusion or pleural empyema (Lasko and
Langer 2009; Chan et  al. 2006; Altmann et  al.
2019). Early, primary thoracoscopic debridement
of the empyema peel can produce results superior
to fibrinolytic therapy (Parikh 2009; Byington
et  al. 2002; Balfour-Lynn et  al. 2005; Bouros
et  al. 2006; Eastham et  al. 2004; Ismail et  al.
2008). Recently, simultaneous clinical observa-
tion of bronchoscopic alveolar lavage combined
with video-assisted thoracoscopy (VAT) has been
recommended (Yue et al. 2019).
VAT surgery or fibrinolytic therapy is most
often sufficient and a more invasive surgical
treatment is rarely indicated (Redden et al. 2017).
However, when medical treatment fails (persis-
tent pyrexia, recurrent effusion, pneumothorax),
surgical debridement and decortication of the
empyema peel are indicated. This is best achieved
via a thoracotomy. Surgical treatment of empy-
ema consists of decortication, which means
444
a b
Fig. 35.4  Thoracoscopic pictures of an organized pleural
empyema. (a) The pleural space is covered with and
bridged by fibrinous sheets. The thoracic wall itself is not
removal of fibrin sheets and fibrous covering of
the pleurae and the insertion of a drain
(Fig. 35.4a–c). The loculated effusion is debrided.
The decortication of the fibrous peel from the vis-
ceral pleura is performed. Any areas of leaking
pulmonary necrosis or lung abscess with a bron-
chopleural fistula is debrided and the abscess
cavity is filled with a serratus anterior muscle flap
(Parikh 2009; Byington et al. 2002; Balfour-Lynn
et  al. 2005; Bouros et  al. 2006; Eastham et  al.
2004; Peters et al. 2020; Jester et al. 2012).
35.8 Complications
The outcome after pleural effusion and empy-
ema differs depending on potential underlying
conditions.
The outcome of fetal hydrothorax is related to
potential underlying conditions and the progres-
sion to hydramnios and hydrops fetalis.
Pleuroamniotic shunting has its own complica-
tions, such as premature rupture of membranes
and possible trauma to tissues caused by the
shunt. Postnatal chylothorax usually has a good
outcome. The prognosis of lymphangiomatosis
and especially of the Gorham-Stout syndrome is
not good.
The outcome of chylothorax after heart sur-
gery is usually good but depends on the underly-
ing cardiac condition. Long-lasting leakage of
lymph and chyle predisposes for hypoprotein-
emia, lymphocytopenia, and infectious compli-
M. Singh and D. Parikh
c
visible. (b) The fibro-purulent material is being sucked
away. (c) Fibrinous sheets and strands are grasped and
removed
cations. Parenteral substitution of the losses and
parenteral feeding are part of the treatment but
have its own complications, such as sepsis,
thrombosis, and parenteral nutrition-induced
liver disease.
Pleural effusion due to central venous line
malposition may cause a life-threatening condi-
tion but if resolved in time, the prognosis is good.
The prognosis of an effusion due to malignant
disease depends on the prognosis of the underly-
ing disease. The outcome in parapneumonic effu-
sion, and even empyema, is usually excellent
without long-term morbidity (de Benedictis et al.
2019) (Fig. 35.3d).
35.9 Conclusion
There is a wide differential diagnosis when a
patient presents with a pleural effusion from fetal
life to childhood. A detailed history and examina-
tion can often give sufficient information as to the
etiology. Pleural fluid aspiration, chest X-rays,
ultrasounds, and CT scans can guide appropriate
investigations and treatment modalities. The
more uncommon and complex etiologies (fetal,
congenital chylothorax, lymphangiomas, and
empyema) require a multidisciplinary approach
to their management.
Acknowledgement  We would like to thank and acknowl-
edge Professor Klaas Bax for the materials we used from
his previous chapter while writing this updated version of
the chapter.
35  Pleural Effusion and Empyema 445

References Ismail M, Nachira D, Meacci E, Ferretti GM, Swierzy M,

Englisch JP et al (2008) Uniportal video-assisted tho-
racic surgery in the treatment of pleural empyema. J
Adams D, Tenor C, Hammil AM et  al (2016) Safety of
Thorac Dis 10(31):S3696–S3703
sirolimus in the treatment of complicated vascular
Jester I, Nijran A, Singh M, Parikh DH (2012) Surgical
anomalies. Pediatrics 137(2):e20153257
management of broncho-pleural fistula in paediatric
Altmann ES, Crossingham I, Wilson S, Davies HR (2019)
empyema and necrotizing pneumonia: efficacy of the
Intra-pleural fibrinolytic therapy versus placebo, or a
serratus anterior muscle digitation flap. J Pediatr Surg
different fibrinolytic agent, in the treatment of adult
47:1358–1362
parapneumonic effusions and empyema. Cochrane
Klam S, Bigras J-L, Hudon L (2005) Predicting out-
Database Syst Rev (10):CD002312
come in primary fetal hydrothorax. Fetal Diagn Ther
Balfour-Lynn IM, Abrahamson E, Cohen G et al (2005)
20:366–370
BTS guidelines for the management of pleural infec-
Lasko D, Langer JC (2009) Chylothorax. In: Parikh DH,
tion in children. Thorax 60:1–21
DCG C, Auldist AW, Rothenberg SS (eds) Pediatric
Bouros D, Tzouvelekis A, Antoniou K, Heffner JE (2006)
thoracic surgery. Springer, London, pp 573–579
Intrapleural fibrinolytic therapy for pleural infection.
Lewis MR, Micic TA, Doull IJM, Evans A (2018) Real-­
Pulm Pharmacol Ther 20(6):616–626. https://doi.
time ultrasound-guided pigtail catheter chest drain
org/10.1016/j.pupt.2006.08.001
for complicated parapneumonic effusion and empy-
Byington CL, Spencer LY, Johnson TA et  al (2002) An
ema in children  – 16-year, single-centre experi-
epidemiological investigation of a sustained high
ence of radiologically placed drains. Pediatr Radiol
rate of pediatric parapneumonic empyema: risk fac-
48(10):1410–1416
tors and microbiological associations. Clin Infect Dis
Light RW (2006) The undiagnosed pleural effusion. Clin
34:434–440
Chest Med 27:309–319
Chan S, Lau W, Wong WHS, Cheng L, Chau AKT,
Parikh DH (2009) Empyema thoracis. In: Parikh DH,
Cheung Y (2006) Chylothorax in children after con-
DCG C, Auldist AW, Rothenberg SS (eds) Pediatric
genital heart surgery. Ann Thorac Surg 82:1650–1657
thoracic surgery. Springer, London, pp 109–129
de Benedictis FM, Carloni I, Osimani P, Cobellis G,
Peters RT, Parikh DH, Singh M (2020) Thoracoscopic
Martino A, Lanza C et al (2019) Prospective evalua-
debridement of empyema thoracis. J Pediatr Surg
tion of lung function in children with parapneumonic
55:2187–2190
empyema. Pediatr Pulmonol 54(4):421–427
Picone O, Benachi A, Mandelbrot L et  al (2004)
Eastham KM, Freeman R, Kearns AM et  al (2004)
Thoracoamniotic shunting for fetal pleural effusions
Clinical features, aetiology and outcome of empy-
with hydrops. Am J Obstet Gynecol 191:2047–2050
ema in children in the north east of England. Thorax
Redden MD, Chin TY, van Driel ML (2017) Surgical ver-
59:522–525
sus non-surgical management for pleural empyema.
Godfrey MS, Bramley KT, Detterbeck F (2019) Medical
Cochrane Database Syst Rev 3(3):CD010651
and surgical management of empyema. Semin Respir
Shen KR, Bribriesco A, Crabtree T, Denlinger C, Eby
Crit Care Med 40(3):361–374
J, Eiken P et  al (2017) The American Association
Haggie S, Gunasekera H, Pandit C, Selvadurai H,
for Thoracic Surgery consensus guidelines for the
Robinson P, Fitzgerald DA (2020) Paediatric empy-
management of empyema. J Thorac Cardiovasc Surg
ema: worsening disease severity and challenges iden-
153(6):e129–e146
tifying patients at increased risk of repeat intervention.
Yue F, Yang Z, Yang F, Liu Y, Zhao L, Chen Z, Gao F
Arch Dis Child 105(9):886–890
(2019) Clinical observation of bronchoscopy alveo-
Harigopal S, Shaw NJ (2009) Respiratory disorders of
lar lavage combined with thoracoscopy in the treat-
the newborn. In: Parikh DH, DCG C, Auldist AW,
ment of empyema in children. Medicine (Baltimore)
Rothenberg SS (eds) Pediatric thoracic surgery.
98(52):e18528
Springer, London, pp 411–426
 Congenital Malformations
of the Lung
Ali A. Mokdad, David M. Gourlay,
and Keith T. Oldham
36.1 Introduction and Historical
Overview
Congenital malformations of the lung are an
impressively diverse group of lesions that affect
the airway and lungs. The first reported clinical
description of these lesions dates back to 1687.
The spectrum of presentation of these malforma-
tions ranges from antenatal diagnosis of malfor-
mations to symptomatic presentation in adulthood
and ranges from asymptomatic to immediately
life threatening. The most common congenital
malformations, congenital pulmonary airway
malformations (CPAM), pulmonary sequestra-
tions, congenital lobar emphysema (CLE), and
bronchogenic cysts, are themselves unique in
their presentation and are considered here.
Optimal management of congenital lung malfor-
mations requires an understanding by the surgeon
of the embryology of lung development, respira-
tory physiology, and anatomy.
36.2 Incidence
Congenital malformations of the lung are uncom-
mon, affecting 1  in 10,000–35,000 live births.
The incidence may be as high as 1 in 2000; how-
A. A. Mokdad · D. M. Gourlay · K. T. Oldham (*)
Division of Pediatric Surgery, Medical College of
Wisconsin, Milwaukee, WI, USA
e-mail: DGourlay@chw.org; KOldham@chw.org
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_36
36
ever, more lesions are detected prenatally with
improving prenatal imaging techniques. In the
USA, there is no gender, racial, or ethnic predi-
lection. Among congenital malformations of the
lung, congenital pulmonary airway malforma-
tions are most commonly encountered.
36.3 Anatomy
The normal lobar and segmental anatomy of
mature lungs and their relationship to adjacent
mediastinal structures are demonstrated in
Figs.  36.1 and 36.2. A detailed review of lung
anatomy is outside the scope of this text, but sev-
eral anatomic points of critical relevance in pedi-
atric thoracic surgery deserve mention. In the
term infant, the carina is located at the fourth or
fifth thoracic vertebral body. The right main stem
bronchus is larger in diameter, shorter in length,
and more vertical in direction than the left main
stem bronchus. In the infant and child, the hilum
of either lung lies directly beneath the fifth inter-
costal space on the lateral chest wall. Thus, a tho-
racotomy incision through this space provides
optimal exposure to the hilum for pulmonary
resection.
447
448 A. A. Mokdad et al.

Fig. 36.1 Development of the trachea and bronchi. JE.  Skandalakis’ Surgical Anatomy: The Embryologic
Anterior and lateral views at the beginning of the seventh and Anatomic Basis of Modern Surgery. Paschalidis
week. The main lobar bronchial tree has developed and Medical Publications, Athens 2004; page 270)
divided into five orders (Modified from Skandalakis
36  Congenital Malformations of the Lung
Fig. 36.2  The major
bronchopulmonary
segments (Modified
from Hood. Techniques
in General Thoracic
Surgery. WB Saunders
Company, Philadelphia
1985; page 84)
36.4 Embryology
Embryologic development of the lung is central
to the understanding of congenital lung malfor-
mations. The primordium of the respiratory sys-
tem develops around the third week of gestation
from a diverticulum of the ventral wall of the
foregut, immediately distal to the caudal pharyn-
geal complex. The tracheoesophageal septum is
complete by the fourth week of gestation, sepa-
rating the developing respiratory system and
foregut. By the end of the sixth week of gesta-
tion, the trachea has divided into the right and left
main stem bronchi (Fig. 36.3). Progressive bron-
chial branching ensues and is complete by the
16th week of gestation. By that time, the pleural
peritoneal canal has closed, and the diaphragm
has formed. The pulmonary arterial vasculature
449
follows the bronchial branching pattern but lags
slightly behind. Notably, in the early stages of
bronchial development, bronchial buds are per-
fused by a systemic capillary plexus derived from
the primitive aorta. As the pulmonary artery
develops, this capillary plexus slowly regresses.
Between 16 and 24 weeks of gestation, the blood-­
air interface matures as respiratory bronchioles
develop and the capillary network expands.
Following 24 weeks of gestation, primitive alve-
oli develop, and surfactant production ensues.
Both the number and size of the alveoli increase
during this time. Many of the congenital pulmo-
nary malformations considered here are evident
by this time. Considering the rapid expansion of
pulmonary parenchyma during this period of
development, one can understand how a congeni-
tal pulmonary lesion can have a severe negative
450
Fig. 36.3  Key anatomic
relations of the
structures of the
pulmonary hilum
(Modified from Oldham
KT. Principles and
Practice of Pediatric
Surgery. Lippincott
Williams & Wilkins,
Philadelphia 2005; page
952) Bronchial Tree
(anterior view)
Stem
Bronchus
impact on both fetal lung development and post-
natal respiratory physiology. A few weeks before
birth, the alveoli develop their definitive mor-
phology as secondary septations appear and pul-
monary angiogenesis is maximized. This process
continues into the first few years of life; over
85% of the alveoli are formed after birth. Because
of the continued development of the alveoli after
birth, pulmonary resection in infants and children
is extraordinarily well tolerated, with little long-­
term functional compromise in respiratory physi-
ology (Mullassery and Smith 2015).
36.5 Etiopathogenesis
The phases of normal lung development
described above have been widely accepted, but
the mechanisms that underlie congenital lung
malformations remain unknown. Clements and
Warner proposed a unifying mechanism to
explain the development of these lung lesion
(Clements and Warner 1987). Their theory
describes an insult to the developing lung such
A. A. Mokdad et al.
Trachea
Esophagus
Fundus
Stomach
that the nature, severity, and timing of the insult
differentially impact the component structures of
the lung, namely, the tracheobronchial tree, the
lung parenchyma, and/or the vasculature as fol-
lows. A severe insult, which completely impedes
the development of a bronchus and its pulmo-
nary blood supply, may cause agenesis of the
entire affected lung, lobe, or segment. A minor
insult causing only a local disruption but not
impacting the overall development of the lung
may result in a local lesion such as a broncho-
genic cyst. An insult that disrupts the develop-
ment of the bronchus without affecting the
pulmonary vasculature may develop into a con-
genital pulmonary airway malformation. A simi-
lar focal insult occurring late in the development
process may impact the bronchus primarily lead-
ing to congenital lung emphysema. An insult that
preferentially interrupts the development of the
pulmonary arteries, thereby allowing the affected
lung to retain its primitive systemic vasculature,
may result in an intralobular sequestration or a
hybrid malformation which comprises overlap-
ping sequestration and CPAM morphologies.
36  Congenital Malformations of the Lung
More contemporary literature underscores the
role of bronchial obstruction in the pathogenesis
of congenital lung malformations (Langston
2003). This is partly driven by the high preva-
lence of occult atretic bronchi in congenital lung
malformations. The exact mechanism by which
bronchial obstruction leads to various lung mal-
formations has not been defined; however, the
degree, level, and timing of the obstruction likely
play a prominent role in explaining the spectrum
of lung anomalies.
36.6 Classification, Pathology,
and Pathophysiology
The classification of congenital lung malforma-
tions remains in flux owing to the diverse nature
of these lesions and a poor understanding of their
pathogenesis. As a result, terminology is variable
and inconsistent across specialty fields. In this
chapter, we ascribe to nomenclature familiar to
pediatric surgeons. As such, we discuss congeni-
tal pulmonary airway malformations, broncho-
pulmonary sequestrations, congenital lobar
emphysema, and bronchogenic cysts (Fowler and
Gould 2015).
36.6.1 Congenital Pulmonary Airway
Malformation
Congenital pulmonary airway malformations,
previously termed congenital cystic adenomatoid
malformations (CCAM), are the most commonly
identified lesions among congenital lung malfor-
mations and may represent up to 60% of lesions
in some series. They are benign hamartomatous
lesions, characterized by overgrowth of terminal
bronchioles in an adenomatoid pattern. They
communicate with the tracheobronchial tree and
have normal pulmonary arterial and venous blood
supply. Involvement is usually unilobar with a
slight predilection for the lower lobes, with right
and left sides equally affected. Bilateral and mul-
tifocal disease are rare. Grossly, CPAMs have
both cystic and solid components. Gross cystic
changes, however, are not universal to all CPAMs.
451
Histologically, they are composed of disorga-
nized cysts lined with ciliated cuboidal or colum-
nar epithelium.
CPAMs are classified by Stocker into five sub-
types (0–4) that reflect different levels of involve-
ment of the tracheobronchial tree (Fowler and
Gould 2015). Type 0 is a proximal bronchial
anomaly, whereas type 4 indicates a peripheral
location in the lung. Lesions between those two
locations are denoted types 1, 2, and 3. The latter
types have varying cyst sizes; type 1 have large
cysts (2–10  cm) and are most prevalent, type 2
have small cysts (0.5–2 cm), and type 3 have very
small cysts (< 0.5 cm). The Stocker classification
has been challenged, as it does not account for
overlap among lesions and does not adequately
classify atypical forms. A more encompassing
classification that is based on pathologic features
was described by Langston (2003); however, the
Stocker system of classification remains well
established in the surgical literature.
36.6.2 Bronchopulmonary
Sequestration
Bronchopulmonary sequestrations are com-
posed of nonfunctional pulmonary tissue that
lacks communication with the tracheobronchial
tree of the normal lung. They account for less
than 10% of congenital lung malformations.
Histologically, they demonstrate immature lung
development, often resembling more peripheral
lung parenchyma. Bronchopulmonary seques-
trations can be intralobar (75%) or extralobar
(25%), depending on whether the lesion is
invested by pleura of the adjacent normal lobe
or by its own pleura, respectively. Typically, the
arterial blood supply is anomalous and most
commonly arises from the descending thoracic
aorta. Venous drainage for intralobar sequestra-
tions is usually via the pulmonary veins, whereas
venous drainage for extralobar sequestrations is
typically systemic via either the azygous or
hemiazygous veins, inferior vena cava, or
directly into the atrium. Intralobar sequestra-
tions occur most commonly on the left side, par-
ticularly involving the left lower lobe. Extralobar
452
sequestrations can be found anywhere from the
neck to below the diaphragm. They are most
commonly located in the left hemithorax in the
posterior mediastinum. Sequestration in the
upper lobe occurs in only 10–15% of cases, and
bilateral sequestrations are infrequent. These
lesions have a well-defined vascular pedicle;
evidence of other structures within the pedicle
raises the suspicion for a communication with
the digestive system which is infrequent but
well described. Notably, extralobar sequestra-
tions are associated with other congenital anom-
alies in up to 40% of patients, most notably,
congenital heart disease, diaphragmatic hernia,
and chest wall/vertebral anomalies.
Not infrequently, components of bronchopul-
monary sequestration and CPAM are found in the
same specimen. These mixed lesions are denoted
“hybrid” lesions and represent a quarter of con-
genital lung lesions in some series. They are usu-
ally perfused by both pulmonary and systemic
arteries and drain into the pulmonary veins. In
Langston’s classification of congenital lung mal-
formations, these lesions are classified in the sub-
category of cystic adenomatoid malformations
with anomalous systemic arterial or venous con-
nections. The Stocker classification does not
include these mixed lesions.
36.6.3 Congenital Lung Emphysema
Congenital lobar emphysema is characterized
by expiratory air trapping within the affected
lobe. The air trapping leads to overdistension of
the affected lobe and compression of the adja-
cent lung and mediastinal structures. CLE most
often affects the left upper lobe and the right
upper and middle lobes (Kunisaki et al. 2019).
Unlike classic emphysema in adults, the lung
parenchyma in CLE appears normally devel-
oped; overinflation is secondary to bronchial
obstruction that causes air trapping within the
affected lobe. Hence, these lesions are rather
more appropriately referred to as congenital
lobar overinflation. Bronchial obstruction can
be related to focal deficiency or absence of the
A. A. Mokdad et al.
cartilaginous components of the lobar bronchus.
Other potential events that yield similar ana-
tomic and physiologic outcomes include endo-
bronchial obstruction due to secretions,
granulation tissue, ingested foreign bodies, or
endobronchial tumors. In addition, extrinsic
compression of a proximal lobar bronchus is
culprit at times and can occur as a result of
mediastinal lymphadenopathy, an aberrant or
enlarged pulmonary artery, ductus arteriosus, or
mediastinal cyst or tumor. Rarely, an apparent
CLE is associated with proliferation of normal
alveoli within the affected lobe without bron-
chial obstruction. This is termed polyalveolar
morphology.
36.6.4 Bronchogenic Cyst
Bronchogenic cysts belong to this spectrum of
bronchopulmonary-foregut cystic malforma-
tions. They are thick-walled, unilocular cysts.
The cyst wall resembles that of a bronchus and is
composed of smooth muscle, cartilage, elastic
tissue, and mucous glands lined by pseudostrati-
fied ciliated columnar or cuboidal epithelium.
Identification of bronchial wall structures distin-
guishes bronchogenic cysts from other cysts
such as esophageal or neurogenic cysts. While
bronchogenic cysts arise from the airway and
often remain intimately attached, they typically
have lost their communication with the airway
during development. They are usually fluid and
mucus filled. Bronchogenic cysts are believed to
occur as a result of abnormal budding of the tra-
cheobronchial tree; therefore they can be found
anywhere along the conducting airways depend-
ing on the point in development of the airway
when the anomalous budding occurred.
Bronchogenic cysts are most often located
within the lung parenchyma and mediastinum or
in the neck, but ectopic bronchogenic cysts have
been reported in paravertebral, paraesophageal,
pericardial, subcarinal, and subcutaneous loca-
tions. Rarely, these cysts may communicate with
adjacent structures including the esophagus and
the stomach.
36  Congenital Malformations of the Lung
36.7 Diagnosis and Differential
Diagnosis
Congenital malformations of the lung are more
commonly detected antenatally. As several other
congenital anomalies may present similarly, the
definitive diagnosis relies on clinical findings,
imaging, and histologic examination.
36.8 Presentation
In contemporary practice, most congenital lung
malformations are detected prenatally. Postna-
tally, these lesions typically present with either
respiratory distress in the newborn period or
recurrent lung infections in early childhood.
36.8.1 Congenital Pulmonary Airway
Malformation
CPAMs are usually identified during the fetal
anatomic survey between 18 and 20  weeks of
gestation. For these, spontaneous resolution is
reported to occur in up to 15% of patients.
Small CPAMs rarely impact the prenatal course
of the fetus and are usually asymptomatic at
birth. About a third ultimately develops recur-
rent pulmonary infections later in life. Large
CPAMs can produce mass effect on surround-
ing structures leading to polyhydramnios,
mediastinal shift, pleural effusions, and fetal
hydrops. Fetal hydrops is associated with a
high risk of fetal or neonatal demise, whereas
mortality approaches zero for a fetus with a
CPAM without hydrops. At birth, large lesions
present with respiratory distress that can range
from mild tachypnea and increased work of
breathing that interferes with feeding to severe
respiratory compromise requiring significant
respiratory support and possibly ECMO and/or
emergent thoracotomy and resection (Palla and
Sockrider 2019).
453
36.8.2 Bronchopulmonary
Sequestration
Extralobar bronchopulmonary sequestrations are
often identified on prenatal ultrasound or in
infancy during surgery for other congenital
anomalies, such as CPAM or congenital dia-
phragmatic hernia. They are often asymptomatic.
In contrast, most intralobar sequestrations pres-
ent later in childhood due to recurrent pulmonary
infections from inadequate tracheobronchial
drainage, or less commonly with hemorrhage.
Both intralobar and extralobar sequestrations can
be present in the newborn period with either
respiratory distress due to mass effect or conges-
tive heart failure because of arteriovenous shunt-
ing within the sequestration.
36.8.3 Congenital Lung Emphysema
CLE is most commonly found in term infants. It
may be challenging to distinguish from acquired
emphysema which can occur in premature infants
as a result of barotruma, oxygen toxicity, and lung
immaturity. However, over half of the affected
infants will demonstrate signs of respiratory dis-
tress in the first few days of life. After birth, spon-
taneous respiration leads to progressive over
distension of the affected lobe and compression of
normal lung. This process can be exacerbated by
the need for positive pressure ventilation. Up to
10–15% may rapidly progress to significant respi-
ratory failure requiring emergency thoracotomy.
For those who are asymptomatic at birth, symp-
toms may develop within the first few months or
years of life with either recurrent chest infections
with progressive distention and increased work of
breathing (Kunisaki et al. 2019).
36.8.4 Bronchogenic Cyst
Bronchogenic cysts are usually discovered inci-
dentally by chest radiography in the child. The
454
symptomatology of bronchogenic cysts varies
with age and depends on the location and size of
the lesion. Most infants present with wheezing,
tachypnea, dyspnea, cyanosis, or, less commonly,
failure to thrive as a result of compression of
adjacent mediastinal structures. Older children
typically present with infectious complications.
Rarely does enlargement of the cyst cause such
mediastinal displacement that they lead to car-
diopulmonary compromise. Rare cases of malig-
nancies within bronchogenic cysts have been
reported.
36.9 Imaging
36.9.1 Ultrasound
An ultrasound performed for antenatal fetal
screening is usually the first imaging modality
that identifies CPAM and bronchopulmonary
sequestration. These lesions appear as echogenic
foci with displacement of adjacent structures
(Chowdhury and Chakraborty 2015). They can
be distinguished by identifying the anomalous
feeding vessels in sequestrations using 3D
Doppler ultrasound. Ultrasound may also dem-
onstrate polyhydramnios, mediastinal shift, pleu-
ral effusions, and/or fetal hydrops. Notably,
similar ultrasound characteristics are typical of
congenital diaphragmatic hernias rendering them
challenging to distinguish from congenital lung
lesions on antenatal ultrasound. Antenatal ultra-
sound is also less likely to identify congenital
lobar emphysema as the process of air trapping
occurs postnatally.
Antenatal ultrasound criteria are used to pre-
dict outcome in a fetus with congenital lung mal-
formation. The CPAM volume ratio (CVR) was
developed to predict the risk of hydrops in CPAM
(Crombleholme et al. 2002). CVR is a measure of
the volume of the lesion, approximated to an
ellipse, and normalized by gestation age;
CVR = (CPAM length × height × width × 0.52)/
head circumference. CVR increases most rapidly
between 20 and 26 weeks of gestation and there-
after plateaus or diminishes with continued fetal
A. A. Mokdad et al.
growth. In Crombleholme et al.’s series, 75% of
fetuses with CVR  >  1.6 developed hydrops.
Among those, 88% were intubated after delivery
and only 53% survived, compared to 94% of
those with a CVR  <  1.6. A lower threshold, as
low as 0.4, has been proposed to help predict a
broader range of fetal morbidities including
mediastinal shift and effusion.
Postnatally, the use of ultrasound in evaluating
congenital lung malformations remains limited
due to lower diagnostic accuracy compared to CT
and MRI.
36.9.2 Plain Radiography
In the postnatal period, a plain chest radiograph
is usually performed first during workup for
respiratory concerns. Since a radiograph is occa-
sionally diagnostic for congenital lung malfor-
mations, further imaging is usually pursued. The
sensitivity of plain chest radiography in detecting
CPAM is around 60% (Fig. 36.4). Stocker type 1
and 2 lesions may manifest as a hyperlucent mul-
ticystic region on a chest radiograph, whereas a
type 3 CPAM or a sequestration may appear as a
radiopaque mass or a retrocardiac opacity
(Fig. 36.5). A chest radiograph will also identify
sequalae of lung lesions such as mass effect,
mediastinal shift, effusion, or pneumothorax. A
bronchogenic cyst most often appears as a
smooth, spherical mass on radiography in either a
paratracheal or hilar location in the posterior
mediastinum. If infected, or if the cyst communi-
cates with the airway, an air fluid level may be
present.
In contrast, the diagnosis of CLE is made most
commonly by plain chest radiography in a new-
born with respiratory distress (Fig. 36.6). Shortly
after birth, a lobe affected by CLE may appear
consolidated on chest radiography as a result of
inadequate clearance of amniotic fluid from the
affected lobe. Subsequently, classic radio-
graphic findings include hyperinflation of the
affected lobe, mediastinal displacement to the
contralateral side, atelectasis of the adjacent
lung, and flattening of the ipsilateral hemidia-
36  Congenital Malformations of the Lung
a c
Fig. 36.4  Preoperative plain chest radiography (a) and
CT scan (b) of a CPAM of the left upper lobe. A cystic
pulmonary mass was identified on prenatal US.  Despite
significant mediastinal shift and compression of adjacent
normal lung parenchyma on both prenatal ultrasound and
these preoperative images the infant was born at term and
phragm. Important in the differential diagnosis
of CLE are tension pneumothorax and macrocys-
tic CPAM.  Tension pneumothorax should be
marked by complete collapse of the entire ipsilat-
eral lung into the hilum, in contrast to CLE where
some compressed aerated normal lung is seen on
the affected side.
36.9.3 Computed Tomography
Contrast-enhanced computed tomography (CT)
is the gold standard imaging modality for evalu-
455
was asymptomatic. An elective resection was performed
at 2 months. At the time of resection, an associated extra-
pulmonary sequestration not seen on preoperative imag-
ing was resected. (c) Postoperative plain chest radiography
3  weeks postop demonstrating re-expansion of the left
lower lobe and absence of mediastinal shift
ating congenital malformations of the lung
(Chowdhury and Chakraborty 2015). Compared
to antenatal ultrasound or fetal MRI, postnatal
CT imaging has a higher accuracy in identifying
congenital lung malformation and delineating its
subtype (Fig.  36.4). Recognizing the risks of
radiation exposure in children, CT is preferably
reserved for preoperative planning, symptomatic
children, or cases of diagnostic uncertainty after
other imaging modalities.
CT scans can define anatomy, delineate the
level of involvement of a lesion, and quantify cyst
size. CT imaging can identify aberrant systemic
456
a b
Fig. 36.5  Preoperative plain chest radiography (a) and
CT scan (b) of an extrapulmonary sequestration. The pre-
natal ultrasound and the plain chest radiography after
birth were interrupted as most likely consistent with a
CPAM or small CDH.  Chest CT nicely demonstrates a
solid 2  cm mass in the left posterior mediastinum.
Although no systemic blood supply could be identified by
blood supply that is suggestive of a pulmonary
sequestration (Fig.  36.5). Hence, CT has sup-
planted the need for conventional angiography to
identify anomalous vasculature in differentiating
pulmonary sequestrations from other diagnoses.
Also, CT scans can confirm complete resolution
of an antenatally diagnosed congenital pulmo-
nary lesion that is not seen on postnatal plain
chest radiography. Congenital lobar emphysema
is usually diagnosed with plain chest radiograph,
but a CT scan would confirm the diagnosis if in
doubt (Fig. 36.6). CT imaging is also helpful in
differentiating a bronchogenic cyst, which
appears as a non-enhancing mass lesion on a CT
scan, from other lesions, such as neuroblastoma,
an aberrant pulmonary artery (pulmonary sling),
or a neuroenteric cyst.
36.9.4 Magnetic Resonance Imaging
Fetal magnetic resonance imaging (MRI) is an
important adjunct to antenatal ultrasound and
should be considered when the diagnosis is not
clear or in the case of high-risk lesions. Postna-
tally, MRI of the chest is an alternative to CT in
evaluating congenital malformations of the lung.
A. A. Mokdad et al.
CT imaging, this was felt to most consistent with an extra-
pulmonary sequestration since the diaphragm appeared
intact and the mass was outside normal lung parenchyma.
The lesion failed to demonstrate signs of resolution on
serial chest CT by 4  months of age and was resected at
4 months of age
MRI spares radiation exposure compared to CT;
however, undergoing an MRI generally is more
likely to require sedation or general anesthesia,
and it is limited in accurately delineating the lung
parenchyma. MR angiography is generally equiv-
alent to CT in making the diagnosis of a broncho-
pulmonary sequestration. It is similarly helpful
mapping and distinguishing a bronchogenic cyst
from a neuroblastoma or a neuroenteric cyst
(Fig. 36.7).
36.9.5 Other Imaging
Beyond ultrasound, plain radiography, CT, and
MRI, other imaging modalities may be important
in the workup of a patient with congenital lung
malformation. Fetal echocardiography is critical
in evaluation for cardiac anomalies or signs of
heart failure with hydrops. A screening echocar-
diography has been obtained traditionally in chil-
dren with CLE due to an association of CLE
congenital heart disease in approximately 15% of
patients. Recent large CLE series, however, have
not demonstrated a high prevalence of congenital
heart disease in this population (Kunisaki et  al.
2019).
36  Congenital Malformations of the Lung 457

a b

Fig. 36.6  Preoperative plain chest radiography (a) and sis of the adjacent normal lung parenchyma. Postoperative
CT scan (b) of a congenital lobar emphysema of the left plain chest radiography (c) following resection. Note the
upper lobe. Note the hyperinflation of the left upper lobe re-expansion of the remaining lung parenchyma and
and associated mediastinal shift and compressive atelecta- absence of mediastinal shift

In older children, bronchoscopy, in addition to

a CT or MR, is helpful to evaluate for reversible
causes of lobar emphysema. For bronchogenic
cysts, an esophagram is obtained in selected
cases to evaluate for foregut communication.

36.10 Management

36.10.1 Congenital Pulmonary
Airway Malformation

36.10.1.1 Antenatal Therapy

Fetal surgery is one of the more controversial
Fig. 36.7  Preoperative MRI of a bronchogenic cyst areas in pediatric surgery, due to the high risk of
458
complications and potential morbidity and mor-
tality for both the fetus and mother. For congeni-
tal lung lesions, treatment should be expectant
management with term delivery and postnatal
evaluation as long as the fetus does not experi-
ence hydrops or physiologic distress. In fetuses
developing fetal hydrops or distress, it is less
clear what treatment should be offered in the
absence of definitive evidence from randomized
trials or long-term follow-up studies (David et al.
Fig. 36.8  An algorithm for the management of prenatally
diagnosed congenital pulmonary malformations. Lesions
with a CVR ≤ 1.6 are at low risk for the development of
hydrops and should expect to delivery at term with little
respiratory compromise. Lesions with a CVR > 1.6 are at
high risk for the development of hydrops. The develop-
A. A. Mokdad et al.
2016). Figure 36.8 describes an approach for the
management of prenatally diagnosed congenital
pulmonary airway malformations. Management
is predicated on an accurate delineation of the
lesion including characterization of the cyst size,
particularly if macrocystic (>5  mm cysts) or
microcystic (<5  mm cysts), measurement of
CVR, and identification of hydrops. An ultra-
sound, fetal MRI, and fetal echocardiogram are
typically pursued. The ultrasound is repeated
ment of hydrops is associated with a high risk of fetal or
neonatal demise due to the mass effect caused by the con-
genital pulmonary mass. Fetal intervention should be con-
sidered in fetuses who develop hydrops and the therapy
tailored to the estimated gestational age and type of lesion
36  Congenital Malformations of the Lung
weekly, particularly if there is a high risk of
developing hydrops.
For fetuses less than 32 weeks of gestational
age, with hydrops or at risk of hydrops
(CVR  >  1.6), maternal steroids are advocated
(Downard et al. 2017). Some series have dem-
onstrated improvement of hydrops, regression
in the size of the lesion, and/or a decrease in
CVR with maternal betamethasone. A favorable
response to antenatal steroids is often achieved
with microcystic lesions, but these may be mixed
with other morphologies. In case no improve-
ment is noted with steroids, a thoracoamniotic
shunt may be considered for macrocystic lesions.
An ultrasound-guided percutaneously placed
thoracoamniotic shunt has the best outcome
with the lowest fetal-maternal risk for a CPAM
with a large cyst. A systematic review of fetuses
with a cystic adenomatoid malformation treated
with thoracentesis or a thoracoamniotic shunt
showed an 88% survival among those without
hydrops and 65% with hydrops (Cavoretto et al.
2008). Open maternal-fetal surgery with pulmo-
nary resection is an alternative for those with
microcystic lesions that do not respond to
maternal steroids. The procedure yields a 50%
probability of survival to discharge from the
NICU and imparts significant risk and potential
morbidity to the mother. Given the technical
complexity, this should only be performed in a
center with experience in open maternal fetal
surgery.
For fetuses over 32  weeks estimated gesta-
tional age with hydrops, an ex utero intrapartum
treatment (EXIT) procedure with thoracotomy
and lobectomy using placental bypass can permit
safe resection and avoid respiratory collapse.
Several series reported an overall survival
between 62 and 100% in fetuses in whom this
approach was performed either because of fetal
hydrops, extensive mediastinal shift, or persis-
tently elevated CVR (David et al. 2016).
36.10.1.2 Postnatal Therapy
Prompt surgical resection should be performed in
any symptomatic newborn. This is most often
lobectomy of the affected area. In patients who
remain asymptomatic after birth, an elective resec-
459
tion is generally agreed upon among experts, but
the timing of resection, surveillance strategy, and
surgical approach are more contentious areas. A
resection is advocated to preclude future respira-
tory compromise from recurrent infections or
other respiratory sequalae and to prevent malig-
nant degeneration. The likelihood of developing
respiratory symptoms following the neonatal
period varies between 3 and 86% in the literature.
The risk of malignant degeneration in a CPAM to
a pleuropulmonary blastoma or epithelial and
mesenchymal malignancies has been estimated to
be 9%. For patients with a positive DICER1 muta-
tion or radiologic features that portend an increased
likelihood for developing pleuropulmonary blas-
toma, early resection is warranted (Downard et al.
2017). Resection beyond the newborn period but
during the first year of life is most common among
pediatric surgeons and has been advocated by
expert groups (Downard et  al. 2017). Delaying
resection beyond the newborn period allows
somatic growth that facilitates pulmonary resec-
tion and permits monitoring a CPAM as 4%
resolve spontaneously. Operating around 6 months
of age allows ample time for compensatory lung
growth, avoids infectious complications, and pre-
cludes inflammation and adenopathy that increase
the difficulty of the resection. The latter reason has
prompted surgeons to recommend intervention as
early as 3  months. Postnatal CT scan is recom-
mended, the timing of which is dependent upon
provider’s preference and operative plans.
Resection most often takes the form of formal
lobectomy; however, for small CPAMs, nonana-
tomical resection is reasonable. Traditionally,
resection is done via open thoracotomy, but sev-
eral contemporary reports have demonstrated
excellent outcomes with thoracoscopic resection.
36.10.2 Bronchopulmonary
Sequestration
Prenatal management of bronchopulmonary
sequestrations is similar to that for CPAM, as
outlined above. Spontaneous involution of
extralobar sequestrations is rare. Postnatal resec-
tion of a sequestration is indicated for multiple
460
reasons. First, asymptomatic extralobar and
intralobar sequestrations pose a risk of recurrent
pulmonary infections or hemorrhage. Second, the
abnormally developed lung tissue is ineffective in
gas exchange and therefore is not of benefit to the
patient. Third, in practice, sequestrations that
present as a posterior mediastinal mass may be
difficult to differentiate from solid tumors,
CPAM, and possibly diaphragmatic hernia. Last,
thoracoscopic resection particularly of extralobar
sequestrations is often technically straightfor-
ward with low risk because there is no attach-
ment to the normal lung.
The goal of surgical resection is to remove
only the abnormal portion of lung. For an extralo-
bar sequestration, this is often easily achieved
unless there is inflammation due to previous
infection or hemorrhage. For intralobar seques-
tration, particularly if previously infected, this
most often necessitates a lobar resection (Palla
and Sockrider 2019).
Endovascular embolization has been employed
as well in the management of pulmonary seques-
tration. Transarterial, including transumbilical
arterial, embolization of the anomalous systemic
arteries feeding a sequestration resulted in regres-
sion of 80–90% of lesions treated (Chowdhury
and Chakraborty 2015). Among those, complete
regression varied between 7 and 53%. Leaving
residual dysplastic tissue, however, is a major
concern for this treatment modality and its adop-
tion has been limited.
36.10.3 Congenital Lobar
Emphysema
Lobar resection is indicated in any symptomatic
patient with CLE. This is commonly achieved by
an open thoracotomy. A thoracoscopic approach
is challenging, owing to a small working space,
inability to deflate the affected lobe, and potential
difficulty operating on the upper/middle lobes
that are implicated most in CLE. In a recent large
series of CLE, a quarter of patients initially
approached thoracoscopically were converted to
an open procedure. Of the overall study cohort,
6/53 (11.3%) patients had successful thoraco-
A. A. Mokdad et al.
scopic lobectomy (Kunisaki et  al. 2019).
Emergent thoracotomy with delivery of the
affected lobe can be lifesaving in a newborn with
severe respiratory distress as a result of CLE.
Although generally unnecessary in newborns,
older children should generally undergo bron-
choscopy preceding thoracotomy to evaluate for
reversible endobronchial lesions not requiring
pulmonary resection. Extrinsic bronchial com-
pression is generally associated with a focal car-
tilaginous defect such that relief of the extrinsic
compression alone is rarely adequate to relieve
the bronchial obstruction. While bronchoplasty is
theoretically attractive, the size of the bronchus
in an infant or child poses an impractical obstacle
to achieve an adequate functional result.
For an asymptomatic or minimally symptom-
atic child with CLE, there are no long-term data
to support a decision not to remove the lesion. A
few studies with short-term follow-up on a small
cohort of selected patients who have not under-
gone resection suggest that there is a low inci-
dence of progression of disease. While the
long-term outcome of these patients with
untreated CLE remains unknown, the postoper-
ative morbidity is minimal and the long-term
outcome after surgical resection for CLE is
excellent.
36.10.4 Bronchogenic Cyst
Resection is typically indicated to alleviate symp-
toms; to prevent future infection, hemorrhage,
airway obstruction, or malignant degeneration;
and to provide pathologic identification.
Generally, simple local resection with a mini-
mally invasive approach can be accomplished
with preservation of adjacent normal lung paren-
chyma. Resolution of active infection prior to
attempted resection is helpful to minimize unnec-
essary pulmonary resection. Occasionally, for-
mal pulmonary resection is required either due to
the anatomic location of the cyst or inflammation
from previous infections. Rarely, it is not possi-
ble to remove the cyst in its entirety without sac-
rificing vital structures; in such instances, partial
cyst resection with ablation of the remaining cyst
36  Congenital Malformations of the Lung
wall can be performed. Long-term follow-up is
necessary, however, as late recurrences have been
reported for partially resected bronchogenic
cysts.
36.11 Short- and Long-Term
Postoperative Outcomes
Pulmonary resection is generally very well toler-
ated, even in the newborn. Most contemporary
pediatric surgical series of pulmonary resection
for congenital pulmonary malformations in the
absence of diffuse lung disease cite a mortality
rate of <2%. Increased mortality from pulmonary
resection can be anticipated in newborns with
non-immune hydrops fetalis, significant medias-
tinal shift, pulmonary hypertension, or other
associated anomalies. Short-term complications
following resection are infrequent and include
prolonged air leak, pneumothorax, hemorrhage,
and infectious complications. Whether performed
by an open or a minimally invasive approach, the
frequency of short-term complications does not
appear to be higher in infants compared to older
pediatric patients (Oldham and Pinkerton 2005).
Due to in part to the continued lung develop-
ment that occurs well into childhood, pediatric
pulmonary resection is associated with an excel-
lent long-term functional outcome (Hall and
Stanton 2017). Most patients who undergo pul-
monary resection as an infant or child, in the
absence of other associated illness, will have nor-
mal growth in both height and weight. Laros
et  al. demonstrated that several decades after
pneumonectomy for tuberculosis or bronchiecta-
sis, the loss in total lung capacity was almost
fully compensated by the remaining lung if resec-
tion occurred in children less than 5 years of age.
The compensatory response was inversely related
to the age at the time of resection. Total lung
capacity 30 years after pneumonectomy was 96,
87, and 70% if the pneumonectomy occurred at
0–5, 6–10, or 31–40  years of age, respectively.
McBride et al. demonstrated a similar adaptation
of lung volume (total lung capacity and vital
capacity) in 15 infants who underwent lobectomy
for congenital lobar emphysema (McBride et al.
461
1980). They also observed evenly distributed pul-
monary blood flow between the operated and
non-operated sides, suggesting compensatory
growth of the pulmonary vascular bed. Despite
normalization of lung volume and pulmonary
blood flow, a decrease in expiratory flow rates
and airway conductance was noted. These later
findings have subsequently been validated by
others. Yet, despite the decreased expiratory flow
rates and airway conductance, Frenckner et  al.
demonstrated normal PaCO2 and PaO2 at rest and
following intense exercise in 16 patients who
underwent lobectomy during infancy or child-
hood for either congenital lobar emphysema or
CPAM (Frenckner and Freyschuss 1982). During
intense exercise, these individuals achieved nor-
mal workload, heart rate, and oxygen uptake.
Similarly, 3–20  years after partial pulmonary
resection for congenital cystic pulmonary lesions
in infancy and early childhood, Warner et  al.
found that the decreased expiratory flow rates did
not appear to impact on exercise endurance. After
excluding patients with other associated anoma-
lies, all of the patients in their series performed at
or above the 25th percentile and half were above
the 50th percentile for age in a treadmill endur-
ance test. In aggregate, whether this occurs
through further lung development or simply by
compensation by the remaining lung, these data
demonstrate the excellent functional outcome
associated with pediatric pulmonary resection
(Lau and Wong 2018).
36.12 Conclusion
The diverse spectrum of congenital malforma-
tions of the lung continues to pose a challenge
to elucidating the underlying etiology and
embryologic mechanisms governing these mal-
formations, evaluating them across the prenatal
and postnatal spectrum, as well as properly
managing them. There is a growing focus on
genetic predispositions and aberrant genetic
pathways that may drive congenital lung mal-
formations. Mortality remains high among
high-risk lesions with suboptimal prenatal ther-
apies. Postnatal management with resection has
462
been largely established; however, timing of
resection, surgical approach, and surveillance
strategy remain areas of controversy and neces-
sitate further research.
References
Cavoretto P, Molina F, Poggi S, Davenport M, Nicolaides
KH (2008) Prenatal diagnosis and outcome of echo-
genic fetal lung lesions. Ultrasound Obstet Gynecol
Off J Int Soc Ultrasound Obstet Gynecol 32:769–783.
https://doi.org/10.1002/uog.6218
Chowdhury MM, Chakraborty S (2015) Imaging of con-
genital lung malformations. Semin Pediatr Surg 24:
168–175. https://doi.org/10.1053/j.sempedsurg.2015.
02.001
Clements BS, Warner JO (1987) Pulmonary sequestration
and related congenital bronchopulmonary-­ vascular
malformations: nomenclature and classification
based on anatomical and embryological consider-
ations. Thorax 42:401–408. https://doi.org/10.1136/
thx.42.6.401
Crombleholme TM, Coleman B, Hedrick H, Liechty K,
Howell L, Flake AW et al (2002) Cystic adenomatoid
malformation volume ratio predicts outcome in pre-
natally diagnosed cystic adenomatoid malformation
of the lung. J Pediatr Surg 37:331–338. https://doi.
org/10.1053/jpsu.2002.30832
David M, Lamas-Pinheiro R, Henriques-Coelho T (2016)
Prenatal and postnatal management of congenital pul-
monary airway malformation. Neonatology 110:101–
115. https://doi.org/10.1159/000440894
Downard CD, Calkins CM, Williams RF, Renaud EJ,
Jancelewicz T, Grabowski J et al (2017) Treatment of
congenital pulmonary airway malformations: a sys-
tematic review from the APSA outcomes and evidence
based practice committee. Pediatr Surg Int 33:939–
953. https://doi.org/10.1007/s00383-­017-­4098-­z
A. A. Mokdad et al.
Fowler DJ, Gould SJ (2015) The pathology of congenital
lung lesions. Semin Pediatr Surg 24:176–182. https://
doi.org/10.1053/j.sempedsurg.2015.02.002
Frenckner B, Freyschuss U (1982) Pulmonary function
after lobectomy for congenital lobar emphysema and
congenital cystic adenomatoid malformation. A fol-
low-­up study. Scand J Thorac Cardiovasc Surg 16:293–
298. https://doi.org/10.3109/14017438209101066
Hall NJ, Stanton MP (2017) Long-term outcomes of con-
genital lung malformations. Semin Pediatr Surg 26:311–
316. https://doi.org/10.1053/j.sempedsurg.2017.09.001
Kunisaki SM, Saito JM, Fallat ME, St Peter SD, Kim
AG, Johnson KN et  al (2019) Current operative
management of congenital lobar emphysema in chil-
dren: a report from the Midwest Pediatric Surgery
Consortium. J Pediatr Surg 54:1138–1142. https://doi.
org/10.1016/j.jpedsurg.2019.02.043
Langston C (2003) New concepts in the pathology of
congenital lung malformations. Semin Pediatr Surg
12:17–37. https://doi.org/10.1053/spsu.2003.00001
Lau C-T, Wong KKY (2018) Long-term pulmonary func-
tion after lobectomy for congenital pulmonary airway
malformation: is thoracoscopic approach really better
than open? J Pediatr Surg 53:2383–2385. https://doi.
org/10.1016/j.jpedsurg.2018.08.029
McBride JT, Wohl ME, Strieder DJ, Jackson AC, Morton
JR, Zwerdling RG et al (1980) Lung growth and air-
way function after lobectomy in infancy for congenital
lobar emphysema. J Clin Invest 66:962–970. https://
doi.org/10.1172/JCI109965
Mullassery D, Smith NP (2015) Lung development. Semin
Pediatr Surg 24:152–155. https://doi.org/10.1053/j.
sempedsurg.2015.01.011
Oldham K, Pinkerton H (2005) Lung. In: Oldham
KT, Colombani PM, Foglia RP, Skinner MA (eds)
Principles and practice of pediatric surgery, 2nd edn.
Lippincott Williams & Wilkens, Philadelphia, PA,
p 951
Palla J, Sockrider MM (2019) Congenital lung mal-
formations. Pediatr Ann 48:e169–e174. https://doi.
org/10.3928/19382359-­20190326-­02
 Congenital Diaphragmatic Hernia
Prem Puri and Nana Nakazawa
37.1 Introduction
Congenital diaphragmatic hernia (CDH) is a
common malformation, characterized by a defect
in the diaphragm through which the abdominal
viscera migrate into the thoracic cavity during
fetal life. The defect is located in the posterolat-
eral diaphragm (Bochdalek hernia) in 90% of
cases and 9% occur as an anteromedial defect
(Morgagni hernia) (McHoney 2015). Total agen-
esis of diaphragm is rare.
37.2 Incidence
The incidence of CDH ranges from 1 in 2500–
5000 births in recent population-based studies.
Approximately 80% of posterolateral diaphrag-
matic hernia occurs on the left side, 15% are
right-sided, and less than 5% are bilateral
(Colvin et al. 2005; Gallot et al. 2007). The size
P. Puri (*)
Department of Pediatric Surgery, Beacon Hospital,
and University College Dublin, Dublin, Ireland
e-mail: prem.puri@ucd.ie
N. Nakazawa
Department of Pediatric Surgery, Juntendo University
Nerima Hospital, Tokyo, Japan
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_37
37
of the defect varies in size from small (2 or
3  cm) to very large, involving most of the
hemidiaphragm.
The incidence of associated anomalies has
been reported to be about 40% in most series.
The most common abnormalities associated with
CDH are cardiovascular anomalies, followed by
skeletal, central nervous system, genitourinary,
gastrointestinal, craniofacial, abdominal wall
defects, and chromosomal and syndromic defects.
Despite significant advances in neonatal resusci-
tation and intensive care, newborn infants with
CDH continue to have high mortality, which is
assumed to be even higher when it includes “hid-
den mortality.” Infants with associated anomalies
have much lower survival rates than those with
isolated CDH. The high mortality and morbidity
in CDH are mainly attributed to pulmonary hypo-
plasia and persistent pulmonary hypertension.
37.3 Etiopathogenesis
and Embryology
The etiology of CDH is still unknown. Although
most cases are isolated, associated malforma-
tions are often observed. Genetic studies have
indicated that chromosomal anomalies are
observed in 10% of CDH cases, and 3–10% of
CDH cases are associated with a syndromic diag-
nosis (Yu et  al. 2020). There are more than 70
syndromes in which CDH is described as a clini-
463
464
cal feature and the genetic causes have been
established in over 20 of these syndromes. In
some syndromes, such as Donnai-Barrow syn-
drome and Fryns syndrome, CDH is present in a
high percentage of affected individuals. Others
include Beckwith-Wiedemann, Simpson-Golabi-­
Behmel, Coffin-Siris, Cornelia De Lange, and
Denys-Drash syndromes. The CDH-associated
aneuploidies include trisomies 2 (mosaic), 9, 13,
16, 18, 21, and 22, Turner syndrome (45, X), and
trisomy X (46, XXX) (Kardon et  al. 2017; Yu
et al. 2020). The recurrence risk in a first-degree
relative is estimated to be 2% in families without
a known family history or known genetic
etiology.
In ∼30% of CDH patients, genomic analyses
have identified a range of genetic defects, includ-
ing chromosomal anomalies, copy number vari-
ants, and sequence variants (Kardon et al. 2017).
The affected genes identified in CDH patients
include transcription factors, such as GATA4,
ZFPM2, NR2F2, MYRF, and WT1, and signaling
pathway components, including members of the
retinoic acid pathway (Kardon et  al. 2017).
Mutations in these genes affect diaphragm devel-
opment and can have pleiotropic effects on pul-
monary and cardiac development.
Nongenetic factors have also been incrimi-
nated in the etiology of CDH. Thalidomide, qui-
nine, phenmetrazine, and nitrofen have been
mentioned as possible agents. A vitamin
A-deficient diet has also been shown to produce
CDH in rats.
Embryogenesis of CDH has been described as
a failure of the closure of the pleuroperitoneal
canal, which occurs during gestational week 8.
Consequently, the abdominal viscera herniate
into the thorax, which is thought to cause pulmo-
nary hypoplasia by the compression of the grow-
ing lung. However, experimental studies have
suggested that the classical view of embryogen-
esis of CDH may have to be revised. Kluth et al.
have shown that pleuroperitoneal canals are not
wide enough to allow herniation of gut loops in
rats. Furthermore, a toxicological nitrofen model
of CDH has shown that abnormalities in the con-
tralateral lung, as well as the ipsilateral side, are
present even before the diaphragm starts to
P. Puri and N. Nakazawa
develop. This concept was explained by Keijzer,
who proposed the so-called dual-hit hypothesis,
in which early retardation in lung development is
caused by nitrofen, whereas the late-gestational
increase in lung hypoplasia is caused by mechan-
ical compression from herniated viscera (Keijzer
et  al. 2000). The nitrofen-induced CDH model
has been widely used to investigate the pathogen-
esis of CDH because the phenotype resembles to
that in human condition. The retinoid signaling
pathway, a key regulator of embryonic morpho-
genesis, is likely to be involved, which is reported
to be disrupted in animal models and CDH neo-
nates (Beurskens et  al. 2010; Doi et  al. 2010;
Nakazawa et al. 2007).
37.4 Diagnosis
Widespread use of obstetrical ultrasonography
has led to an increased frequency of antenatal
diagnosis of CDH. Displacement of the mediasti-
num, the absence of a stomach bubble in the
abdomen, and the presence of abdominal organs
in the chest are signs of fetal diaphragmatic her-
nia. Incidence of pregnancies with polyhydram-
nios associated with CDH ranges from 15 to
75%. The presence of abdominal contents intra-
thoracically in a transverse sonographic scan, at
the level of a four-chamber view of the heart, is
required for diagnosis. A right-sided CDH is
more difficult to identify because the echo-
genicity of the fetal liver is similar to that of the
lung. Identifying the gallbladder in the fetal chest
may be the most reliable sign in these cases.
Prenatal diagnosis offers the advantage that
the mother can be referred to an institution with
maternity services and a neonatal intensive care
unit where the baby can be electively delivered,
resuscitated, and undergo surgery by a fully pre-
pared team of neonatologists and surgeons.
Advanced ultrasound examination will discrimi-
nate the condition from other intrathoracic
lesions, and other malformations will be looked
for. Prenatal magnetic resonance imaging has
been shown to be effective in confirming the
diagnosis of CDH and detecting additional infor-
mation that may affect prognosis.
37  Congenital Diaphragmatic Hernia
If lethal congenital anomalies are also
detected, prenatal diagnosis may lead to termina-
tion of pregnancy. In selected cases, in utero
intervention may be possible.
The onset and severity of symptoms depend on
the amount of abdominal viscera in the chest and
the degree of pulmonary hypoplasia. The most
severely affected infants present with respiratory
distress at birth. Other infants with CDH develop
cyanosis, tachypnea, and grunting respirations
within minutes or hours of birth. Physical examina-
tion reveals a scaphoid abdomen, an increased
anteroposterior diameter of the thorax, and medias-
tinal shift. Breath sounds are absent on the affected
side. Associated congenital anomalies may also be
seen or revealed on further examination.
The definitive diagnosis of CDH is made post-
natally by plain radiography of the chest and
abdomen by demonstration of air-filled loops of
the bowel in the chest and a paucity of gas in the
abdomen (Fig. 37.1a and b). The diaphragmatic
margin is absent, there is a mediastinal shift to
the opposite side, and only a small portion of
lung may be seen on the ipsilateral side.
Although most CDH infants present in the
first 24  h of life, 10–20% of affected infants
a b
Fig. 37.1 (a) X-ray of the chest in a newborn showing
large left-sided Congenital Diaphragmatic Hernia with
Mediastinal shift to the right. (b) X-ray of the chest and
465
present later. The symptoms and signs of those
patients are nonspecific and include recurrent
chest infections, vomiting, abdominal pain,
diarrhea, anorexia, failure to thrive, or an
abdominal chest X-ray in an asymptomatic
patient. Some children present acutely with vol-
vulus or strangulation or acute respiratory dis-
tress. Chest X-ray with an in situ nasogastric
tube is reliable for the diagnosis. Even if the
hernia is asymptomatic, it should be repaired to
prevent complications.
37.5 Differential Diagnosis
Differential diagnosis includes congenial pul-
monary airway malformation (CPAM) and
mediastinal cystic lesions (e.g., cystic teratoma,
neuroenteric, bronchogenic, and thymic cysts). A
congenital lung cyst in the neonate may produce
a radiological picture mimicking that of dia-
phragmatic hernia but can be distinguished from
CDH by the presence of an outline of the lung
and the diaphragm on chest X-ray and by the
presence of a normal intestinal gas pattern on
abdominal X-ray. The diagnosis of CDH can also
upper abdomen showing large right-sided diaphragmatic
hernia
466
be confused with other congenital thoracic condi-
tions, such as eventration of the diaphragm, con-
genital esophageal hiatal hernia, and primary
agenesis of the lung.
37.6 Prognostic Factors
Several factors have been proposed as predictors
for poor diagnosis. Intrathoracic herniation of the
stomach and/or liver has been proposed as pre-
dictors for poor diagnosis. The determination of
lung-to-head ratio (LHR), first described in 1996,
has been suggested to be predictive of postnatal
survival. LHR < 1.0 is reported to be correlated
with poor outcome. Since LHR increases with
gestation, the values of LHR measured at differ-
ent gestational ages were not found to be compa-
rable. Therefore, the observed LHR normalized
by dividing it by the expected value for that ges-
tation, the observed-to-expected LHR (O/E LHR)
has become the preferred method, which has
been shown to be a good predictor of neonatal
mortality. O/E LHR is reported to be associated
with death when the value is less than ~20%
(Ruano et al. 2012).
The observed-to-expected total fetal lung
volume (O/E TLV), percent predicted lung vol-
umes (PPLV), and percent liver herniation,
need for extracorporeal membrane oxygenation
(ECMO), and development of chronic lung dis-
ease in MRI studies have been shown to be pre-
dictors of mortality (Jeanty et al. 2014; Walleyo
et al. 2013; Oluyomi-Obi et al. 2017).
Recently, the CDH study group developed the
Score for Neonatal Acute Physiology-II to pre-
dict not only mortality, but also the need for
ECMO in CDH patients on the first day of life, by
analyzing a multivariable logistic regression
adjusted for hernia side, gestational age, liver
position, center, ventilation mode, and O/E LHR
(Snoek et  al. 2016). Furthermore, low birth
weight, patch repair, and need for ECMO corre-
late with more severe pulmonary hypertension at
1 month of age (Wynn et al. 2013).
P. Puri and N. Nakazawa
37.7 Prenatal Treatment
As recent advances in imaging and evaluation
have provided early and accurate prenatal diag-
nosis of CDH, prenatal intervention aiming to
reverse or improve associated lung hypoplasia is
becoming a more acceptable alternative for better
prognosis and outcome.
Fetal surgery with primary repair of the defect
seemed to be a promising approach, but clinical
application of anatomical fetal CDH repair was
abandoned once it became clear that it was not
possible in fetuses with liver herniation and that
those without did not benefit from the interven-
tion (Harrison et al. 1993, 1997).
The concept of tracheal occlusion to improve
fetal lung development came from experimental
studies. Back in the 1970s, Alcorn et al. demon-
strated that serial ligation and drainage of the
fetal lamb lung led to pulmonary hyperplasia
(Alcorn et al. 1977). The first report of fetal tra-
cheal occlusion for the treatment of CDH in
humans was by Harrison et  al. (1981). An ex
utero intrapartum therapy (EXIT) procedure was
developed to reverse the occlusion.
Recent advancements in minimally invasive
surgery have led to the current less-invasive tech-
nique, fetal endoscopic tracheal occlusion
(FETO). Usually, the tracheal balloon is placed
endoscopically in a one-port technique, followed
by planned balloon removal by repeat fetoscopy
or ultrasound-guided balloon puncture (Deprest
et al. 2006; Kovler and Jelin 2019). A recent large
international randomized control trial, the
Tracheal Occlusion to Accelerate Lung growth
trial (TOTAL trial), investigating the role of fetal
therapy for severe and moderate pulmonary
hypoplasia reported that in fetuses with isolated
severe congenital diaphragmatic hernia on the
left side, FETO performed at 27 to 29 weeks of
gestation resulted in a significant benefit over
expectant care with respect to survival to dis-
charge, and this benefit was sustained to 6 months
of age (Deprest et al. 2021). FETO increases the
risks of preterm, prelabor rupture of membranes
37  Congenital Diaphragmatic Hernia
and preterm birth. (Deprest et  al. 2014, 2021;
Verla et al. 2018).
Various other medical strategies for lung
hypoplasia, such as steroid administration with or
without thyrotropin-releasing hormone, vita-
mins, or stem cell therapy, have been tested in the
last decades in different CDH animal models, but
their impact on the human situation has yet to be
addressed (Eastwood et  al. 2015; Jeanty et  al.
2014). Especially, regenerative medicine includ-
ing stem cell therapy and tissue engineering
seems to be a promising field for further treat-
ment strategies in CDH.
37.7.1 Preoperative Management
An infant with respiratory distress requires
endotracheal ventilatory support. Mask ventila-
tion should be avoided as it will distend the
stomach and further compromise respiratory
status. These babies should be paralyzed and
sedated to prevent swallowing and noncompli-
ance with ventilation and to minimize stressful
stimuli that cause pulmonary vasoconstriction.
The concept of “gentle” ventilation and permis-
sive hypercapnia was proposed to minimize
barotraumas by strictly limiting the peak infla-
tion pressure.
Despite many advances in the diagnosis and
treatment of CDH, morbidity and mortality
remain high, largely due to pulmonary hyperten-
sion (PH), along with pulmonary hypoplasia and
cardiac dysfunction (Gupta and Harting 2020).
Pulmonary hypertension (PH) is sustained,
supranormal pulmonary arterial pressure, and
among patients with CDH, is driven by abnormal
pulmonary vasculature, including alterations at
the molecular, cellular, and tissue levels, along
with pathophysiologic pulmonary vasoreactivity
(Harting 2017). Inhaled nitric oxide (iNO)
induces pulmonary vasodilatation without sys-
temic hypotension and can be beneficial in the
context of pulmonary hypertension. However, its
relevance for CDH patients remains controver-
sial. Short-term improvement in oxygenation in
467
selected patients has been observed with positive
effect on stabilizing the patient during transport
or awaiting ECMO cannulation; however, inhaled
NO does not reduce the need for ECMO itself
(McHoney 2015; Oliveira et  al. 2000; Harting
2017).
High-frequency oscillatory ventilation (HFOV)
provides effective ventilation while decreasing
barotrauma but has not been shown to improve the
mortality or morbidity rates in CDH (Puligandla
et  al. 2015; Snoek et  al. 2016). Any changes in
HFOV settings must be monitored carefully, as
high airway pressures may cause lung hyperinfla-
tion, with adverse effects on venous return, pulmo-
nary vascular resistance, and ultimately in cardiac
output (Logan et al. 2007).
Infants who fail to respond to optimal therapy
may be placed on ECMO. This is a life support
system that employs partial heart-lung bypass,
providing rest to the lungs for long periods of
time during which it is hoped that the lung and,
in particular, the lung vasculature will mature.
Optimal patient selection for ECMO in CDH
requires refinement of non-ECMO support tech-
niques, so that this higher risk but higher poten-
tial reward modality is focused primarily on
those patients with more severe CDH as defined
by smaller lungs, worse birth physiology, anat-
omy, and larger defects (Kays 2017). Although
widely used, a Cochrane review found that the
ECMO benefit remains unclear (Mugford et al.
2008).
37.7.2 Timing of Surgery
The timing of surgical repair has gradually
shifted from an emergency repair to a policy of
stabilization using a variety of ventilatory strate-
gies prior to operation. However, several studies
provided no strong advantage for a delayed
(when stabilized) or early (within 24–48 h after
birth) repair (Moyer et al. 2002; Okuyama et al.
2017). The optimal timing of repair of CDH, in
relation to treatment with ECMO, also remains
controversial. Recently, Tan et al. (2019) reported
468
that both the mean oxygenation index (OI) on
day1 and the best oxygenation index on day1 are
highly predictive of mortality in CDH. Oxygen-
ation indices in the first 48 h poorly predicted the
timing of surgery. The authors concluded that
pre-Op-OI < 3 may be a cutoff for optimal timing
for surgery in infants for CDH repair.
37.7.3 Surgical Technique
Open (laparotomy or thoracotomy) or minimally
invasive (laparoscopic or thoracoscopic) tech-
niques can be chosen for the surgical repair;
however, the optimal approach is still a matter of
discussion among surgeons. The most com-
monly preferred approach is abdominal. This
offers good exposure, easy reduction of the
abdominal viscera, and recognition and correc-
tion of associated gastrointestinal anomalies. A
subcostal transverse muscle cutting incision is
made on the side of the hernia (Fig. 37.2). The
Fig. 37.2  A subcostal transverse muscle cutting incision
is made on the side of the hernia
P. Puri and N. Nakazawa
contents of the hernia are gently reduced in the
abdomen. On the right side, the small intestine
and the colon are first reduced and the liver is
withdrawn last. After the hernia is reduced, an
attempt is made to visualize the ipsilateral lung.
This is usually done by retracting the anterior
rim of the diaphragm. Often, a hypoplastic lung
can be observed at the apex. Most diaphragmatic
defects can be sutured by direct sutures of the
edges of the defect. Usually, the anterior rim of
the diaphragm is quite evident. However, the
posterior rim may not be immediately apparent
and may require dissection for delineation. The
posterior rim of the diaphragm is mobilized by
incising the overlying peritoneum. The defect is
closed by interrupted nonabsorbable sutures
(Fig.  37.3a and b). Occasionally, the posterior
rim is absent altogether, in which case the ante-
rior rim of the diaphragm is sutured to the lower
ribs with either periosteal or pericostal sutures.
If the defect is large, it may not be possible to
repair it using direct sutures. Various techniques
have been described and include the use of prer-
enal fascia, rib structures, the latissimus dorsi
muscle, rotational muscle flaps from the thora-
coabdominal wall, and prosthetic patches. The
operations involving muscle flaps are too long
and complex for critically ill patients and can
lead to unsightly chest deformities. Prosthetic
materials have been advocated for repair of large
defects (Fig.  37.4). Numerous types of patches
are commercially available (natural vs. syn-
thetic, absorbable vs. nonabsorbable), but the
ideal material has yet to be identified. Right-
sided CDH has been found to require patch
repair more commonly than left-sided CDH due
to larger defect size or complete agenesis (Collin
et  al. 2016). If the abdominal cavity is small,
gentle stretching of the abdominal wall will
enable safe closure in most of the patients. The
use of a chest drain is controversial. The argu-
ment against the use of a chest drain is in avoid-
ance of barotraumas as it increases the
transpulmonary pressure gradient.
37  Congenital Diaphragmatic Hernia 469

a b

Fig. 37.3  Posterior rim of the diaphragm is mobilized by incising the overlying peritoneum. (a, b) The defect is closed
by interrupted nonabsorbable sutures

37.7.4 Postoperative Management period, the so-called honeymoon period, but

The infant is transferred to the intensive care unit,
kept warm, and given maintenance requirements
of intravenous fluids. The vital signs are moni-
tored closely with regular blood gas analyses and
monitoring of preductal and postductal oxygen-
ation. Ventilatory support should be continued
postoperatively. Some infants show improvement
in oxygenation in the immediate postoperative
deteriorate 6–24 h later. This deterioration is due
to pulmonary hypertension and persistent fetal
circulation, with an increase in pulmonary artery
resistance, elevated pulmonary artery pressure,
and right-to-left ductal and preductal shunting
leading to hypoxemia. Pulmonary hypertension
is probably caused by multiple factors, such as
increased abdominal pressure with impaired vis-
ceral and peripheral perfusion, limited diaphrag-
470 P. Puri and N. Nakazawa

malities have frequently been described in

CDH survivors. Developmental delay and
abnormalities, in both motor and cognitive
skills, have been documented. Nutritional and
growth-related problems are common in CDH
patients. Gastroesophageal reflux, which is one
of the possible causes of this issue, has fre-
quently been found in CDH survivors.
Sensorineural hearing loss is also a well-­
described morbidity in CDH patients. Generally
accepted risk factors for hearing loss in neo-
nates include hyperventilation, ototoxic medi-
cations, severe hypoxemia, prolonged
mechanical ventilation, and ECMO therapy.
Musculoskeletal abnormalities have also been
described in CDH survivors. Chest asymmetry
Fig. 37.4 Patch placement for large diaphragmatic
and pectus deformity are the most commonly
defects
documented, followed by vertebral deformity.
Recurrence of hernia has been reported more
matic excursion, overdistension of the alveoli in frequently in CDH patients and is repaired with
the hypoplastic lungs with diminished alveolar-­ a prosthetic patch. Because of significant mor-
capillary blood flow, release of vasoactive cyto- bidity in survivors with CDH, it has been sug-
kines, and deterioration in pulmonary compliance gested that these patients should be cared for in
after surgical repair. Sudden deterioration in the multidisciplinary follow-up clinics.
patient’s oxygenation status should raise the sus-
picion of pneumothorax, and a chest tube should
be inserted if the diagnosis is confirmed. Infection 37.9 Congenital Diaphragmatic
complications, including pneumonia and septice- Eventration (CDE)
mia, are not uncommon.
Eventration of the diaphragm has been described
as an abnormally high or deviated position of all
37.8 Outcome or part of the hemidiaphragm. Eventration may
be congenital or acquired as a result of phrenic
Recent improvements in the treatment of infants nerve palsy. Congenital diaphragmatic eventra-
with CDH have increased the survival of more tion (CDE) is a developmental abnormality that
severely affected infants. Long-term follow-up of results in muscular aplasia of the diaphragm,
those patients has led to the recognition of pul- which initially has fully developed musculature,
monary and extrapulmonary morbidities that and becomes atrophic secondary to phrenic nerve
were not previously recognized. damage and disuse. Although this section deals
Pulmonary morbidity, including pulmonary with congenital eventration, the clinical features
hypoplasia, bronchopulmonary dysplasia, per- and principles of the management are similar in
sistent pulmonary hypertension, reactive air- congenital and acquired forms of eventration.
way disease, and limited postnatal alveolar The estimated incidence is reported to be 1  in
growth, is the most common and significant 10,000 neonates (Vanamo et al. 1996). It is more
problem in CDH infants surviving beyond the common in males and more commonly affects
neonatal period. Neurodevelopmental abnor- the left hemidiaphragm.
37  Congenital Diaphragmatic Hernia 471

37.10 Clinical Features 37.11 Diagnosis

Clinical features range from being asymptom-
atic to severe respiratory distress. Patients may
present later in infancy with repeated attacks of
pneumonia, bronchitis, or bronchiectasis.
Occasionally, patients present with gastrointes-
tinal symptoms of vomiting or epigastric dis-
comfort later in childhood. In patients with
phrenic nerve palsy, there may be a history of
difficult delivery. They may present with tachy-
pnea, respiratory distress, or cyanosis. Physical
examination reveals decreased breath sounds on
the affected side mediastinal shift during inspi-
ration and a scaphoid abdomen. Occasionally,
CDE patients exhibit associated malformations,
such as hypoplastic lung, congenital heart dis-
ease, or cryptorchidism (Wu et al. 2015).
The diagnosis of eventration is usually made on a
chest X-ray. Frontal and lateral chest X-rays will
show an elevated diaphragm with a smooth,
unbroken outline (Fig. 37.5a and b). Fluoroscopy
is a useful investigation for differentiating a com-
plete eventration from a hernia. Paradoxical
movement of the diaphragm is seen if complete
eventration is present. Ultrasonography is also
useful for the identification of abnormal organs
underneath the eventration.

a b

Fig. 37.5 (a) Right-sided CDE with liver visible in the right chest and mediastinal shift to the left. (b) Lateral X-ray
image of right CDE
472
37.12 Management
Asymptomatic patients, without a significant
underlying pulmonary abnormality, may be
treated expectantly. Also, those patients with
incomplete phrenic nerve palsy without paradox-
ical movement should be treated conservatively
as normal function will usually return. Symptom-
atic patients, especially those with respiratory
distress, need prompt supportive care with endo-
tracheal intubation and ventilation with humidi-
fied oxygen to minimize the diaphragmatic
excursions. A nasogastric tube is passed to
decompress the stomach, and intravenous fluids
are commenced. Surgery is undertaken once the
patient’s condition is stabilized.
37.13 Operative Repair
The first successful surgical repair of the dia-
phragmatic eventration was described in 1923
(Morrison 1923). Plication of the diaphragm has
been used for many years to treat eventration
(Fig. 37.6). Plication increases both tidal volume
and maximal breathing capacity and has been suc-
Fig. 37.6 Eventration repaired by plication of the
diaphragm
P. Puri and N. Nakazawa
cessful in many clinical series. An abdominal
approach through a subcostal incision is preferred
for left-sided eventration but a thoracic approach
through a posterolateral incision via the sixth
space may be used for right-sided lesions. The
transabdominal approach allows good visualiza-
tion of the entire diaphragm from the front to back
and easier mobilization. Diaphragm plication can
also be done via laparoscopy or thoracoscopy.
37.14 Outcome
The prognosis of patients with CDE without
underlying pulmonary hypoplasia is usually
excellent. Mortality is related to pulmonary
hypoplasia. Postoperative complications include
pneumonia, pleural effusions, and recurrence.
37.15 Conclusions
Although advances in neonatal resuscitation,
intensive care treatment, and technical equipment
have improved the overall survival rate up to 90%
in highly specialized centers, CDH remains a
therapeutic challenge due to its associated comor-
bidities, lung hypoplasia, and pulmonary hyper-
tension. Optimal perioperative stabilization and
management by an interdisciplinary team of
pediatric surgeons, neonatologists, and anesthe-
tists is crucial for neonates with this potentially
life-threatening condition.
References
Alcorn D, Adamson TM, Lambert TF, Maloney JE,
Ritchie BC, Robinson PM (1977) Morphological
effects of chronic tracheal ligation and drainage in the
fetal lamb lung. J Anat 123(Pt 3):649–660
Beurskens LWJE, Tibboel D, Lindemans J, Duvekot
JJ, Cohen-Overbeek TE, Veenma DCM et  al (2010)
Retinol status of newborn infants is associated
with congenital diaphragmatic hernia. Pediatrics
126(4):712–720
Collin M, Trinder S, Minutillo C, Rao S, Dickinson J,
Samnakay N (2016) A modern era comparison of right
versus left sided congenital diaphragmatic hernia out-
comes. J Pediatr Surg 51(9):1409–1413
37  Congenital Diaphragmatic Hernia
Colvin J, Bower C, Dickinson JE, Sokol J (2005)
Outcomes of congenital diaphragmatic hernia:
a population-­ based study in Western Australia.
Pediatrics 116(3):e356–e363
Deprest J, Jani J, Cannie M, Debeer A, Vandevelde M,
Done E et al (2006) Prenatal intervention for isolated
congenital diaphragmatic hernia. Curr Opin Obstet
Gynecol 18(3):355–367
Deprest J, Nicolaides K, Benachi A, Gratacos E, Ryan G,
Persico N, Sago H, Johnson A, Wielgos M, Berg C,
Van Calster B, Russo FM (2021) Randomized trial of
fetal surgery for severe left diaphragmatic hernia. N
Engl J Med 385:107–118
Deprest J, Brady P, Nicolaides K, Benachi A, Berg C,
Vermeesch J et al (2014) Prenatal management of the
fetus with isolated congenital diaphragmatic hernia in
the era of the TOTAL trial. Semin Fetal Neonatal Med
19(6):338–348
Doi T, Sugimoto K, Ruttenstock E, Dingemann J, Puri
P (2010) Prenatal retinoic acid upregulates pulmo-
nary gene expression of PI3K and AKT in nitrofen-­
induced pulmonary hypoplasia. Pediatr Surg Int
26(10):1011–1015
Eastwood MP, Russo FM, Toelen J, Deprest J (2015)
Medical interventions to reverse pulmonary hypo-
plasia in the animal model of congenital diaphrag-
matic hernia: a systematic review. Pediatr Pulmonol
50(8):820–838
Gallot D, Boda C, Ughetto S, Perthus I, Robert-Gnansia
E, Francannet C et  al (2007) Prenatal detection and
outcome of congenital diaphragmatic hernia: a French
registry-based study. Ultrasound Obstet Gynecol
29(3):276–283
Gupta VS, Harting MT (2020) Congenital diaphragmatic
hernia-associated pulmonary hypertension. Semin
Perinatol 44(1):151–167
Harrison MR, Ross NA, de Lorimier AA (1981)
Correction of congenital diaphragmatic hernia in
utero. III.  Development of asuccessful surgical tech-
nique using abdominoplasty to avoid compromise of
umbilical blood flow. J Pediatr Surg 16(6):934–942
Harrison MR, Adzick NS, Flake AW, Jennings RW, Estes
JM, MacGillivray TE et al (1993) Correction of con-
genital diaphragmatic hernia in utero: VI. Hard-earned
lessons. J Pediatr Surg 28(10):1411–1417. discussion
1417–8
Harrison MR, Adzick NS, Bullard KM, Farrell JA, Howell
LJ, Rosen MA et  al (1997) Correction of congenital
diaphragmatic hernia in utero VII: a prospective trial.
J Pediatr Surg 32(11):1637–1642
Harting MT (2017) Congenital diaphragmatic hernia-­
associated pulmonary hypertension. Semin Pediatr
Surg 26(3):147–153
Jeanty C, Kunisaki SM, MacKenzie TC (2014) Novel
non-surgical prenatal approaches to treating congeni-
tal diaphragmatic hernia. Semin Fetal Neonatal Med
19(6):349–356
Kardon G, Ackerman KG, McCulley DJ, Shen Y, Wynn
J, Shang L et  al (2017) Congenital diaphragmatic
473
hernias: from genes to mechanisms to therapies. Dis
Model Mech 10(8):955–970
Kays DW (2017) ECMO in CDH: is there a role? Semin
Pediatr Surg 26(3):166–170
Keijzer R, Liu J, Deimling J, Tibboel D, Post M (2000)
Dual-hit hypothesis explains pulmonary hypoplasia in
the Nitrofen model of congenital diaphragmatic her-
nia. Am J Pathol 156(4):1299–1306
Kovler ML, Jelin EB (2019) Fetal intervention for con-
genital diaphragmatic hernia. Semin Pediatr Surg
28(4):150818
Logan JW, Cotten CM, Goldberg RN, Clark RH (2007)
Mechanical ventilation strategies in the management
of congenital diaphragmatic hernia. Semin Pediatr
Surg 16(2):115–125
McHoney M (2015) Congenital diaphragmatic her-
nia, management in the newborn. Pediatr Surg Int
31(11):1005–1013
Morrison JMW (1923) Eventration of diaphragm due
to unilateral phrenic nerve paralysis. Arch Radiol
Electrotherap 28:72–75
Moyer V, Moya F, Tibboel R, Losty P, Nagaya M, Lally
KP (2002) Late versus early surgical correction for
congenital diaphragmatic hernia in newborn infants.
Cochrane Database Syst Rev 3:CD001695
Mugford M, Elbourne D, Field D (2008) Extracorporeal
membrane oxygenation for severe respiratory fail-
ure in newborn infants. Cochrane Database Syst Rev
3:CD001340
Nakazawa N, Montedonico S, Takayasu H, Paradisi F,
Puri P (2007) Disturbance of retinol transportation
causes nitrofen-induced hypoplastic lung. J Pediatr
Surg 42(2):345–349
Okuyama H, Usui N, Hayakawa M, Taguchi T (2017)
Appropriate timing of surgery for neonates with con-
genital diaphragmatic hernia: early or delayed repair?
Pediatr Surg Int 33(2):133–138
Oliveira CA, Troster EJ, Pereira CR (2000) Inhaled nitric
oxide in the management of persistent pulmonary
hypertension of the newborn: a meta-analysis. Rev
Hosp Clin 55(4):145–154
Oluyomi-Obi T, Van Mieqhem T, Ryan G (2017) Fetal
imaging and therapy for CDH-current status. Semin
Pediatr Surg 26(3):140–146
Puligandla PS, Grabowski J, Austin M, Hedrick H,
Renaud E, Arnold M et  al (2015) Management of
congenital diaphragmatic hernia: a systematic review
from the APSA outcomes and evidence-based practice
committee. J Pediatr Surg 50(11):1958–1970
Ruano R, Takashi E, da Silva MM, Campos JADB,
Tannuri U, Zugaib M (2012) Prediction and prob-
ability of neonatal outcome in isolated congeni-
tal diaphragmatic hernia using multiple ultrasound
parameters. Ultrasound Obstet Gynecol 39(1):42–49
Snoek KG, Capolupo I, Morini F, van Rosmalen J,
Greenough A, van Heijst A et al (2016) Score for neo-
natal acute physiology-II predicts outcome in congen-
ital diaphragmatic hernia patients. Pediatr Crit Care
Med 17(6):540–546
474
Tan YW, Ali K, Andradi G, Sasidharan L, Greenough A,
Davenport M (2019) Prognostic value of the oxygen-
ation index to predict survival and timing of surgery in
infants with congenital diaphragmatic hernia. J Pediatr
Surg 54(8):1567–1572
Vanamo K, Rintala R, Lindahl H, Louhimo I (1996)
Congenital eventration of the diaphragm  – 45 years’
perspective. Pediatr Surg Int 11(2–3):123–125
Verla MA, Style CC, Olutoye OO (2018) Prenatal inter-
vention for the management of congenital diaphrag-
matic hernia. Pediatr Surg Int 34(6):579–587
Walleyo A, Debus A, Kehl S, Weiss C, Schönberg SO,
Schaible T et  al (2013) Periodic MRI lung volume
assessment in fetuses with congenital diaphragmatic
P. Puri and N. Nakazawa
hernia: prediction of survival, need for ECMO, and
development of chronic lung disease. AJR Am J
Roentgenol 201(2):419–426
Wu S, Zang N, Zhu J, Pan Z, Wu C (2015) Congenital
diaphragmatic eventration in children: 12years’ expe-
rience with 177 cases in a single institution. J Pediatr
Surg 50(7):1088–1092
Wynn J, Krishnan U, Aspelund G, Zhang Y, Duong J,
Stolar CJH et al (2013) Outcomes of congenital dia-
phragmatic hernia in the modern era of management. J
Pediatr 163(1):114–119.e1
Yu L, Hernan RR, Wynn J, Chung WK (2020) The influ-
ence of genetics in congenital diaphragmatic hernia.
Semin Perinatol 44(1):151169
 Extracorporeal Membrane
Oxygenation
38
Brian P. Fallon, Samir K. Gadepalli,
Jason S. Frischer, Charles J. H. Stolar,
and Ronald B. Hirschl

Table 38.1  ELSO registry neonatal respiratory failure

38.1 Introduction cases (2015–2020)
Total runs Survived % Survived
Extracorporeal membrane oxygenation (ECMO) is CDH 1202 616 51
a life-saving technology that affords partial heart MAS 629 574 91
and/or lung bypass for extended periods. ECMO is PPHN/PFC 529 384 72
a supportive rather than a therapeutic modality, as it Sepsis 107 55 51
RDS 27 23 85
provides sufficient gas exchange and perfusion in
Pneumonia 22 10 45
patients with acute, reversible cardiac or respira- Air leak syndrome 5 4 80
tory failure. It provides a finite period to “rest” the Other 1252 894 71
cardiopulmonary systems during which time they CDH congenital diaphragmatic hernia, MAS meconium
are spared insults from traumatic mechanical venti- aspiration syndrome, PPHN persistent pulmonary hyper-
lation and perfusion impairment. ECMO was first tension of the newborn, PFC persistent fetal circulation,
implemented in newborns in 1974. Since then, the RDS respiratory distress syndrome
Extracorporeal Life Support Organization (ELSO)
has recorded 71,079 neonatal and paediatric 38.3 demonstrate the common neonatal and paedi-
patients treated with ECMO for a wide range of atric respiratory and cardiac diagnoses along with
cardiorespiratory disorders. Tables 38.1, 38.2 and survival with ECMO support. In the neonatal popu-
lation, congenital diaphragmatic hernia (CDH) is
B. P. Fallon · S. K. Gadepalli currently the most common indication for ECMO
University of Michigan, Ann Arbor, MI, USA and carries with it a survival rate of 51%. Other
e-mail: bfallon@med.umich.edu;
frequent diagnoses (with survival rates in parenthe-
samirg@med.umich.edu
ses) include meconium aspiration syndrome
J. S. Frischer
(MAS; 91%), persistent pulmonary hypertension
Cincinnati Children’s Hospital Medical Center,
Cincinnati, OH, USA of the newborn (PPHN; 72%), sepsis (51%), respi-
e-mail: jason.frischer@cchmc.org ratory distress syndrome (RDS; 85%) and pneu-
C. J. H. Stolar monia (45%). Acute respiratory distress syndrome
Stanley Morgan Children’s Hospital, (ARDS) is the most common aetiology amongst
New York, NY, USA the paediatric population requiring ECMO for
e-mail: unknown_user_518314@meteor.springer.com
respiratory support and has a 69% survival. Other
R. B. Hirschl (*) common indications include non-ARDS respira-
C.S. Mott Children’s Hospital, University of
tory failure (65%), viral pneumonia (71%), bacte-
Michigan, Ann Arbor, MI, USA
e-mail: rhirschl@med.umich.edu rial pneumonia (74%) and aspiration (67%). The

© Springer Nature Switzerland AG 2023 475

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_38
476 B. P. Fallon et al.

Table 38.2  ELSO registry pediatric respiratory failure ECMO), and (b) an alveolar-arterial oxygen gra-
cases (2015–2020) dient (A-aDO2) > 610 mmHg for more than 8 h.
Total % Older infants and children do not have as well-
runs Survived Survived defined criteria for high mortality risk. A ratio of
ARDS, not post-op/ 328 227 69
arterial PaO2 to FiO2 (P:F ratio) less than 60–100 in
trauma
Acute respiratory 316 208 65 hypoxic respiratory failure or an arterial pH less
failure, non-ARDS than 7.0–7.25 due to hypercapnic respiratory fail-
Viral pneumonia 282 203 71 ure is commonly used to identify candidates for
Bacterial pneumonia 193 143 74 ECMO.  The Paediatric-Pulmonary Rescue with
Aspiration pneumonia 40 27 67 Extracorporeal Membrane Oxygenation Predic-
ARDS, post-operative/ 25 18 72
trauma
tion (P-PREP) score has been developed to predict
Pneumocystis 1 1 100 in-hospital mortality for children over 7  days of
pneumonia age prior to the initiation of ECMO. It considers
Other 2023 1299 64 mode of support (V-A or V-V), duration of
ARDS acute respiratory distress syndrome mechanical ventilation, P:F ratio, arterial pH, pri-
mary pulmonary diagnosis and comorbid condi-
Table 38.3  ELSO registry neonatal and pediatric cardiac tions. Indications for support in patients with
failure cases (2015–2020) cardiac pathology are based on clinical signs of
Total runs Survived % Survived cardiovascular failure such as hypotension despite
Congenital defect 2949 1559 53 the administration of inotropes or volume resusci-
Cardiac arrest 97 45 46 tation, metabolic acidosis, oliguria (urine output
Cardiogenic shock 499 287 58 <0.5 cc/kg/h) and decreased peripheral perfusion.
Cardiomyopathy 108 65 60 In addition, the gestational age should be at
Myocarditis 137 108 79
least 30 weeks due to the increased risk for intra-
Other 2691 1576 59
cranial haemorrhage (ICH) and the birth weight at
least ~1.5 kg secondary to cannula size limitations.
most common cardiac indications for ECMO, The length of mechanical ventilation prior to
amongst both neonatal and paediatric populations, ECMO, and its associated toxicity from prolonged
are congenital heart defects and cardiogenic shock. exposure to high concentrations of oxygen and
Survival amongst neonatal patients with these con- elevated positive pressure ventilation, should be no
ditions is 48% and 50%, respectively; for paediat- longer than 14  days due to the development of
ric patients, survival is 56% and 59%, bronchopulmonary dysplasia. Babies with lethal
respectively. congenital anomalies should not be considered for
Candidates for ECMO are expected to have a ECMO support. Treatable conditions such as total
reversible cardiopulmonary disease process. anomalous pulmonary venous return and transpo-
ECMO should be considered when predicted sition of the great vessels, which may initially
mortality is greater than 50% despite maximal masquerade as pulmonary failure, can be corrected
ventilatory and medical management; it is with surgery, but may require ECMO resuscitation
strongly recommended when predicted mortality initially. Therefore, echocardiogram should be
reaches 80%. Obviously, these criteria are sub- rapidly obtained to determine cardiac anatomy.
jective and vary between institutions. Criteria for There should be no evidence of significant neuro-
mortality risk in neonatal respiratory failure have logic injury such as seizures. Patients with sugges-
been suggested to identify infants with a high tion of a small ICH (grades I–II) should be
risk of mortality. These include (a) the oxygen- considered candidates for ECMO on an individual
ation index (OI), calculated as FiO2 * mean air- basis and monitored closely for worsening of the
way pressure * 100/PaO2 (OI > 25 is predictive haemorrhage. In fact, all gross active bleeding or
of a 50% mortality rate and is a relative indica- major coagulopathy should be corrected prior to
tion for ECMO, while OI > 40 equates with 80% ­initiating ECMO, but patients with such derange-
mortality and mandates implementation of
38  Extracorporeal Membrane Oxygenation
a b
IVC
SVC
Gravity
Lung
IVC
SVC
Gravity
Lung
Fig. 38.1 (a–c) The goal of ECMO support is to provide
gas exchange, oxygen delivery and often haemodynamic
support. There are three standard configurations for extra-
corporeal circulation: venoarterial (V-A), venovenous
(V-V) and double-lumen single-cannula venovenous (DL
V-V) bypass. V-A bypass allows for support of both the pul-
monary and cardiac systems. Venous blood is drained from
the right atrium (RA) through the internal jugular vein (IJ),
and oxygenated blood is returned to the aorta via the carotid
artery (CA) (a). In larger children, the common femoral
vein (CFV) and common femoral artery (CFA) can be used
for drainage and return, respectively. Potential disadvan-
tages of this arrangement include the sacrifice of a major
artery; risk of gaseous or particulate emboli into the sys-
temic circulation; reduced pulmonary perfusion; increased
afterload, which may reduce cardiac output; non-pulsatile
flow; perfusion of the coronaries by relatively hypoxic left
ventricular blood; and limb ischaemia with CFA or CFV
cannulation. V-V and DL V-V avoid these disadvantages
and provide pulmonary support but do not provide haemo-
477
IVC
SVC
Gravity
Lung
dynamic/cardiac support. V-V bypass is accomplished with
drainage from the IJ/RA with reinfusion into the inferior
vena cava (IVC) via the CFV (b) or drainage from the IVC/
CFV with reinfusion into the IJ/RA. DL V-V is carried out
by means of the IJ (c). A major disadvantage of V-V and DL
V-V ECMO is that a fraction of freshly infused blood recir-
culates back into the circuit and requires approximately a
20% increase in flow rate. In summary, it is recommended
that patients who require only respiratory support use V-V
or DL V-V bypass and those who necessitate cardiac sup-
port use V-A ECMO. If necessary, one can convert V-V or
DL V-V to V-A support. Additional cannulae can be placed
for hybrid cannulation to support particular physiologic cir-
cumstances. A second venous reinfusion cannula (V-AV)
can be used to provide oxygenated blood to the heart of a
patient on femoral V-A support who is recovering heart
function but still has poor lung function (North-South syn-
drome). Patients on V-V support who subsequently develop
cardiac failure can benefit from a carotid or femoral arterial
cannula to convert to V-VA
478
Temp Probe Connector
Arterial Line Heat Exchanger
Bridge
Venous
Line
Oxymetrics Probe
Fig. 38.2  The circuit is comprised of three main compo-
nents: a roller pump, a membrane oxygenator and a heat
exchanger. Right atrial blood is drained by gravity siphon
into a venous servomechanism, which acts to ensure that
venous return to the circuit is adequate for the current pump
flow. To do so, the servo detects diminished venous return,
slows or shuts off the pump and sounds an alarm, hence stop-
ping blood flow and relieving the risk of introducing air into
the circuit, cavitation and injury to the right atrium. Next, a
roller or centrifugal pump, with continuous servoregulation
and pressure monitoring, perfuses the blood through the
membrane oxygenator. The oxygenator is typically a cham-
B. P. Fallon et al.
Membrane
Oxygenator
Bladder Pump
ber containing hollow fibres through which gas flows, with
blood flowing around the outside of the fibres with low resis-
tance. The size of the oxygenator is chosen based on the
patient’s size. The oxygenated blood flows through a heat
exchanger and is then returned to the patient. A bridge is con-
structed to connect the venous line shortly after exiting the
patient and the arterial line just prior to entering the patient so
that during weaning the patient and the circuit can easily
form two separate circuits. In the case of femoral cannula-
tion, an additional smaller cannula can be placed into the
superficial femoral artery or posterior tibial artery for perfu-
sion of the limb distal to the primary arterial return cannula
38  Extracorporeal Membrane Oxygenation 479

Common carotid a.

Vagus n.

Internal jugular v.

Fig. 38.3  Open cut-down cannulation can be performed
in the neonatal or paediatric intensive care units under
adequate sedation, with proper monitoring, while percuta-
neous cannulation should be performed with fluoroscopic
and ultrasound guidance, most often in the operating
room. The patient is positioned with the head at the foot of
the bed, supine and the head and neck hyperextended over
a shoulder roll and turned to the left. Local anaesthesia is
administered over the proposed incision site. A transverse
cervical incision is made over the sternomastoid muscle,
one finger’s breadth above the right clavicle. The platysma
muscle is divided with electrocautery. Self-retaining
retractors are placed and dissection is carried out with the
sternomastoid muscle retracted to expose the carotid
sheath. Using sharp dissection and meticulous haemosta-
sis, the sheath is opened and the internal jugular vein,
common carotid artery and vagus nerve are identified. All
vessels must be handled with extreme care as to avoid
spasm. The vein is dissected free first and mobilized over
proximal and distal ligatures. Occasionally, it is necessary
to ligate the inferior thyroid vein. The common carotid
artery lies medial and posterior, contains no branches and
is mobilized in a similar fashion. The vagus nerve should
be identified and protected from injury
480
Fig. 38.4  The patient is then systemically heparinized
with 50–100 U/kg heparin, which is allowed to circulate
for 3  min. The arterial cannula (usually 10  Fr for new-
borns) is measured so that the tip will lie at the junction of
the brachiocephalic artery and the aorta (2.5 cm, one-third
the distance between the sternal notch and the xiphoid).
The venous cannula (12–14 Fr for neonates) is measured
to have its tip in the distal RA (6 cm, one-half the distance
between the suprasternal notch and the xiphoid process).
For V-A bypass, the carotid artery is ligated cranially.
Proximal control is obtained with an angled clamp, and a
transverse arteriotomy is made near the ligature. Stay
sutures, using 5-0 or 6-0 polypropylene, are placed
through the full thickness of the medial and lateral proxi-
mal edges of the arteriotomy to help prevent subintimal
dissection. The sutures are gently retracted and the clamp
slowly released as the arterial catheter is inserted into the
carotid artery to its proper position. The cannula is then
fastened into place with two silk ligatures (2-0), with a
small piece of vessel loop, on the anterior aspect, inside
the ligatures to protect the vessel from injury during
decannulation
B. P. Fallon et al.
Fig. 38.5  In preparation for the venous cannulation, the
patient is given succinylcholine to prevent spontaneous res-
piration. The vein is then ligated cranially. Gentle traction is
placed on the lower ligature to help decrease back bleeding,
and a venotomy is made close to the proximal ligation. The
drainage catheter is passed to the level of the RA and
secured in a manner similar to that used for the arterial cath-
eter. The cannulas are debubbled with back bleeding and
heparinized saline after each individual cannula is placed.
Once in position, the cannulas are connected to the ECMO
circuit and bypass is initiated. Both cannulas are then
secured to the mastoid process using 2-0 suture. The wound
is irrigated, meticulous haemostasis obtained and the skin
closed with a running nylon suture. The site is covered with
a sterile dressing and the circuit tubing is fixed securely to
the bed. For V-V and DL V-V bypass, the procedure is
exactly as described above including dissection of the
artery so that a future switch from V-V to V-A ECMO can
be accomplished, if necessary. The catheter tip should be in
the mid-right atrium (6 cm in most neonates) with the arte-
rial portion of the catheter pointed towards the ear. This
directs the oxygenated blood flow towards the tricuspid
valve. Cannula position is confirmed by chest X-ray and by
transthoracic echocardiogram when necessary. The venous
cannula should optimally be located in the inferior aspect of
the right atrium and the arterial cannula at the ostium of the
innominate artery and the aorta. With a double-lumen
venous catheter, the tip should be in the mid-right atrium
with reinfusion of oxygenated blood flow towards the tri-
cuspid valve. Percutaneous cannulation is most often per-
formed for DL V-V bypass, although it can be used for the
V-V cannula configuration demonstrated in Fig. 38.1b and
femoral V-A bypass. The technique, along with concerns
for potential complications, is described in the references
by Jarboe et al. and Berdajs et al.
38  Extracorporeal Membrane Oxygenation 481

ments should be considered candidates on an indi- normal), anti-factor Xa activity (anti-Xa; 0.3–
vidual basis (Figs. 38.1, 38.2, 38.3, 38.4 and 38.5).0.7  IU/mL) or thromboelastography (TEG). A
complete blood count should be obtained every
6 h and coagulation profiles daily. In order to pre-
38.2 Patient Management vent haemorrhage, platelets are transfused to
on ECMO maintain a platelet count above 100,000/mm3,
and some authors sustain fibrinogen levels above
Once the cannulas are connected to the circuit, 100  mg/dL.  The haematocrit should remain
bypass is initiated and flow is slowly increased to above 40% using red blood cell transfusions so
100–150 mL/kg/min so that the patient is stabi- that oxygen delivery is optimized.
lized. Continuous inline monitoring of the venous Volume management of patients on ECMO is
(prepump) SvO2 and arterial (postpump) PaO2 as extremely important. It is imperative that all
well as pulse oximetry is vital. The goal of V-A inputs and outputs be diligently recorded and
ECMO is to maintain a mixed venous PO2 of electrolytes monitored every 6 h. All fluid losses
37–40 mmHg and saturation (SvO2) of 65–80%. should be repleted and electrolyte abnormalities
V-V ECMO is more difficult to monitor due to corrected. All patients should receive mainte-
variation in the degree of recirculation, which nance fluids as well as adequate nutrition using
may produce a falsely elevated SvO2 assessment. hyperalimentation. Trophic enteral feeds should
Inadequate oxygenation and perfusion are indi- be attempted and advanced as tolerated, but
cated by metabolic acidosis, oliguria and hypo- patients will typically require some amount of
tension. Arterial blood gasses should be parenteral nutrition to meet their caloric and
monitored frequently with PaO2 and PaCO2 nutrient needs. The first 48–72 h of ECMO typi-
maintained as close to normal level as possible. cally involves fluid extravasation into the soft tis-
As soon as these parameters are met, all vasoac- sues. The patient becomes oedematous and may
tive drugs are weaned, and ventilator levels are require volume replacement (crystalloid, colloid
adjusted to “rest” settings (typically peak inspira- or blood products) in order to maintain adequate
tory pressure 15–20  cmH2O, positive end-­ intravascular and bypass flows, haemodynamics
expiratory pressure 5–10 cmH2O, respiratory rate and urine output greater than 1  cc/kg/h. By the
12–20 and FiO2 0.21–0.3). Mild sedation and third day of bypass, diuresis of the excess extra-
analgesia are provided usually with fentanyl and cellular fluid begins and can be facilitated with
midazolam, but the use of a paralyzing agent is the use of furosemide or even continuous renal
avoided. Alternatively, patients can be extubated replacement therapy (CRRT) if necessary.
and managed without sedation to promote wake- Surgical procedures, such as CDH repair, may
fulness and rehabilitation. Routine antimicrobial be performed while the child remains on bypass.
prophylaxis is not recommended. Surveillance Haemorrhagic complications are a frequent mor-
blood, urine and tracheal cultures can be per- bidity associated with this situation and increase
formed, though practice varies and their benefit is mortality. To avoid these complications, prior to
not clear. the procedure, the platelet count should be greater
Anticoagulation is administered throughout than 100,000/mm3, a fibrinogen level above
the ECMO course in order to preserve a circuit 150  mg/dL, an ACT reduced to 180–200  s and
free of thrombus. Unfractionated heparin (typi- ECMO flow increased to full support. It is imper-
cally starting at 15–30 U/kg/h) is the current stan- ative that meticulous haemostasis be obtained
dard of care, though there is increasing use of the throughout the surgery. Fibrinolysis inhibitor
direct thrombin inhibitor bivalirudin. The level of aminocaproic acid (100 mg/kg) just prior to inci-
anticoagulation should be monitored hourly sion followed by a continuous infusion (30 mg/
using activated clotting time (ACT; maintained at kg/h) until all evidence of bleeding ceases is a
180–220 s or 1.5 times normal), activated partial useful adjunct.
thromboplastin time (aPTT; 50–60 s or 1.5 times
482
As the patient improves, the flow of the circuit
may be weaned at a rate of 10–20 mL/h as long as
the patient maintains good oxygenation and per-
fusion. With V-A bypass, flows should be
decreased to 30–50  mL/kg/min and the ACT
should be at a higher level (200–220 s) to prevent
thrombosis. Moderate conventional ventilator
settings are used, but higher settings can be used
if the patient needs to be weaned from ECMO
urgently. If the child tolerates the low flow, all
medications and fluids should be switched to vas-
cular access on the patient, and the cannulas may
be clamped with the circuit bypassing the patient
via the bridge. During V-V, there is no need to
decrease or discontinue flow, but instead a trial
off of ECMO is performed by removing the
sweep flow through the artificial lung. Under
both circumstances, the patient is observed for
2–4 h and, if this is tolerated, ECMO is discontin-
ued and decannulation performed. This should be
executed under sterile conditions with muscle
relaxants administered to prevent air aspiration
into the vein. The catheters are removed, the ves-
sels are ligated and the wound is closed.
38.3 Complications
Extracranial bleeding is a common complication
of the heparinized ECMO patient either at the site
of cannulation or at other sites and is noted in
17% of neonatal respiratory cases, 26% of paedi-
atric respiratory cases and 39% of all cardiac
cases. Bleeding at the site of cannulation can
often be treated with local pressure or the place-
ment of topical haemostatic agents such as
Gelfoam, Surgicel or topical thrombin. For all
sites of bleeding, the platelet count should be
increased to >100,000 mm3 and the ACT lowered
to 180–200 s. Sometimes the temporary discon-
tinuation of anticoagulation and normalization of
the coagulation status are warranted to help stop
the haemorrhage with a second circuit available
in the event acute clotting of the circuit should
occur. Aggressive surgical intervention is war-
ranted if bleeding persists.
Neurologic sequelae are a serious morbidity
of the ECMO population and include learning
B. P. Fallon et al.
disorders, motor dysfunction and cerebral palsy.
These outcomes appear to be as much due to
hypoxia and acidosis prior to the ECMO course
as the time on ECMO itself. ICH is the most dev-
astating complication, occurring in 7% of new-
born patients with an associated 72% mortality
amongst newborns who have ICH on
ECMO.  Frequent comprehensive neurologic
exams should be performed and cranial ultra-
sounds obtained frequently while on ECMO
based on local protocols. Blood pressure should
be carefully monitored and maintained within
normal parameters to help decrease the risk of
ICH.  If necessary, electroencephalograms may
be helpful in the neurologic evaluation.
Acute tubular necrosis (ATN), marked by oli-
guria and increasing blood urea nitrogen and cre-
atinine levels, is often seen in the ECMO patient
during the initial 48 h, at which time renal function
is expected to improve. If improvement does not
occur, consideration must be towards poor tissue
perfusion. This may be due to low cardiac output,
insufficient intravascular volume or inadequate
pump flow, all of which should be corrected. In the
event of continued renal failure, haemofiltration or
haemodialysis can be ­ performed to maintain
proper fluid balance and electrolyte levels and are
reported to be required in 26% of cases.
38.4 Conclusion
As of January 2020, 32,385 neonates (87% sur-
vival) and 10,346 paediatric patients (72% sur-
vival) have been treated with ECMO for
respiratory failure and 8830 neonatal (69% sur-
vival) and 12,538 paediatric (72% survival)
patients for cardiac failure. Tables 38.1, 38.2 and
38.3 demonstrate the common neonatal and pae-
diatric respiratory and cardiac diagnoses along
with survival with ECMO support. In the neona-
tal period, the most common disorders treated
with ECMO are CDH, MAS, PPHN, sepsis, RDS
and cardiac support. For the paediatric popula-
tion, viral and bacterial pneumonia, ARDS, acute
respiratory failure (non-ARDS) and cardiac
­disease are the most common pathophysiologic
processes requiring ECMO intervention.
38  Extracorporeal Membrane Oxygenation
Recent medical advances, such as permissive
hypercapnea, inhaled nitric oxide and the use of
oscillatory ventilation, have spared numerous
babies from ECMO, yet many children still ben-
efit from this modality.
In summary, any patient with reversible cardio-
pulmonary disease, who meets criteria, should be
considered an ECMO candidate. ECMO provides
an excellent opportunity to provide “rest” to the
cardiopulmonary systems thus avoiding the addi-
tional lung or cardiac injury which otherwise would
be associated with maintaining life support.
Further Reading
Bailly DK, Reeder RW, Zabrocki LA, Hubbard AM,
Wilkes J, Bratton SL et  al (2017) Development and
validation of a score to predict mortality in children
undergoing extracorporeal membrane oxygenation for
respiratory failure. Crit Care Med 45(1):e58–e66
483
Berdajs D (2015) Bicaval dual-lumen cannula for veno-
venous extracorporeal membrane oxygenation:
Avalon(c) cannula in childhood disease. Perfusion
30:182–186
Campbell BT, Braun TM, Schumacher RE et  al (2003)
Impact of ECMO on neonatal mortality in Michigan
(1980–1999). J Pediatr Surg 38:290–295
Extracorporeal Life Support Organization (2020)
International registry report of the extracorporeal life
support organization. University of Michigan Medical
Center, Ann Arbor
Hirschl RB, Bartlett RH (2012) Extracorporeal life
support in cardiopulmonary failure. In: Coran AG,
Adzick NS, Krummerl T, Laberge JM, Shamberger R,
Caldamone A (eds) Pediatric surgery, 5th edn. Mosby,
New York, pp 89–102
Jarboe MD, Gadepalli SK, Church JT et al (2017) Avalon
catheters in pediatric patients requiring ECMO:
placement and migration problems. J Pediatr Surg
53:159–162
Kim ES, Stolar CJ (2000) ECMO in the newborn. Am J
Perinatol 17:345–356
Kim ES, Stolar CJ (2003) Extracorporeal membrane oxy-
genation for neonatal respiratory failure. In: Puri P
(ed) Newborn surgery. Arnold, London, pp 317–327
 Part V
Spina Bifida and Hydrocephalus
 Spina Bifida and Encephalocoele
Martin T. Corbally
39.1 Introduction
Neural tube defects (NTD: spina bifida (SB),
encephalocoele) are potentially serious congeni-
tal deformities of the spine and spinal cord that
can have a major impact on the quality of life, not
only for the child but on the entire family. The
precise aetiology is uncertain. Although the inci-
dence appears to be decreasing, there remain a
significant number of newborns with this condi-
tion each year. In the past 10 years, an awareness
of the benefits of peri-conceptual folic acid and
improved nutrition has significantly decreased
the incidence of NTD.  The impact of antenatal
screening and therapeutic abortion in some juris-
dictions has clearly further reduced the incidence.
Surviving children face a varied future, directly
related to the severity of their NTD and to the
quality of early interventional services and long-­
term support structures. The patient with NTD is
likely to require the expertise of many services
and specialists over their lifetime including the
paediatric surgeon/neurosurgeon/urologist/ortho-
paedic/ophthalmic surgeon/paediatric radiolo-
gist/social workers/continence nurses and many
other varied disciplines. Children with SB face
the prospect of multiple surgical, urological and
orthopaedic interventions for the duration of their
M. T. Corbally (*)
Royal College of Surgeons in Ireland, Dublin, Ireland
e-mail: martin.corbally@khuh.org.bh
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_39
39
lives and must cope with the effects of poor or
zero ambulation, bladder and renal failure, hydro-
cephalus and the complexity of multiple shunt/
bladder or bowel and shunt procedures.
The management objective for these children
aims to provide as normal a life as possible, to
minimise the effect of their disability in areas
such as mobility, continence and education. In
addition, urinary system monitoring is essential
to safeguard against the complications of a neu-
ropathic bladder and renal failure.
39.2 Embryology
A NTD is a congenital defect of the spine and
neural tube with failure of fusion of the vertebral
arches and, to a varying degree, the development
of the covering muscles and skin. In some cases,
the neural tube will protrude externally as a neu-
ral plaque without any covering of skin or mus-
cle, as in a myelomeningocoele, but in others, the
neural tube is closed, but there is a defect of the
vertebral arch and muscles through which the
dura and arachnoid protrude (meningocele) or it
is entirely covered by skin (spina bifida occulta).
Essentially, the defect arises as an abnormality
of fusion of the neural tube. At the start of the 4th
week of foetal life, the neural plate (precursor of
the neural tube) is a broad flat plate in its cranial
portion that will become the brain and a narrow
caudal portion that will become the spinal cord.
487
488 M. T. Corbally

At about 22 days, the embryo undergoes ventral
flexion, and that portion of the neural tube cranial
to the point of flexure (the mesencephalic flexure)
is recognisable as the future forebrain, the point at
the flexure is the midbrain, and the point caudal to
this point is the hindbrain. Rapid elongation over
the next 7 days occurs with the narrow caudal por-
tion (future spinal cord) occupying up to 60% of
the neural tube. One of the most important events
of the 4th week is conversion of the neural plate
into a neural tube by a series of infolding of the
plate called neurulation of the tube. This process
begins along the future occipito-cervical region of
the plate and progresses caudally. During this pro-
cess, the lateral edges of the plate meet and fuse in
the midline while detaching from the surface
ectoderm, which then fuse and so cover the neural
tube completely. The tube remains open at both
ends during this process through small openings
called the cranial and caudal neuropores. The cra-
nial neuropore closes completely at day 24 and
the caudal neuropore at day 26. The neural folds
have essentially closed by 4 weeks. Subsequently,
the mesodermal somites form around the closed
cord, and the meninges, vertebral column and
muscles result.
Failure of part of the neural tube to close dis-
rupts both the process of differentiation of the
central nervous system and the induction of the
vertebral arches and can result in a variety of
anomalies. This most commonly affects the cau-
dal end of the spinal cord, which affects the lum-
bar and sacral regions of the central nervous
system. Involvement of the cranial end of the tube
can result at its most extreme end in anencephaly
and form an encephalocoele in less severe cases.
Less severe anomalies of fusion are failure of
the arch to fuse, with or without meningocele
protrusion (spina bifida occulta v meningocele,
respectively). More severe defects result in fail-
ure of the neuroectoderm with protrusion of the
neural tube itself (myelomeningocoele). A simi-
lar process in the brain results in an encephalo-
coele. The process of lack of fusion can occur
anywhere along the length of the spinal canal,
with varying levels of severity.
39.3 Classification
39.3.1 Anencephaly
Failure of the cranial end of the neural tube to
close can result in disruption of the differentiation
of the CNS and is represented by an exposed mass
of undifferentiated neural tissue (Table  39.1).
These embryos often survive to late pregnancy
but usually do not survive much after birth.

Table 39.1  Types of
neural tube defects
Anencephaly Brain and skull poorly developed Death inevitable
Myelomeningocoele Failure of closure of neural tube Significant lesion
Failure of muscle and skin Variable outcome
formation Urgent closure
Exposed neural tissue 90% need VP shunt
Distal limb innervation affected
Neuropathic bladder
Meningocoele Failure of spinal fusion Usually no neural
consequences
Dural sac protrudes Rarely bladder
function affected
Skin covered defect
Encephalocoele Usually occipital Variable outcome
Defect in cranial bone
Herniation of meninges and Sometimes shunt
Brain to varying degree needed
Spina bifida Occulta hamartoma at site Excellent outcome
Sinus occasionally
Skin intact
Bony vertebral arch deficient
39  Spina Bifida and Encephalocoele
Fig. 39.1  MRI showing occipital encephalocoele
39.3.2 Encephalocoele
Although it is controversial whether this is truly a
NTD defect, it is probably best considered as
such for the purposes of this discussion
(Fig.  39.1). It results from bony defects in the
cranial vault, which lead to a herniation of the
meninges, with or without differentiated brain
tissue. The condition is more rightly called a
meningoencephalocoele if dura and brain tissue
herniate or a meningohydroencephalocoele if a
part of the ventricular system also herniates.
39.3.3 Spina Bifida Occulta
At its mildest extreme, the vertebral arches of a
single vertebra fail to fuse, but there is no underly-
ing abnormality of the neural tube. The defect may
occur anywhere along the spine but is most com-
monly found in the lumbosacral region, and its
presence may only be signalled by the presence of
a small tuft of hair, small dimple, pigmented skin
or vascular lesion overlying the lesion.
39.3.4 Meningocoele
Abnormalities of the neural arch without under-
lying neural tube defects with the formation of a
489
Fig. 39.2  Lumbosacral myelomeningocoele
cystic swelling at the site of the lesion are called
a meningocoele. This cystic swelling is lined by
dura and arachnoid.
39.3.5 Myelomeningocoele
In this, the severest form of spina bifida, the tis-
sues overlying one or more vertebra are deficient
so that the neural tube tissue itself protrudes as a
neural plaque to the surface (Fig.  39.2). This
plaque may be completely or partially covered by
arachnoid or exposed. It most typically involves
the lumbar region, but any portion of the spine
can be affected.
39.4 Aetiology
The aetiology of NTD is clearly multifactorial,
and no single agent, either genetic or teratogenic,
has been identified. However, there is evidence of
a genetic influence in some cases, e.g. SB is more
common in some parts of India and Ireland (1.1%)
and is relatively rare in African Americans
(0.035%). In addition, the presence of NTD in one
sibling increases the risk among subsequent sib-
lings to 1 in 20. In families with two NTD siblings,
the risk increases to 1 in 8. In certain syndromes,
e.g. Meckel syndrome, an autosomal recessive dis-
order, craniorachischisis may be seen. NTD may
also be seen in the Waardenburg syndrome, which
may result from Pax-3 gene abnormalities.
490
It also seems likely that environmental factors
are important in the development of NTD. There
has been a significant decline in the incidence of
NTDS over the past four decades. Factors noted
have been the association of maternal diabetes,
the anti-epileptic drug sodium valproate and
hyperthermia. Valproate may interfere with folate
metabolism, and there is also evidence that a sig-
nificant number of infants with NTD may have
gene mutations that are involved in folate and
vitamin B metabolism, especially mutations of
5,10-methylenetetrahydrofolate reductase and
methionine synthase reductase. The administra-
tion of peri-conceptual folic acid has probably
been the single most important factor in the
acknowledged decline in the incidence of
NTD. Despite increased awareness of the benefit
of peri-conceptual folic acid and its widespread
use, there remains a significant incidence of the
problem. However, in women with a history of
folic acid intake and a SB child, it appears that
the severity of the lesion is much reduced.
39.4.1 Incidence
The worldwide incidence of NTD is reported to
be as high as 400,000 per annum. However, the
use of peri-conceptual folic acid has reduced the
incidence by 70% in the past 20 years and has
also reduced the severity of the lesion. In Ireland,
the rate fell from 32 in 1979 to 22 per 10,000 in
1982 and continues to fall.
39.5 Diagnosis
39.5.1 Antenatal
It is more preferable to make the diagnosis ante-
natally, which allows for counselling and facili-
tates transfer to a paediatric surgical centre for
appropriate management. In addition, if foetal
intervention is available and appropriate, then
this can be considered (vide infra). Antenatal
diagnosis may be made by careful ultrasound
examination, chorionic villous sampling, mater-
nal alpha-fetoprotein (AFP) or amniocentesis.
M. T. Corbally
Varying sensitivities have been reported, but in
experienced hands, ultrasound is a sensitive tech-
nique to detect a NTD. If antenatal ultrasound is
suspicious of a NTD, then maternal AFP com-
bined with amniocentesis for AFP and acetylcho-
linesterase assay is confirmatory.
With improved prenatal care, NTDs are com-
monly detected before birth, and arrangements
can then be made for counselling and for the
delivery and care of the infant. Delivery should
be scheduled close to a surgical centre and con-
sideration given to delivery by Caesarean section,
which may confer a significant functional benefit
to the child. Rarely, hydrocephalus may need
antenatal drainage to facilitate delivery.
39.6 Clinical Features
The diagnosis of a NTD is usually straightfor-
ward at birth if the lesion is a myelomeningo-
coele, meningocoele or encephalocoele. Lesser
lesions may not be clinically obvious and require
more detailed investigations.
39.6.1 Myelomeningocoele
A myelomeningocoele presents as a large open
lesion anywhere along the spinal column,
although the lumbar and sacral areas are the more
frequently involved. It is unfortunately the com-
monest form of NTD.  Typically, there is a thin
membrane covering the exposed neural plaque,
which may be intact and appear cystic. Usually,
however, the neural plaque is open to the environ-
ment, and CSF leaks constantly. If there is little
or no CSF leak, the lesion will be raised. There
may be occasional hamartomatous lesions asso-
ciated with it is, such as a haemangioma, a lipoma
or a naevus (Fig. 39.3). There may be associated
deformities of the lower limbs with hip disloca-
tion or subluxation, hypoplastic lower limbs,
genu recurvatum and talipes (Fig. 39.4). In addi-
tion, there may be obvious hydrocephalus
although this is uncommon at birth. The vertebral
anomalies can be significant with severe kypho-
sis evident at birth. Neurological deficits include
39  Spina Bifida and Encephalocoele
Fig. 39.3  MRI showing lipomyelomeningocoele
Fig. 39.4  Bilateral talipes deformities of the feet
motor and sensory loss to the lower limbs. The
effects of neural involvement include paralysis of
lower limb muscle groups often with p­ reservation
of nerve supply to antagonistic groups, which
results in more severe deformity. Occasionally,
there will be complete loss of innervation as in a
flaccid paralysis, while often, there will be an
upper motor neuron lesion and a resultant spastic
paresis.
While the internal anal sphincter is preserved
due to its autonomic nerve supply, the innerva-
491
tion to the external sphincter (puborectalis and
pelvic floor muscles) is likely to be lost, which
results in a patulous anus. This may result in rec-
tal prolapse and will not have any immediate
impact on bowel emptying, but clearly may affect
faecal continence later.
At least 90% of patients with myelomeningo-
coele have a neuropathic bladder with distur-
bances of detrusor and sphincter muscle activity.
This is manifest at birth by constant dribbling of
urine, but some do have an intermittent urinary
stream. The management of the child with a NTD
and a neuropathic bladder is quite complex and
should involve the surgeon and paediatric conti-
nence nurse specialist. Frequent monitoring by
renal ultrasound should be performed to allow
early detection and intervention in the presence of
upper tract dilatation, which occurs as a result
of poor bladder compliance. Early introduction
of clean intermittent catheterisation (CIC) may
be required in some infants.
39.6.2 Meningocoele
A meningocoele is uncommon and presents as a
sac and skin covered defect in the lower spinal
column with no abnormality of the underlying
neural tube. There is usually no neurological
defect, and the cord is normal. In addition, the
lower limbs are normal. Rarely, some neural
fibres may be adherent or contained within the
sac, and these require careful dissection from the
sac during closure. This may be apparent on
ultrasonic exam or MRI pre-operatively.
39.6.3 Spina Bifida Occulta
Spina bifida occulta may be difficult to detect on
clinical grounds and may only become apparent
at a later stage during incidental spinal imaging.
However, the presence of a tuft of hair, pigmented
naevus or vascular malformation in the midline
along the spinal column may indicate the under-
lying vertebral anomaly of occulta. While the spi-
nal cord is normal and there may be no overt
evidence of neurological impairment, it is impor-
492
tant to be aware that functional disorders of the
urinary tract may be related to an underlying
occulta and would warrant a search for an
­underlying lesion. Occasionally, there may be a
sinus-­like tract connecting to the meninges and
lying over the spine itself. This is an occasional
cause of spinal sepsis and meningitis. It should
not be confused with the quite common sacral or
posterior anal dimple, which is a skin dimple
attached to a normal coccyx and not related to
NTD.  It should be remembered that the defect
usually includes a vertebral arch abnormality and
normal cord and meninges but rarely, the cord
may be tethered and may be the cause of gait
abnormalities or subsequent bladder abnormali-
ties. Neurosurgical assessment is needed.
39.6.4 Encephalocoele
Encephalocoele is a midline defect in the bones
of the skull, which allows protrusion of meninges
only or gross herniation of brain tissue. In the lat-
ter condition, there may also be an associated
microcephaly or other macrostructural cerebral
anomalies. Often, these include Dandy-Walker
cyst formation, hydrocephalus and dysplasia of
the cerebellum and optic pathways. The usual
bony site is the occiput, but frontal encephalo-
coeles are more commonly seen in Asia.
There may be other congenital lesions, such as
NTD at other sites, cleft palate and cardiac, lung
and renal anomalies.
39.7 Management
39.7.1 Myelomeningocoele
Surgical closure of myelomeningocoele was not
regularly attempted until the early twentieth cen-
tury, when survival of 23% was reported. The
advent of asepsis and antibiotics improved sur-
vival, and surgical closure became more widely
practised. Patients with extensive paralysis and
hydrocephalus were not offered closure, but the
back lesion was allowed to slowly granulate and
epithelialise. However, death from infection and
M. T. Corbally
uncontrolled hydrocephalus was common.
Improved surgical and anaesthetic techniques
and antibiotics, and the development of reliable
valve-regulated shunts, led to more aggressive
management of patients even with severe lesions.
However, many survivors were noted to have a
poor quality of life with mental impairment,
shunt and renal problems that made their man-
agement difficult and tended to overwhelm exist-
ing medical resources. A review of the selection
process generated a return to a conservative
approach in patients with extensive paralysis,
severe hydrocephalus, kyphosis and major asso-
ciated anomalies, in the firm belief that the sever-
ity of the lesion was not compatible with an
acceptable quality of life. Reports of unselected
treatment for all patients with myelomeningo-
coele suggested that early mortality, the fre-
quency of mental impairment, poor mobility,
pressure sores, incontinence and other issues dic-
tated a selected approach for all patients.
However, the continued and increased survival of
patients initially regarded as being of poor poten-
tial outcome indicated that survival could not
always be based on the clinical appearance of the
lesion or extent of associated problems alone.
Moreover, children surviving this initial conser-
vative approach often suffered greater disabilities
as a result of a non-operative attitude. A decision
to withhold treatment cannot therefore be sup-
ported on clinical or ethical grounds alone. It
must be noted, however, that this approach is not
universally accepted and that parental wishes
must also be considered. Nevertheless, the cur-
rent standard of practice for children with myelo-
meningocoele is that the defect should be closed
within the first 24–48  h of life, to place a
ventriculo-­peritoneal (VP) shunt if hydrocepha-
lus is present and to monitor and treat aggres-
sively their problems for the rest of their life.
Patients with myelomeningocoele should be
transferred to a paediatric surgical centre and be
prepared for early closure of the defect. It is
important to protect the defect from contamina-
tion with faecal matter so chlorhexidine-soaked
gauze is applied to the lesion, and this should be
changed frequently. Broad-spectrum antibiotics
are usually given, and the baby is allowed to feed
39  Spina Bifida and Encephalocoele
on demand. Upon arrival at the surgical centre,
the baby undergoes a variety of investigations,
such as a cranial ultrasound, spinal X-ray (include
back, pelvis and skull), muscle charting, thor-
ough examination to rule out other problems,
orthopaedic assessment and a social work con-
sult. An MRI scan may also be performed at this
stage to document the presence or absence of
other spinal lesions, although this can be per-
formed at a later date and a spinal ultrasound is
probably just as sensitive in the first 6 weeks of
life. The surgeon should meet with both parents
and discuss the management plan in detail and, in
particular, the likely problems that may occur in
the future. These include the possible need to
treat hydrocephalus with a VP shunt; the possi-
bility of shunt malfunction and its consequences;
the likelihood of a neuropathic bladder and its
significance; the possibility of orthopaedic treat-
ment for talipes, dislocated hip, etc.; and the
issues of continence and intellect. The procedure
is then scheduled for the next available time but
should generally be within 24–48 h.
Although most patients with myelomeningo-
coele undergo surgical closure after birth, consid-
erable effort has been focused on in utero repair
in selected patients. This presents an alternative
in their management and may carry significant
advantages to the infant in terms of neurological
outcome. The proposed mechanism of improved
outcome with foetal intervention is to lessen the
hindbrain herniation associated with Arnold-­
Chiari malformation and so reduce the frequency
of significant hydrocephalus and shunt proce-
dures. The results of randomised trials, such as
the MOMS (Management of Myelomeningocoele
Study), which offered foetal intervention between
19 and 25 weeks of gestation, have shown that
foetal intervention is best performed at 25 weeks
of gestation to minimise preterm delivery, cho-
rioamniotic membrane separation and premature
labour. While not freely available at every paedi-
atric surgical centre, foetal surgical intervention
has some benefits over postnatal surgery as in:
1. Less significant hydrocephalus, as measured
by a significant incidence of shunt placement
at 12 months of age
493
2. Great chance of independent walking at 30
months
3. Less risk of any degree of hindbrain hernia-
tion at 12 months
However, it also carries a greater risk to the
mother, as in an increased risk of uterine dehis-
cence and placental abruption. Significant prema-
turity carries risks to the foetus/baby, as in
respiratory distress syndrome. A mini-­hysterotomy
may offset some of these adverse outcomes.
39.7.2 Operative Approach
The procedure is carried out under general anaes-
thesia, with the patient prone and in a warm
ambient temperature (Fig.  39.5). A small roll
may be placed beneath the hips and the lesion
and surrounding skin prepped with an aqueous
solution. It may be useful to cover the natal cleft
with a non-porous tape to exclude the area from
the sterile field.
Operating loupes are useful during all parts of
the procedure. It is wise to plan the orientation of
the incision before commencing the procedure as
this may impact on the ease of the closure espe-
cially with large lesions.
The skin edge is incised just at its junction
with the lesion and the membranes close to the
plaque carefully dissected from the plaque. The
plaque should be separated from all epithelial
elements so as to prevent a theoretical epithelioid
inclusion at a later date. When the plaque has
been freed from all local attachments, the dura is
incised on its lateral surface in an elliptical man-
ner around the neural plaque. This involves incis-
ing down to the underlying fascia and
subsequently mobilising the dura so as to allow
closure of the dura over the plaque. Closure is
effected by a running 6/0 or 7/0 suture through-
out the length of the dural sac. If possible, the
lumbar fascia can be mobilised to cover the dural
repair although this is not essential. A small
Redivac drain is left in situ and the skin closed
over the repaired defect. The skin is closed using
a series of interrupted nylon sutures with alternat-
ing Steri-Strips. If the skin closure seems a prob-
494
a c
b d
Fig. 39.5 (a) Position of the patient on the operating
table and an elliptical incision at the junction of the mem-
brane and the skin, (b) membrane being eplised to free the
lem, the skin may be undermined, or very rarely,
Z flaps are used. It is normal practice to cover the
repair with a semi-permeable dressing. Post-­
operatively, the patient is nursed prone or in a
lateral position, the drainage is monitored, and
antibiotics are continued until drain removal. If
there is excessive CSF leakage, it is generally not
due to a problem with dural closure but to increas-
ing hydrocephalus, and a ventriculo-peritoneal
shunt is indicated.
39.7.3 Meningocoele
Unlike myelomeningocoele, there is usually no
urgency to close a meningocoele. The surgical
procedure can be scheduled electively after
M. T. Corbally
neural plaque, (c) plaque lying in the dural layer, (d) dura
is closed with a continuous suture, (e) skin is closed with
interrupted sutures
appropriate investigations. There is little risk of
hydrocephalus although there may be neural ele-
ments adherent within the sac and an ultrasound
and/or MRI scan should be obtained pre-­
operatively. The skin edge is incised in an ellipti-
cal fashion around the defect and the protruding
dura exposed. The dural sac is opened vertically
on its lateral aspect taking care to avoid any
adherent neural tissue (rare) and the sac then
repaired, removing the herniating portion. The
skin is closed over a drain.
39.7.4 Encephalocoele
If the defect is small and contains little or no
brain tissue, then closure is within the experience
39  Spina Bifida and Encephalocoele 495

of a paediatric surgeon; however, if there is sig- the cerebellum and medulla through the foramen
nificant brain tissue and if there is associated magnum, elongation of the aqueduct of Sylvius
microcephaly, then a non-operative approach making it liable to blockage and various bony
may be indicated. Although there is generally defects of the upper cervical vertebra and occiput.
little urgency about closure of an encephalocoele,
there is the potential risk of further herniation,
which may compromise the child and make clo- 39.7.6 Clinical Features
sure more difficult. In general, these should be
closed as soon as possible after appropriate imag- The most obvious is that of a symmetrically
ing (MRI) is performed. enlarged head, either at birth or developing over
Surgery is performed with the child prone and the next few weeks. An ultrasound examination
intubated. In the case of occipital lesions, a trans-will easily show dilated ventricles, and serial
verse incision is made over the apex of the lesion. measurement of the head circumference will
The dural sac is exposed and opened away from show increasing deviation over the standard mea-
the bone edges. Brain tissue should be preserved surements. The anterior fontanelle is wide and
unless necrotic or grossly dysmorphic or likely to bulging, and the sutures will appear separated.
interfere with dural closure. The dura is closed Consideration should be given to ventriculo-­
with a continuous suture and a small drain left in peritoneal shunt insertion when there is a rapid
situ. An acute rise in intracranial pressure may increase in head circumference or when there is
require an urgent VP shunt following repair. clear evidence of significant hydrocephalus on
Anterior encephalocoeles and meningocoeles ultrasound, CT or MRI. Newborn infants with a
are complex and may require the input of other NTD and hydrocephalus tend not to have many
specialist services, such as neurosurgery and/or symptoms of increased intracranial pressure as
otolaryngology. the open fontanelles and sutures can accommo-
date to some extent. However, internal strabis-
mus and setting-sun sign due to pressure on soft
39.7.5 Hydrocephalus orbital plates are seen with significant and
untreated hydrocephalus. In addition, optic nerve
Hydrocephalus results from an imbalance in the damage or occipital lobe damage may result in
production and absorption of cerebrospinal fluid. visual deficits if the hydrocephalus is not treated.
Obstruction to the flow of CSF out of the ven- A description of the technique of VP shunt
tricular system by the Arnold-Chiari malforma- insertion is beyond the scope of this chapter.
tion, tumour, aqueductal stenosis, haemorrhage However, shunt valves are selected on their open-
or obstruction of the fourth ventricle (Dandy-­ ing pressure, which is the pressure that the valve
Walker cyst) causes a non-communicating hydro- will open to allow CSF leave the ventricular sys-
cephalus and is the most common type seen. Free tem and enter the peritoneal cavity. Since it is
flow of CSF due to lesions of the choroid plexus possible to overdrain and cause a slit ventricle
or following inflammatory conditions causes syndrome, it is probably best not to use low pres-
communicating hydrocephalus. In the newborn, sure systems except in the very small, preterm
increasing pressure within the ventricles and cra- infant. The author’s preference is to use a unitised
nial vault is somewhat compensated by the open single medium pressure system in the majority of
fontanelles. Most patients with myelomeningo- cases.
coele have hydrocephalus, and approximately Shunts are mechanical devices and are subject
90% of these will ultimately require insertion of to problems such as blockage, breakage,
a VP shunt to control it. In this group, hydroceph- ­malfunction and infection. Many or all of these
alus is associated with the Arnold-Chiari malfor- problems can be found in the life of a single
mation, which includes caudal displacement of shunt.
496
39.8 Long-Term Management
Most patients survive the trauma of back closure
and surgical treatment of hydrocephalus.
However, as many as 23% may die within 1 year
of birth. Surviving patients face ongoing prob-
lems in relation to mobility, shunt issues, prob-
lems with faecal and urinary continence, problems
specific to their neuropathic bladder and educa-
tional, intellectual and social issues. As many as
75% have normal intelligence, but a significant
number require special educational support. Many
of these problems can and perhaps should be
addressed at a special clinic to cope with the needs
of this particular group. Individual care plans can
readily be worked out for each patient, and this
can be modified on a fluid basis at the SB clinic. It
can often be difficult to have all involved special-
ists attend such a clinic, but the benefits to the par-
ents and child are significant. Increasingly, local
agencies provide comprehensive care to NTD
patients and their families, such as Enable Ireland,
and the impact of a variety of voluntary agencies,
such as “The Association for Spina Bifida and
Hydrocephalus”, cannot be overstated. Of partic-
ular importance is the ongoing surveillance of the
urinary tract, especially the results of 6 monthly
renal ultrasound exams. This is chiefly to detect
the early development of upper urinary tract dila-
tation due to a non-compliant neuropathic blad-
der. When this occurs, the parents are instructed in
the technique of clean intermittent catheterisation
(CIC), which is performed every 4 h on average
and serves to empty the bladder of urine and pre-
vent reflux of static urine from high vesical pres-
sures. Approximately 10–15% of patients will not
have their high pressure bladder controlled by
CIC, and a vesicostomy may be necessary. Older
children and children with small volume high
pressure bladders will need the expertise of a pae-
diatric urologist to assess the need for bladder
augmentation. The input of paediatric continence
nurses is invaluable, especially in the performance
of urodynamic studies and also in instructing
older children in the often difficult task of over-
coming manipulative skills to facilitate self CIC.
Bowel problems are generally treated using
medications such as stool softeners and/or regu-
M. T. Corbally
lar stimulant enemas; however, social “conti-
nence” can be achieved in up to 85–90% of
children using a regular washout enema contain-
ing 200–300 ml of water and a stimulant such as
toilax. This generally gives clean results lasting
up to 24 h. When this fails, consideration should
be given to performing an antegrade colonic
enema (ACE) procedure using the appendix as a
catheterisable conduit.
Ongoing issues of mobility and joint deformi-
ties need the continued input of orthopaedic sur-
geons and occupational physiotherapists. In
addition, social workers play a significant role in
helping the family adapt the home environment
to cope with mobility and toileting issues and to
secure proper state funding for their needs.
The incidence of NTD continues to decline,
but for those born with this condition, it can
impose severe restrictions on the quality of their
lives. The current standard of practice is that all
patients with myelomeningocoele should be
offered surgical repair within the first 24–48 h of
life. Improvements in valve regulated ventriculo-­
peritoneal shunts have contributed greatly to
quality of life. Long-term review in special mul-
tidisciplinary clinics facilitates review of renal
function, status of urinary tracts, status of their
valve-shunt and management of continence and
education and social issues.
39.9 Conclusion
The incidence of NTDs continues to decline.
This may be due to the combined impact of peri-­
conceptual folic acid, antenatal diagnosis and ter-
mination of pregnancy. Early surgical closure is
important to reduce infective complications.
Foetal surgery between 19 and 25 weeks has a
role and can improve overall outcomes and per-
formance; however, it is limited to select centres
and includes the risk of premature delivery with
its attendant problems. Following the initial mea-
sures to close the back and secure adequate CSF
drainage, the myriad of issues that can envelop a
child with a NTD and the family are best dealt
with at a multidisciplinary clinic. This is espe-
cially important to provide a one-stop assessment
39  Spina Bifida and Encephalocoele
of shunt performance, mobility and continence
issues. Early back closure is important to prevent
complications such as infection. Monitoring for
hydrocephalus is important in order to insert a
shunt at an appropriate time and thus preserve
function.
Selected References
Corbally MT (2006) Spina bifida. In: Puri P, Höllwarth
ME (eds) Pediatric surgery. Springer surgery atlas
series. Springer, Berlin, pp 419–426
Martin T. Corbally (2009) in Pediatric Surgery, Eds Puri
and Hollwarth. Chapter 79. 765–774, 2009. Pubs
Springer
Dias MS (2005) Neurosurgical management of myelome-
ningocoele. Pediatric Rev 26(2):50–60
497
Finnell RH, Gould A, Spiegelstein O (2003) Pathobiology
and genetics of neural tube defect. Epilepsia 44(Suppl
3):14–23
Hunt GM (1990) Open spina bifida: Outcome for a com-
plete cohort treated unselectively and followed into
adulthood. Dev Med Child Neurol 32:108–118
Larsen WJ (2001) In: Sherman LS, Potter SS, Scott
WJ (eds) Human embryology, 3rd edn. Churchill
Livingstone, New York
Lorber J (1971) Results of treatment of myelomeningo-
coele: An analysis of 524 unselected cases, with spe-
cial reference to possible selection for treatment. Dev
Med Child Neurol 13:279–303
Mitchell LE, Azdick NS (2004) Spina bifida. Lancet
364:1885–1895
Sacco A, Ushakov F, Thompson D et  al (2019) Fetal
surgery for open spina bifida. Obstet Gynaecol
21:271–282
Walsh DS, Adzick NS (2003) Foetal surgery for spina
bifida. Semin Neonatol 8:197–205
 Hydrocephalus
Geraint Sunderland, Jonathan Ellenbogen,
and Conor Mallucci
40.1 Introduction
The central dogma of hydrocephalus is a simple
one: cerebrospinal fluid (CSF) is constantly pro-
duced and reabsorbed. Disturbances in CSF flow
and absorption lead to an accumulation within the
ventricular system. Untreated, this leads to rising
intracranial pressure (ICP) and active distension
of ventricles in the majority of cases (Rekate
2011). The management of hydrocephalus is far
from simple, however, as causes are numerous
and heterogeneous. There are many nuances, and
expert management is essential. Patients will
often have multi-system disorders and complex
needs, and thus there are some fundamental prin-
ciples all clinicians involved in their care should
know. Hydrocephalus is really a catch-all term
encompassing many disparate aetiologies, each
with their own vagaries of treatment. Defining
hydrocephalus is thus problematic; however, the
vital concern is whether there is associated raised
ICP.  Ventricular volume, while a useful adjunct
especially in the radiologically surveillance of
patients or in the acute setting, is secondary.
Small ventricles ↑ normal intracranial pressure
G. Sunderland (*) · J. Ellenbogen · C. Mallucci
Department of Neurosurgery, Alder Hey Children’s
NHS Foundation Trust, Liverpool, UK
e-mail: geraintsunderland@doctors.org.uk;
jellenbogen@doctors.org.uk
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_40
40
Radiological definitions, therefore, such as
Evans ratio (ventriculomegaly=(maximal frontal
horn diameter)/(transverse inner skull diame-
ter) ≥ 0.3), the diameter (>2 mm) of the temporal
horns, rounded third ventricle walls, etc. are
crude guides and not universally applicable.
The aim of this chapter is not to provide an
exhaustive guide to hydrocephalus in all its
guises, but rather to provide the non-specialist
with a practical approach to the management of
patients with hydrocephalus, both treated and
untreated.
40.2 Historical Overview
Hydrocephalus has been recognised as a patho-
logical entity and studied since the time of
Hippocrates in the fifth century BC. All the key
figures of the history of medicine, including, but
not limited to, Galen, Vesalius, Sylvius and
Abulcasis, have at one point turned their enquir-
ing minds to its study. Early writers sought to
understand its origin, believing the soul to reside
within the ventricles (Aschoff et al. 1999).
Progress in understanding the pathology did
not occur until the physiology of CSF production
was explored, firstly by Thomas Willis in the sev-
enteenth century who identified the choroid
plexus as the organ of origin. The CSF circula-
tion was defined between the nineteenth and into
the twentieth century with two of the contributors
499
500
immortalised in anatomical nomenclature:
Magendie and Luschka. In the early twentieth
century, the practice of neurosurgery as we know
it today was born due to pioneers such as Harvey
Cushing, Sir Victor Horsley and Walter Dandy
who all had a hand in developing hydrocephalus
surgery (Lifshutz and Johnson 2001).
The first documented shunt surgery was per-
formed in 1905, but the technique subsequently
fell into disrepute due to high mortality. There
was a resurgence in the 1950s, thanks to Frank
Nulsen and Eugene Spitz ,who pioneered ven-
triculoatrial shunting and subsequently ventricu-
loperitoneal shunting (Rachel 1999). Spitz also
developed the first working valve alongside
hydraulics engineer, John Holter. Numerous sub-
sequent valve designs have been developed
including the Wade-Dahl-Till valve named after
its designer’s hydraulics engineer, Stanley Wade,
world-famous children’s author Roald Dahl and
neurosurgeon Kenneth Till. The development of
CSF shunting revolutionised the management of
hydrocephalus, transforming it from a disabling
and often fatal condition to an eminently treat-
able one.
40.3 Incidence
The prevalence of hydrocephalus in paediatric
populations is 88/100,000 globally (Dewan
et al. 2018a). Hydrocephalus is less prevalent in
younger adults (11/100,000), but peaks in the
elderly (175/100,000) in the over 65s and is
>400/100,000  in those over 80, due to the
occurrence of normal pressure hydrocephalus
(NPH). It is also associated with congenital
malformations of the brain and spinal cord in a
significant proportion of cases in infants.
Timely and effective surgical treatment is
essential, as complications related to delay in
treatment can be life-threatening. Once treated,
the majority of patients will require lifelong
support and follow-­up, with up to 85% requir-
ing further surgical intervention in their life-
time (Stone et al. 2013).
G. Sunderland et al.
40.4 Aetiopathogenesis
Doctors love good taxonomy, and those inter-
ested in hydrocephalus are no different; thus,
numerous classifications have been devised to
varying success. Classically, hydrocephalus was
either ‘obstructive’ or ‘communicating’, terms
coined by one of the fathers of modern neurosur-
gery Walter Dandy in 1914. Subsequent grouping
by aetiology into ‘congenital’ or ‘acquired’
causes has been used, but has little clinical rele-
vance (Dandy and Blackfan 1914).
The ‘obstruction’ in obstructive hydrocepha-
lus relates to a mass lesion (tumour, swelling) or
some physical impediment (pus, blood) to the
CSF circulation pathway prior to the point of
reabsorption. Communicating hydrocephalus
occurs due to obstruction also, though the
obstruction is more functional, existing at the
point of reabsorption, the prime example of com-
municating hydrocephalus being blood and blood
breakdown products occluding the fine filtration
substrate of the arachnoid granulations in cases
of intraventricular haemorrhage. The terms
obstructive and communicating are increasingly
outdated and, in fact, give a gross over-­
simplification of a hugely complex and still
poorly understood physiology (Tomycz et  al.
2017). They do, however, provide a workable
scheme around which we can make clinical deci-
sions. The exception to this two-classification
rule is where CSF production outstrips reabsorp-
tion. This is very rare, however, limited to cho-
roid plexus tumours and in some circumstances
of infection, such as ventriculitis, due to inflam-
mation and increased ependymal blood flow.
In practical terms, whether ‘obstructive’ or
‘communicating’, CSF accumulates within the
cranial CSF compartment, and pressure rises.
The normal CSF pressure is 10–15  cm/H2O, as
measured in the lumbar theca in the lateral posi-
tion (7–12 cm/H2O in infants). Unchecked eleva-
tions in ICP can lead to coma and eventually
death.
Hydrocephalic aetiology and age are inextri-
cably linked resulting in the three peaks in inci-
40 Hydrocephalus
dence. The vast majority of neonates and infants
will develop hydrocephalus either secondary to
intraventricular haemorrhage (IVH) of prematu-
rity or in association with congenital malforma-
tions, chief amongst which is spina bifida. The
rest of the cases (at least in the developed world)
are due to tumours, with a small number occur-
ring following episodes of meningoencephalitis
(the second leading cause globally) (Dewan et al.
2018b). During adulthood and middle age, the
incidence is low and typically limited to cases of
subarachnoid haemorrhage and as a consequence
of obstructive tumours. With advancing years,
the neurodegenerative condition, idiopathic nor-
mal pressure hydrocephalus (NPH), results in a
third and final peak of hydrocephalus cases, sum-
marised in Table 40.1. This division into different
age groups and how this impacts on management
neatly demonstrates the importance of under-
standing the pathoanatomical basis of a patient’s
hydrocephalus. Careful consideration and under-
standing of this is crucial to making rational man-
agement decisions.
Table 40.1  Common aetiologies causing hydrocephalus
by age group
Age group Common aetiologies
Neonates/ • Post-haemorrhagic (IVH of prematurity)
infants • Post-infective (meningitis/ventriculitis)
• Spina bifida and related neural tube
defects
• Other congenital malformations
(craniofacial syndromes, arachnoid
cysts, Dandy-Walker malformation,
X-linked aqueductal stenosis, etc.)
Children • Tumours (benign/malignant)
• Post-haemorrhagic (vascular
malformations)
• Aqueduct stenosis
• Post-infective (meningitis/ventriculitis)
• Idiopathic intracranial hypertension
Adults • Post-haemorrhagic (aneurysmal
(<65) subarachnoid haemorrhage)
• Tumours (benign/malignant)
• Idiopathic intracranial hypertension
• Trauma
Adults • Idiopathic normal pressure
(>65) hydrocephalus
• Post-haemorrhagic (aneurysmal
subarachnoid haemorrhage)
501
40.5 Pathophysiology
Understanding the pathophysiology of hydro-
cephalus requires some understanding of the
physiology of CSF production and circulation.
Approximately 80% of CSF is produced by the
choroid plexus; tufts of capillaries enclosed in
modified ependyma (membranous lining of the
ventricles). This active process is supplemented
(20%) by passive egress of fluid from the brain
substance into the interstitial space, perivascular
‘Virchow-Robin’ spaces and onwards then to the
body of CSF within the ventricular system.
CSF is produced at around 0.33 ml/min, result-
ing in a total of 20 ml/h and about 500 ml/day in
an adult (Kimelberg 2004). At any point, there is
about 150 ml of CSF within the neuroaxis, half of
which is in the intracranial compartment. Volumes
are obviously less in infants, proportional to body
weight, but approximate adult values are reached
by 5 years of age. Choroid plexus is found pre-
dominantly within the lateral ventricles but also
within the third and fourth ventricles. From the
point of secretion, CSF normally circulates from
the lateral ventricles through the foramina of
Monro into the third ventricle and then onwards
through the aqueduct of Sylvius into the fourth
ventricle in the posterior fossa. From the fourth
ventricle, the CSF exits into the spinal subarach-
noid space either via the midline foramen of
Magendie or via the lateral foramina of Luschka
into the basal cisterns and around the cerebral cor-
tex in the subarachnoid space (Sakka et al. 2011).
CSF is then reabsorbed into the cerebral venous
system, via outpouchings of arachnoid mater
known as arachnoid granulations. The arachnoid
granulations bulge into the dural venous sinuses,
where CSF is resorbed depending on a hydrostatic
pressure differential (Khasawneh et al. 2018).
40.6 Pathology
One of the key factors to consider, certainly when
timing surgical intervention in young children, is
whether the skull plates are fused at the cranial
502
sutures. A child with a closed fontanelle and fused
sutures will not be able to accommodate for a
relative increase in CSF volume without incurring
dangerous elevations in ICP, for example, as their
ICP dynamics conform to the Monro-Kellie doc-
trine (Fig. 40.1). Conversely, a premature neonate
may accommodate, usually with a rapidly increas-
ing head circumference. This can be temporised
relatively safely with episodic ventricular taps.
Fig. 40.1  Monro-Kellie doctrine
G. Sunderland et al.
The cranial vault is a fixed volume (approx.
1700 ml). The contents are the brain (1400 ml),
CSF (150 ml) and blood (150 ml).
An increase in one of these components
requires displacement, and a reduction in one or
both of the other two components or the intracra-
nial pressure will rise.
40 Hydrocephalus
40.7 Diagnosis
40.7.1 Clinical Presentation
Clinically speaking, the vast majority of hydro-
cephalus will present one of five ways:
1. On serial imaging of the premature neonate at
risk for germinal matrix haemorrhage and
intraventricular haemorrhage
2. Following investigations for expanding head
circumference—usually instigated in the
community by a GP or health visitor
3. Following imaging and investigations per-
formed for symptoms of raised ICP
(Table 40.2)
Table 40.2  Symptoms and signs of raised ICP by age
group
Age group Symptoms and signs
Neonates and • Irritability
infants • Floppy, peripherally cool
• Reduced responsiveness
• Vomiting
• Posturing (arching back)
• Sunsetting eyes (combination of
upper eyelid retraction and failure
of upgaze)
• Distended scalp veins
• Bulging/tense anterior fontanelle
• Splaying of suture lines
• Crossing centiles on head
circumference chart
• Desaturation/bradycardic episodes
Young children • Headache
• Clutching head
• Vomiting
• Non-specific behavioural change
• Change in appetite/anorexia
• Collapse (hydrocephalic attack—
associated with transient increases
in ICP: coughing, sneezing, etc.)
• Sunsetting eyes
• Crossing centile lines on head
circumference chart
• Head circumference out of
proportion with body weight/
height
• Visual disturbance (blurring/visual
field defect)
• Papilloedema
503
Table 40.2 (continued)
Age group Symptoms and signs
Older children • Headache
and adolescents • Vomiting
• Collapse (hydrocephalic attack—
associated with transient increases
in ICP: coughing, sneezing, etc.)
• Visual disturbance (blurring/visual
field defect)
• Papilloedema
4. Subsequent to routine fundoscopic examina-
tion by the child’s optometrist and identifica-
tion of papilloedema
5. Incidentally discovered when investigated for
an unrelated cause, e.g. trauma
Radiological imaging forms the foundation of
neurosurgical diagnosis. Cranial imaging is
essential to the diagnosis of hydrocephalus, typi-
cally based on the presence of ventriculomegaly.
It bears repeating, however, that normal-sized
ventricles do not necessarily mean there is no
problem with CSF flow and/or raised intracranial
pressure (Dinçer and Özek 2011).
40.7.2 Plain Radiography
Plain radiographs of the skull do not have a role
in the routine workup of hydrocephalus, although
they can give clues to the underlying pathology.
Signs of raised pressure include a ‘copper beaten’
or ‘thumbprinted’ appearance of the calvarium
and splaying of the cranial sutures (Fig. 40.2). In
addition, the aetiology of the raised pressure may
be noted, e.g. traumatic fractures or abnormal
calcification associated with a pineal region mass
obstructing the aqueduct. Plain radiographs are,
however, vital in the longer-term management of
the hydrocephalus patient. Shunt series XR
(skull, neck soft tissues, chest and/or abdomen)
allow clinicians to follow the tract of the shunt
(radio-opaque) to ensure all components are in
continuity and exclude a shunt ‘fracture’ or dis-
connection, which could compromise shunt func-
504
Fig. 40.2  Plain radiograph of infant skull demonstrating
the ‘copper beaten’ or ‘thumbprinted’ appearance of the
calvarium due to moulding of the inner table over the
cerebral cortex reflecting high ICP.  Note the prominent
coronal suture. Also note the discontinuity in the shunt
(arrow) where the distal catheter has disconnected from
the valve
tion (Fig. 40.3). Furthermore, a skull XR permits
accurate evaluation of the setting in programma-
ble shunt valves (see Sect. 40.9.1).
40.7.3 Ultrasonography
Transcranial ultrasound is a very useful screen-
ing tool in neonates. It requires an open anterior
fontanelle, which provides a window to assess
intracranial structures. With minimal training
and experience, operators (typically neonatolo-
gists) can achieve a diagnosis of intracerebral/
intraventricular haemorrhage, with or without
ventriculomegaly. It is vital to record the ven-
tricular index (VI) to communicate effectively
with colleagues. VI is defined as the width (in
mm) from the falx to the lateral wall of the lat-
eral ventricle at the level of the foramen of
Monro, when viewed in the coronal plane
(Fig. 40.4). As there are both left and right ven-
tricles, a value is given for each as these are fre-
quently subtly different. Blood within the
ventricles, for example, in cases of IVH of pre-
G. Sunderland et al.
maturity, will be echo-bright and should be obvi-
ous to all (Fig.  40.4). Skilled and experienced
operators can identify other underlying aetiolo-
gies, such as tumours or anatomical abnormali-
ties; however, ultrasound is not ideally suited to
this, and confirmatory cross-sectional imaging
would be required (Dorner et al. 2018).
40.7.4 Computed Tomography (CT)
Computed tomography (CT) is a ubiquitous,
rapid and easy-to-interpret tool in the diagnosis
of hydrocephalus. A plain unenhanced CT will
successfully diagnose hydrocephalus and the
underlying aetiology in the vast majority of
cases. CT diagnosis of hydrocephalus is based
primarily on the presence or absence of ventricu-
lomegaly (Fig. 40.5). Features suggestive of ven-
triculomegaly include ballooning of the frontal
horns of the lateral ventricles, the presence of
dilated temporal horns of the lateral ventricles
(not typically visible in health) and periventricu-
lar low density suggesting transependymal pas-
sage of CSF under pressure. Other features
include enlarged and rounded third ventricle,
effacement of the cortical sulcal pattern and sur-
face subarachnoid spaces especially at the vertex
of the skull and Sylvian fissure and upward bow-
ing of the corpus callosum. Radiological criteria
quoted in the literature, such as Evan’s ratio, are
not routinely used in current clinical practice, and
imaging must be interpreted in the clinical con-
text (history and examination) and with reference
to previous imaging (Dinçer and Özek 2011).
CT is also frequently used to assess patients
after insertion or revision of ventricular catheters
for CSF diversion. The scan is useful to confirm
satisfactory positioning as well as the state and
size of the ventricles and can also reassure sur-
geons of the absence of complications, e.g. post-­
operative haematoma or IVH, in the event of
difficult shunt revision surgery. Many neurosur-
geons will perform a baseline or ‘well’ CT sev-
eral weeks after recovery from shunt insertion or
revision to capture the ventricular configuration
at baseline. This can be used as a comparator
40 Hydrocephalus 505

a b

c d

Fig. 40.3 (a–d) Plain radiograph ‘shunt series XR’ dem- tubing through its course, the programmable valve evident
onstrating the course of a right parietal VP shunt transiting on the cranial XR and the distal tubing coiled in the
from the cranium to abdomen. Note the continuity of the abdomen
506
should the patient present with suspicion of shunt
dysfunction in the future (Pople 2002).
Despite the many advantages, CT is unfortu-
nately less good at defining underlying causes
VI
Fig. 40.4  Cranial ultrasound taken in the coronal plane
via the anterior fontanelle in a premature neonate. Note
intraventricular haemorrhage (yellow arrow), dilated
ventricles (red arrows) and ventricular index (VI)
demonstrated
Fig. 40.5 Axial unenhanced CT demonstrating gross
acute hydrocephalus. Note ballooning of the frontal horns
of the lateral ventricles, dilated rounded third ventricle
G. Sunderland et al.
than is magnetic resonance imaging (MRI). In
addition, it exposes the child to radiation, the
cumulative doses of which can be significant
across a lifetime of treatment.
40.7.5 Magnetic Resonance Imaging
(MRI)
MRI is the investigation of choice for delineating
the ventricular size and the pathoanatomical sub-
strate of hydrocephalus. It has many advantages
over CT, not least of which is the avoidance ion-
ising radiation exposure. As with CT scanning,
multiplanar imaging allows the assessment of
anatomical structures but with greater tissue
determination. Fine membranous structures, such
as the floor of the third ventricle, can be identi-
fied; useful as this is typically bowed down and
backwards into the pre-pontine space, particu-
larly in cases of obstructive hydrocephalus where
the obstruction is at the level of the aqueduct of
Sylvius or fourth ventricle. Modern scanning
technology permits high-resolution and accurate
and prominent occipital horns. Note the periventricular
lucency around the frontal horns (white arrows), which
represents transependymal flow of CSF under pressure
40 Hydrocephalus
a b
c d
Fig. 40.6 (a) Axial T2-weighted MRI demonstrating
multi-loculated hydrocephalus due to multiple septations
(red arrows). (b) Axial T2 FLAIR demonstrating hydro-
cephalus with periventricular oedema due to transependy-
mal flow. (c) Coronal T2-weighted MRI showing marked
image representation of CSF spaces and the abil-
ity to perform assessments of CSF flow. This can
allow the identification of septations or cysts
within the ventricles, common following intra-
ventricular haemorrhage and infection. These
would need to be traversed and fenestrated in
order to successfully treat that patient’s hydro-
507
chronic hydrocephalus with expanded temporal horns
(yellow arrows). (d) Sagittal T2-weighted MRI.  Note
expanded third ventricle with thinned corpus callosum
and downward bowing of the third ventricle floor (blue
arrows)
cephalus (Fig.  40.6). Pathological lesions, such
as tumours, vascular malformations (e.g. vein of
Galen malformation), cysts and even anatomical
variants, are identified with an extremely high
degree of accuracy by MRI; thus, for the assess-
ment of de novo hydrocephalus, MRI is
mandated.
508
Like CT, MRI has a role in the post-operative
imaging of patients with hydrocephalus; it is par-
ticularly useful in the assessment of internal CSF
diversion procedures, such as endoscopic third
ventriculostomy, where particular imaging
­paradigms (e.g. time-resolved 2D phase contrast)
can be applied directly to confirm and even quan-
tify CSF flow, for example, across a newly
formed stoma or through a pre-existing CSF
pathway, e.g. across the craniocervical junction,
in cases of Chiari malformation (Fig.  40.13)
(Dinçer et al. 2011).
40.8 Differential Diagnosis
The differential diagnosis for a child presenting
with symptoms and signs of raised intracranial
pressure is wide. Hydrocephalus is rarely the pri-
mary disease entity but rather a consequence of
other underlying pathology. Accurate diagnosis
obtained from imaging, blood tests, clinical his-
tory and examination will identify an underlying
cause in the vast majority of cases.
40.9 Management
Hydrocephalus is an eminently manageable con-
dition surgically speaking, albeit this may be
complicated in a small proportion of cases. The
treatment involves CSF diversion or ‘shunting’
from the intraventricular compartment of the
brain to another compartment either intra- or
extracranial. These shunts may be physically
implanted or internally formed by creating an
ostomy between cavities. The absolute indication
for the treatment of hydrocephalus is the pres-
ence of signs and symptoms of raised intracranial
pressure (Table 40.2). Caution should be applied
and consideration of alternate management strat-
egies made where available, especially in the
presence of active/proven CNS/CSF infection,
low body weight or other remediable causes. The
mode of treatment is dependent on a multitude of
factors outlined below.
Lesional hydrocephalus, i.e. secondary to a
focal mass lesion causing CSF pathway obstruc-
G. Sunderland et al.
tion, is usually best treated by the removal of the
causative lesion. Examples of these might include
an enlarging colloid cyst obstructing both foram-
ina of Monro or a tumour within the posterior
fossa compromising CSF egress via the fourth
ventricle. In the case of acute symptomatic
hydrocephalus, secondary to an operable or
potentially operable lesion, a short-term CSF
diversion may be employed to gain control of the
immediate situation allowing time for further
assessment. Insertion of an external ventricular
drain (EVD) into the frontal horn of the lateral
ventricle (by convention the right side as this is
non-dominant in the majority of individuals)
allows control of raised ICP and more detailed
investigations and definitive treatment.
Occasionally, despite resection of the offending
lesion, the hydrocephalus persists, and in these
cases, diversion of CSF to an alternative site of
reabsorption is required.
40.9.1 Implantable CSF Shunts
Implantation of a CSF shunt is one of the most
commonly performed procedures in neurosur-
gery worldwide. While one of the core proce-
dures a neurosurgeon masters at the very
beginning of their training and, thus, often dis-
dained as ‘just a shunt’, one needs to be wary of
creeping complacency. The consequences and
potential long-term sequelae of a poorly thought-­
out or carelessly performed operation are serious
and potentially dangerous. Complications of
shunt surgery are detailed later in this chapter.
CSF flow is most frequently diverted into the
peritoneal cavity via a ventriculoperitoneal or
‘VP’ shunt (VPS). Alternatives include direct
drainage into the central venous circulation, a
ventriculoatrial or ‘VA’ shunt (VAS) and much
less frequently the pleural cavity (ventriculopleu-
ral shunt).
Shunts consist of a proximal (ventricular)
catheter, a distal catheter and an interposed one-­
way valve. A valveless system may be implanted,
though outside of the developing world these are
limited to complex cases, usually following mul-
tiple failed revision surgeries. Shunt catheters are
40 Hydrocephalus
manufactured from hydrophobic silicone which
is biologically inert, resists bacterial colonisation
and retains its flexibility over decades, all crucial
to longevity. Over the last two decades, manufac-
turers have devised numerous ways to improve
the durability of shunts and avoid the two main
complications, blockage and infection. There is
level I (double-blind RCT) evidence to support
the preference of antibiotic (rifampicin and
clindamycin)-impregnated tubing to reduce the
rate of infection-related failure and revision sur-
gery (Mallucci et al. 2019).
The major differences between shunt systems
are in the valve, and there is a wealth of choice
available on the market. It should be stated at the
outset that all evidence to date, including one
RCT published in 1998, has failed to categori-
cally establish superiority of one valve design
over another (Drake et  al. 1998; Kestle et  al.
2000). More research in this area is needed.
The most frequent valves encountered are dif-
ferential pressure valves (DPVs). These rely on a
hydrostatic pressure gradient across the valve
from inlet to outlet to drive valve opening and
CSF flow. The alternative, flow regulating valves,
are more rarely used. DPVs may have a fixed
opening pressure (most common); manufacturers
a b
Fig. 40.7  Pictorial illustration of the process of shunt
valve reprogramming using the MIETHKE proGAV 2.0
system, one of the numerous programmable shunts on the
509
supply valves with a variety of pressure settings
or that alternatively have a programmable func-
tion whereby the opening pressure can be
adjusted manually by the use of a transcutaneous
handheld device (Fig.  40.7). It is important to
realise that although valves may be of equivalent
opening pressures, they may have different inter-
nal resistances and behave quite differently
in vivo. Numerous other adjustments and techni-
cal features, including gravity-assisted valves
and anti-syphon devices, are available, but their
discussion alone could fill a book and is thus out-
side of the scope of this chapter.
Selection of valves and shunt types is ulti-
mately a matter of preference and tailoring to
patient and clinician requirements. To the
untrained observer, these choices may seem ran-
dom but are often based on individual or depart-
mental collective experience (sometimes painful).
Programmable valves, for example, may be
inserted to allow adjustment in demand as young
children (especially neonates) grow or in the
elderly with NPH to prevent CSF overdrainage
and complications thereafter (Serarslan et  al.
2017). Gravity-assisted valves might be preferred
again in younger patients to accommodate for
rapid growth and change from predominantly
c
market. (a) The setting is first confirmed. (b) A strong
magnetic ring is used to reprogram the valve to the desired
setting. (c) Post-reprogramming setting confirmation
510
recumbent to upright posture and the effect these
have on the ‘syphoning’ effect (Schatlo et  al.
2013). Finally, pragmatic concerns are also valid;
considerations such as the size of the valve and
the risk of tissue breakdown over it (particularly
in neonates), or favouring one programmable
interface over another, are taken into account and
weighted appropriately.
For general surgeons who may come across
intraperitoneal catheters, it is important to be
aware that a significant proportion of resistance to
CSF flow comes the distal catheter and is very
much related to its length. Shortening a catheter in
the peritoneum can have a marked impact on the
function of shunt system so care should be taken
to preserve shunt catheters when encountered.
40.9.2 Insertion of VP Shunt:
The Technique
After induction of general anaesthesia and endo-
tracheal intubation, the patient receives prophylac-
tic intravenous antibiotics and is positioned on the
operating table. The positioning should allow
access to the insertion site on the cranium (usually
parieto-occipital or frontal) and the site of distal
implantation (neck, chest or abdomen). The ideal
insertion point permits a catheter trajectory which
accesses a body of CSF (typically the lateral ven-
tricle) but which traverses the least amount of
brain, avoiding eloquent cortex and crucial deep
brain structures (basal ganglia). The tip of the
catheter bears multiple perforations, and these
should be lying free in CSF away from the ventri-
cle wall and choroid plexus to prevent ingrowth of
tissue and thereby catheter blockage. In the cir-
cumstance of small or ‘slit’ ventricles, this may
not be possible. A sub-optimally placed shunt may
still work but will be more prone to failure. A num-
ber of ‘classical’ cranial insertion points have been
described based on surface anatomy; however,
these pre-date the CT scan and image guidance era
and so are now used to approximate insertion only.
The commonly used reference points include:
In the parietal position: Keen’s point, three fin-
ger’s breadths above and behind the pinna and
G. Sunderland et al.
Frazier’s point, 6 cm cephalad from the inion
and 3 cm from the midline.
Frontally: Kocher’s point dictates that burr holes
are placed in the midpupillary line just ante-
rior to the coronal suture.
As noted above, the standard of care now is to
insert ventricular catheters using image guid-
ance. There are a number of systems available,
but broadly speaking, they utilise either real-time
ultrasound guidance or cross-sectional (CT or
MRI) data. Ultrasound guidance permits accurate
localisation of the fluid-filled ventricles, and
operators can first adjust to a suitable trajectory
and then guide the echo-bright catheter to the
desired target. Neuronavigation systems require
volume cross-sectional imaging to be performed
pre-operatively. This is then uploaded to the
image guidance console where entry point, target
and resulting trajectory can be planned.
Avoidance of blood vessels and vital structures is
thus ensured. These systems may rely on optical
or electromagnetic (EM) technology. Optical
systems use an infrared camera which ‘sees’
reflective markers attached to working instru-
ments and a reference array which remains fixed
relative to the patient’s head. EM technology uses
an EM emitter, which creates a magnetic field
around the target. The relative positions of a ref-
erence probe (attached to the patient) and work-
ing instruments are then computed. The results
are equivocal; a virtual, real-time representation
of your instrument’s position in 3D space relayed
on a multiplanar (axial/coronal/sagittal or even
3D) reconstruction of the patient’s scan with sub-­
millimetric accuracy.
Shunt insertion is best done by two surgeons,
with the opening of the cranium and the perito-
neal cavity done simultaneously to shorten oper-
ative time. The cranial opening should be done
with a small burr hole and a minimal dural open-
ing, just wide enough to admit the catheter to
minimise CSF bypass. In neonates, a drill is often
not required to open the soft calvarium. The
abdominal exposure is done carefully in layers to
avoid visceral injury. Leaflets of the posterior
rectus sheath and peritoneum are reflected and
held in haemostats to identify and preserve the
40 Hydrocephalus
opening. The catheter is then tunnelled in the
subcutaneous layer with care paid to avoiding
vascular structures in the neck and inadvertent
entry into the thoracic compartment. The distal
catheter is placed into the peritoneal compart-
ment under direct vision ensuring free passage
and the wounds are closed in layers with absorb-
able sutures.
Post-operative radiological evaluation of a
shunt’s placement may be performed with CT or
MRI. There is unlikely to be significant improve-
ment in the degree of ventriculomegaly at this
stage. In circumstances of VP shunt revision,
post-operative cross-sectional imaging can reas-
sure there is no haemorrhage associated with the
removal of the existing shunt—a common cause
of early failure of revised shunts (Fig. 40.8).
Fig. 40.8  Post-operative axial CT scan in a patient with
symptoms of early shunt failure following a ventricular
catheter revision surgery. CT demonstrates intraventricu-
lar haemorrhage along the shunt trajectory most com-
monly due to bleeding from avulsed choroid plexus that
was adherent to the revised catheter
511
40.9.3 Complications of CSF Shunts
Shunt surgery is the mainstay of hydrocephalus
management. It has been hugely successful since
its adoption some seven decades ago. It has, how-
ever, been plagued by a number of complications
some of which persist despite numerous itera-
tions of technique and improvements in implanted
technology. Between 11% and 25% of all patients
will require shunt revision surgery within the first
year of implantation and 85% during their life-
time, the most common and important reasons
being mechanical failure (obstruction or overd-
rainage) and infection (Khan et  al. 2013; Wu
et al. 2007; Reddy et al. 2014).
Mechanical Dysfunction and Blockage
Any shunt may become obstructed at any point
along its length, and this is the most common
type of shunt malfunction accounting for approx-
imately three quarters of all failures (Kestle et al.
2000). The most common site is the proximal
catheter, where it is often noted at the time of
revision that fronds of choroid plexus have
migrated into the perforations at the tip of the
catheter (Paff et al. 2018). There are various theo-
ries as to why this may happen but none have
been definitively proven (Harris and McAllister
2nd. 2012). Accurate placement with the distal
tip floating free within the body of CSF and not in
contact with ventricle walls is believed to reduce
the risk of proximal obstruction (Hayhurst et al.
2010). For years, it was believed that frontally
placed catheters were less prone to this complica-
tion; however, this has not been born out in the
literature (Dickerman et al. 2005).
The valve is the next most common site of
shunt failure. Valve failure can result in either
over- or more typically underdrainage. Implanta-
tion of a valve with too low a fixed pressure will
result in low-pressure symptoms and require revi-
sion. This still qualifies as a shunt failure even if
the system is working as designed. One study has
demonstrated superiority in terms of survival for
512
programmable shunts over non-­ programmable
shunts when corrected for age and hydrocephalus
aetiology; this has not been replicated in other
studies, however (McGirt et al. 2007).
A number of theories regarding valve occlusion
and the reasons underlying it exist. In reality, a
number of factors will contribute. In the acute
phase, active bleeding within the ventricular sys-
tem puts the valve at high risk of failure. The blood
can coagulate under low flow conditions within
the fine calibre valve mechanism. This is not
uncommon following proximal revisions as friable
adherent choroid plexus has a propensity to avulse
and bleed when existing ventricular catheters are
removed. CSF infections predispose shunts to fail,
and commonly, this is at the level of the valve. The
pathological process is not fully understood, but it
is likely that the innate immune response and
resulting inflammatory reaction to infection cause
the build-up of cellular debris within the shunt.
Distal occlusions (beyond the valve) are
uncommon but do occur, and distal patency must
therefore always be confirmed at the time of
shunt exploration surgery. Common causes for
distal failure are coiling within the layers of the
abdominal wall, due to either tube expulsion by
bowel peristalsis or, more likely, poor implanta-
tion technique. Entrapment within an abdominal
pseudocyst or coiling within an intra-abdominal
compartment due to adhesion formation is also
commonly noted at distal revision. Finally, raised
intra-abdominal pressure, for example, due to
chronic constipation or mass lesion, has also
been reported as a cause of shunt failure.
Overdrainage
Overdrainage of the ventricular system can occur
when an inappropriately low-pressure valve is
inserted into a shunt system, such that the cere-
bral mantle is allowed to collapse away from the
overlying dura. This can result in tearing of deli-
cate traversing draining veins and the formation
of subdural haematomas. Similarly, overdrainage
can be a positional phenomenon due to the
siphoning of fluid into the peritoneal cavity when
upright. This can be limited by utilising shunt
designs with incorporated anti-siphon systems or
gravitational valves which mitigate overdrainage
G. Sunderland et al.
in the upright position. Occasionally, mechanical
failure of shunt mechanisms can occur wherein
the resistance to drainage fails and patients expe-
rience low-pressure symptoms.
Chronic overdrainage of the ventricular sys-
tem, especially in the context of previous infec-
tion, can result in small, non-compliant ventricles
that do not expand when the shunt is blocked
even when the pressure is very high. This is a par-
ticularly difficult problem termed the ‘slit ventri-
cle syndrome’ and is one reason that ventricular
volume is not a sure determinant of shunt patency.
Chronic low pressure may be manifest in a thick-
ened skull vault as well as thickened, enhancing
meninges visible on post-contrast MRI. Subdural
fluid collection (hygromas) may also be visible.
Mechanical failure in children is notoriously
difficult to accurately diagnose except in cases of
acutely raised ICP where children may present
vomiting, lethargic or even in coma. More subtle
presentations with headache, behavioural change,
abnormal posturing and change in appetite/feed-
ing may be all that is evident. Remember that a
large proportion of shunted patients have com-
plex needs and often global developmental delay
and so are unable to volunteer much information.
Parents of shunted children are coached to have a
low threshold for presentation to hospital if they
are concerned. Parents are also very sensitive to
changes in behaviour and are often very experi-
enced in identifying subtle changes that may be
due to shunt malfunction. The emergency clini-
cian or junior neurosurgeon dismisses parental
concerns at their peril!
Disconnection
Shunt components have a frustrating habit of dis-
connecting, even in cases where they were prop-
erly and diligently secured (with a non-absorbable
braded tie). Shunt material can become tethered
along its length by adjacent tissue reaction. Older
materials had a susceptibility to become calcified
and stiff also. Rapid growth with tethering of dis-
tal/proximal components or rapid/extreme neck
movements, such as those witnessed in gener-
alised tonic-clonic seizures, can test shunt con-
nections or fracture old and brittle tubing. In
these cases, the presentation is often more sub-
40 Hydrocephalus
acute with subtle symptoms or signs of over- or
underdrainage. Patients may even complain of
pain, redness or swelling at the site of disconnec-
tion or fracture.
Infection
Infection remains an important complication,
though rates have significantly improved in the
last four decades, from between 14% and 24%
reported in the mid-1970s to between 2% and
10% quoted in contemporary literature.
Inoculation of the shunt almost universally occurs
at the time of surgery, and infections present
within 30 days of the index operation in the vast
majority of cases (Kulkarni et al. 2001). It is usu-
ally due to skin commensal organisms, most often
Staphylococcus species (Bayston 2018). Risk fac-
tors for shunt infection include young age, in par-
ticular premature neonates, post-operative CSF
leak, improper technique and excess handling of
the shunt prior to insertion and previous shunt
infection (Simon et al. 2009; McGirt et al. 2003).
Patients with infection may present with obvi-
ous symptoms and signs of CNS infection:
pyrexia, nausea and vomiting, headache,
­meningism and confusion. In addition, anorexia
and abdominal pain may reflect a peritoneal reac-
tion to infected CSF. There may also be tender-
ness and erythema tracking along the shunt tract.
More often, however, patients present with a less
well-defined illness, and often the only symp-
toms of infection are due to partial or complete
shunt obstruction and resulting raised
ICP. Diagnosis can therefore be challenging, and
a ‘shunt tap’ is often advocated (Fig. 40.9). CSF
aspirated aseptically via the shunt reservoir is
sent for microscopy, gram stain, cell count and
culture. Unfortunately, due to low numbers of
viable bacteria present in the CSF of infected sys-
tems, microbiological culture can be negative in
up to 25% of ‘confirmed’ infected cases. The
CSF cell count is most instructive in this sce-
nario, and a pleocytosis with predominant neu-
trophilia is highly suggestive of shunt infection
particular with concordant clinical details (recent
surgery, unwell patient).
Staphylococci that colonise a shunt survive in
a biofilm and in deeper layers downregulate their
513
Fig. 40.9  Photograph demonstrating the technique of
‘shunt tap’ via a butterfly needle inserted into the subcuta-
neous shunt reservoir. Strict adherence to aseptic precau-
tions is essential to prevent inadvertent bacterial
inoculation of the shunt system
metabolic activities and cell division markedly
(Bayston 2018). The result is a relative resistance
to antibiotics, and surgical explantation is neces-
sary to manage the infection. The shunt is
replaced with an EVD to maintain CSF diversion,
and systemic intravenous antibiotics, with or
without intrathecal antibiotics, are administered
for 10–14 days before reimplantation (Fig. 40.10
(EVD)).
Improvements in operative technique, stan-
dardisation of practices, antimicrobial steward-
ship and most recently the adoption of antibiotic
impregnated catheters have all contributed to
improvements in infection rates. The consider-
able morbidity associated with infection man-
dates continued vigilance however, and timely
and effective treatment of shunt infections is
essential.
Ventriculoatrial Shunts
VA shunts are indicated in patients with concom-
itant intra-abdominal pathologies precluding the
use of the peritoneum as a drainage site. Common
coexistent pathologies include necrotising
enterocolitis, peritonitis and extensive abdominal
surgery. Complications specific to VA shunts
include the need for repeated lengthening of the
short distal catheter, higher risk of bacteraemia
514
Fig. 40.10  External ventricular drainage
and sepsis as well as the risks of specific vascular
complications, such as thrombosis, microemboli
with resultant pulmonary hypertension, macro-
emboli with pulmonary embolism and vascular
perforation.
40.9.4 Ventriculoperitoneal Shunts
and Abdominal Surgery
Paediatric general surgeons will at times have to
assess patients with VP shunts and potential
abdominal pathology in both the acute and elec-
tive settings.
If, during an elective procedure without infec-
tion or contamination of the peritoneum, a perito-
neal catheter is encountered, then it should
merely be pushed aside gently and excluded from
the field. Significant contamination of the field
mandates externalisation of the shunt. The distal
shunt tubing is assumed to be contaminated and
G. Sunderland et al.
is delivered externally through the skin to con-
tinue draining into a collection bag. Once the
abdominal infection is treated, the external length
of tubing is cut, and a new clean length of distal
shunt tubing can be safely reimplanted.
Laparoscopic techniques are now common-
place, and experience has shown these to be
entirely safe in the presence of a VP shunt
(Rosenfeld et al. 2019; Walker and Langer 2000;
Jackman et  al. 2000). Early animal studies
reported significant rises in ICP on induction of
pneumoperitoneum as one would see in an ICP
trace anytime a patient coughed or strained
(Rosenthal et  al. 1997). There was concern that
prolonged raised intra-abdominal pressure could
significantly impair valve or shunt function or
potentially even lead to retrograde flow and sig-
nificant pneumocephalus. Studies of shunted
patients during laparoscopic surgery have not
identified such risk or adverse events, and there
are only scattered case reports of shunt failure
40 Hydrocephalus
associated with laparoscopy amongst the many
thousands of cases performed in the last 20+
years. In vitro studies of shunt valves show them
to be resistant to retrograde flow at pressures far in
excess of that used for pneumoperitoneum (Neale
and Falk 1999). While there is no good evidence
to support safety concerns in this patient cohort,
caution is still advised over high insufflation pres-
sures (>16 mmHg) and prolonged surgery (>3 h)
(Sankpal et al. 2011). Routine discussion with a
neurosurgeon regarding the safety of laparoscopy
in shunted patients is probably not necessary
unless significant abdominal contamination is
anticipated, in which circumstance externalising
the shunt could be considered in advance.
Assessment of the acute abdomen with a shunt
in situ is a difficult task and requires a holistic
approach. Children may present with abdominal
symptoms secondary to shunt malfunction, and it
is important to define early if there are clinical or
radiological signs of shunt dysfunction and not
concentrate solely on the abdomen; similarly, it is
vital that neurosurgeons seek an experienced
opinion early in cases where the shunt seems to
be functioning but the patient is symptomatic.
If there are abdominal symptoms and signs of
peritonitis mandating laparotomy, then the shunt
should be externalised, the CSF cultured, and
antibiotics started. If the shunt is obviously
infected in addition, i.e. signs of raised ICP,
meningism and erythema tracking along the
shunt, then the shunt is removed in its entirety
and only reimplanted once effective antibiotic
treatment of both the abdominal and CNS infec-
tion is completed.
Appendicitis is a common acute abdomen pre-
sentation, and children with VP shunts are not
excepted from this. There is no evidence to sug-
gest these patients have a worse outcome than
non-shunted patients, and with caution, there
should be no shunt-related complications (Ein
et  al. 2006; Barina et  al. 2007). In an emergent
appendicectomy, if the appendix is inflamed but
not perforated, it is reasonable to leave the shunt
in situ; if, however, there is any peritoneal soil-
ing, then the shunt must be externalised.
Other common, but usually non-emergent,
complications of peritoneal catheters include
515
peritoneal pseudocysts and ascites. Pseudocysts
are wall-less fluid collections accumulating
between matted bowel loops; they may be com-
plicated by infection. If there are no features of
infection, then simple repositioning of the cathe-
ter in another portion of the peritoneum is all that
is required.
40.9.5 Endoscopic Third
Ventriculostomy (ETV)
Endoscopic exploration of the ventricular system
with therapeutic intent had been explored in the
early twentieth century (Demerdash et al. 2017).
Early forays into neuroendoscopy were aimed at
the management of hydrocephalus but had lim-
ited success. There were a handful of stumbling
attempts to revive the technique in the subsequent
decades, but these never translated to routine
practice. These early failures were due largely to
technological limitations, and so unsurprisingly,
it wasn’t until the 1970s with the inception of the
Hopkin’s rod endoscope that the technique really
developed. Endoscopic applications have since
exploded throughout the whole breadth of surgi-
cal practice, and neurosurgery is no exception.
The re-introduction of neuroendoscopy at this
time was followed by a rapid broad adoption, and
since then, a multitude of applications have been
developed focused on, but not limited to, the
management of hydrocephalus. The most ubiqui-
tous of these is the endoscopic third ventriculos-
tomy (ETV).
ETV is primarily indicated in cases of obstruc-
tive hydrocephalus, the aim being to provide a
route whereby CSF can bypass the point of
obstruction and enter the subarachnoid space,
creating a functional shunt from the third ventri-
cle into the basal cisterns, from where it may cir-
culate and be reabsorbed. The major advantages
of ETV over an implanted shunt are that it obvi-
ates the need for an implanted foreign body, thus
avoiding risks of infection, material degradation,
disconnection, misplacement, etc., and it main-
tains a more ‘physiological’ CSF dynamic, some-
thing that even modern shunt valves struggle to
replicate.
516 G. Sunderland et al.

40.9.6 ETV: The Technique
The technique of ETV is demonstrated in
Fig.  40.11. A rigid endoscope is navigated into
the frontal horn of the lateral ventricle through a
frontal pre-coronal burr hole placed in the midpu-
pillary line. After ventricular cannulation, the
endoscope is introduced, and the operator orien-
tates themselves noting the thalamic veins run-
ning toward the foramen of Monro. The foramen
of Monro transmits CSF from the lateral to third
ventricle and is bounded posteriorly by the ante-
rior pole of the ipsilateral thalamus and anteriorly
by the fornix—an important part of the limbic
system that links the hippocampus to mammil-
lary bodies. The fornices are integral to working

a b
b

c
a

c a a

c d

Fig. 40.11  Endoscopic images representing key stages
and anatomy encountered at ETV. (a) View from the endo-
scope within the right lateral ventricle. Note the choroid
plexus and posterior caudate vein (a), anterior caudate
vein (b) and thalamostriate vein (c) converging towards
the foramen of Monro (d). (b) Endoscopic image from
within the third ventricle. Visible are the mammillary bod-
ies (a) and pituitary infundibulum (b) bordering the thin
membranous tuber cinereum (c). Just visible is the basilar
artery (d) through the tuber cinereum. (c) A Fogarty cath-
eter is used to perforate the floor of the third ventricle and
the balloon (a) inflated to expand the ‘stoma’. (d) View
with Fogarty catheter withdrawn demonstrating a perfora-
tion (a) through to underlying pre-pontine cistern. Note
the fronds of arachnoid visible
40 Hydrocephalus 517

memory function and thus must not be injured or structure, and interfering with it risks causing
put under excessive stretch. The endoscope is IVH.  Furthermore, the choroid plexus is not
then navigated through the foramen of Monro merely a CSF ‘pump’ but generates intracranial
into the third ventricle. On the third ventricle pressure (ICP), maintains CSF homeostasis and
floor, the landmarks are reliably seen, the mam- provides micronutrients, proteins and hormones
millary bodies posteriorly and the pituitary infun- for neuronal and glial development, maintenance
dibulum (a red blush at the anterior apex of the and function. Concerns regarding long-term neu-
third ventricle). Within a triangle drawn between rocognitive outcome and the potential neurode-
these points is the tuber cinereum where the velopmental impairment exist (Spector et  al.
membranous floor of the third ventricle is thin 2015).
and through which it is safe to pass. The tuber
cinereum is punctured, and this ‘stoma’ then
expanded sequentially with balloon or endo- 40.9.8 Indications for ETV
scopic forceps. Further membranes, in particular
the membrane of Liliequist, should be identified ETV is considered in cases of obstructive hydro-
and fenestrated, and a clear view to the pre-­ cephalus including obstructions in the caudal
pontine cistern and the basilar artery is seen. The portion of the third ventricle (pineal region),
floor of the third ventricle is seen to billow as if,
aqueduct, fourth ventricle and foramen magnum.
in the wind when the endoscope is withdrawn, a It has an established role in the primary manage-
reassuring sign that the intervention has been ment of aqueduct stenosis (Kulkarni et al. 2018).
successful (Fig. 40.11). Post-operatively, ventric- ETV enjoys an overall success rate of 60–90%,
ulostomy patency can be investigated using phase the significant variety due to age and aetiological
contrast MRI or ‘CSF flow study’. On a midline factors. Rates improve with advancing age (chil-
sagittal cut, the flow of CSF across the stoma and dren > 1 year) and with specific aetiologies, i.e.
turbulence within the pre-pontine cistern can be those with an anatomical obstruction such as
identified via a dark ‘flow void’ in this path aqueduct stenosis, posterior fossa tumours and
(Fig. 40.12) (Dinçer et al. 2011). Dandy-Walker malformation. Aetiologies with
inferior rates of success are those classically
termed ‘communicating’, i.e. post-haemorrhagic
40.9.7 ETV with Choroid Plexus and post-infective.
Coagulation The commonest causes seen in neonates are
germinal matrix IVH and meningitis; thus, ETV
In a technique akin to that pioneered in the 1920s, is not a viable option in this cohort. The
choroid plexus coagulation sees the surgeon Hydrocephalus Research Network in the USA
endoscopically cauterising the choroid plexus developed and in 2010 published the ETV
with the aim of reducing CSF production. This Success Score (ETVSS), a reliable and practical
may be done in conjunction with an ETV (ETV-­ predictive score based on age, aetiology and prior
CPC), a sort of ‘belt and braces’ approach. Trials CSF shunt (Table 40.3) (Kulkarni et al. 2010).
run in Uganda and subsequently in the USA in
the 2000s showed advantage over ETV alone
especially in younger children (<1 year) with cer- 40.9.9 Complications of ETV
tain hydrocephalic aetiologies (Stone and Warf
2014; Warf 2005). There remains controversy ETV is a low-risk procedure, with major risks
over the efficacy of this two-stage technique as exceedingly rare (Bouras and Sgouros 2011).
opponents argue that CSF production and drive The headline potential immediate complication
through the stoma help to maintain patency and of ETV is the risk of damage to the basilar artery
thus reduce failure in the long term. In addition, which lies in wait immediately below the site of
the choroid plexus is a very vascular and friable stoma formation. The extremely rare but likely-­
518 G. Sunderland et al.

a b

c d

Fig. 40.12  CSF ‘flow’ studies employ various MRI appearances of a third ventriculostomy. The yellow arrows
sequences performed to produce a time-resolved 2D highlight the CSF flow void produced by a ‘jet’ of CSF
phase contrast with velocity encoding study. Images a–d moving from the third ventricle to the pre-pontine space
show midline sagittal sections with typical post-operative

Table 40.3 The Score  = Age  + Aetiology  + Previous shunt

Kulkarni ETV 0 <1 month Post-infectious Previous shunt
Success Score 10 1 month to <6 months No previous shunt
(ETVSS)
20 Myelomeningocoele
Intraventricular
haemorrhage
Non-tectal brain
tumour
30 6 months to <1 year Aqueductal stenosis
Tectal tumour
Other
40 1 year to <10 years
50 >10 years
40 Hydrocephalus
immediately fatal complication of basilar artery
injury is one which every neurosurgeon hopes to
avoid. ETV, therefore, while a relatively simple
procedure, should be performed by or under the
supervision of an experienced operator.
Other potential complications include but are
not limited to intraventricular haemorrhage with
blood obscuring the operator’s view, injury to the
thalamus or fornix usually as a consequence of
poorly placed burr hole (due to inadequate or
inaccurate planning) and misaligned trajectory,
CSF leak and rarely infection.
In the long term, ventriculostomies may close
over, although why this should occur is not well
understood. Patients thus require follow-up just
as a shunted patient would but probably not for as
long. A patient who is asymptomatic 1 year on
from ETV probably does not require routine fol-
low-­up, unless they have ongoing neurosurgical
concerns. A high index of suspicion should be
maintained, however, for late failure of ventricu-
lostomy particularly in patients presenting with
symptoms of raised ICP.  For this reason, some
surgeons opt to leave a ventricular access device
(VAD) or ‘Ommaya reservoir’ in situ following
ETV to permit ventricular access and evaluation
of CSF pressure, as well as immediate decom-
pression by therapeutic aspiration of CSF in the
event of raised pressure.
40.9.10 Ventricular Access Devices
and Ventriculosubgaleal
Shunt
In the circumstances of a premature neonate with
expanding head circumference, ventricular index
and clinical evidence of raised intracranial pres-
sure and bradycardic and apnoeic episodes with-
out other identified cause (sepsis, reflux, etc.),
consideration of CSF diversion is needed. In the
first instance, a ventricular or subdural aspiration
of CSF can be performed for both diagnostic and
therapeutic purposes via the wide-open anterior
fontanelle. A ventricular tap (up to 15 ml/kg or no
more than 30 ml total) will inform the operator of
opening pressure (often not high surprisingly)
and CSF appearance/consistency (clear, straw
519
coloured, blood stained, engine oil). Subsequent
microbiological examination and culture is
essential to diagnose or rule out CSF infection or
ventriculitis. Furthermore, the extent and dura-
tion of the clinical response is most informative.
Repeated ventricular taps (twice weekly) were
often recommended for neonates below the 2 kg
threshold for shunt surgery, or in those not clini-
cally stable, for transfer to a neurosurgical unit.
Nowadays, repeated CSF access via this route is
avoided, and placement of a ventricular access
device or VAD is more commonly performed
(Fig.  40.13). In those infants too unstable for
transfer or surgery, serial ventricular tap is still a
viable option however. The VAD provides a read-
ily accessible silastic ‘bubble’ under the skin,
which allows safer access to CSF, without the
need for repeated traversing of the cerebral cor-
tex. This can readily be converted to an external
ventricular drain, by attaching a butterfly needle
to a ‘Becker’ CSF collection and monitoring sys-
tem. Furthermore, the reservoir can be connected
to an internalised distal catheter at a later stage,
thus converting it to a VP or VA shunt.
Unfortunately, even with diligent aseptic tech-
nique, there is a relatively high rate of infection
Fig. 40.13  Diagrammatic representation of a ventricular
access device (VAD) such as an Ommaya reservoir
520
(10%), the risk increasing with more extreme
prematurity (Spader et al. 2015).
An attractive alternative to VAD then is the
implantation of a ventriculosubgaleal shunt
(VSGS) in which a ventricular catheter is con-
nected to a valve and distal tubing implanted in a
subgaleal pocket created at the time of insertion
by blunt undermining of the galea aponeurotica.
CSF drainage can then occur in a regulated fash-
ion due to the inline valve. CSF accumulates in
a b
Fig. 40.14 Ventriculosubgaleal shunts (VSGS). (a)
Diagrammatic representation of positioning within the
frontal horn of the lateral ventricle with the distal tubing
G. Sunderland et al.
the pocket which then expands. The CSF is
slowly resorbed via the lining of the pocket. The
pocket can expand considerably, and this is a fre-
quent concern for parents and inexperienced cli-
nicians; however, as long as it is not under
tension, no long-term sequelae are seen, and it is
generally well tolerated (Eid et  al. 2018)
(Fig. 40.14). At such time as the infant is fit for
distal implantation, the distal catheter can be sim-
ply tunnelled to an alternative and permanent
coiled in the subgaleal pocket. (b) Photograph and (c)
coronal T2-weighted MRI demonstrated a subgaleal CSF
collection and swelling under tension
40 Hydrocephalus 521

location (cardiac atrium or peritoneal cavity). Table 40.4  Summary of the Papile and Volpe systems
There is no demonstrable difference in the rates for grading intraventricular haemorrhage of prematurity
based on cranial ultrasound findings
of conversion to VP shunt between VAD and
VSGS; however, the average age and body weight Grade Papile Volpe
I Unilateral or Germinal matrix
at the time of conversion are greater in VSGS
bilateral germinal haemorrhage with no IVH
patients, and the rate of infection is lower (Wang matrix or IVH occupying <10% of
et al. 2014). haemorrhage ventricular area on
parasagittal view
II IVH without IVH occupying 10–50% of
ventricular the ventricular area on
40.9.11 Common Clinical dilatation parasagittal view
Presentation III IVH with IVH occupying 50% of
ventricular ventricular area on
Three common presentations will now be dis- dilatation parasagittal view +/–
periventricular
cussed to give the non-specialist some insight echodensities
into their management. IV IVH extending Periventricular venous
into adjacent haemorrhagic infarction
Managing Post-haemorrhagic Hydrocephalus brain parenchyma (PVHI), cystic
periventricular
of Prematurity (PHHP)
leukomalacia
Premature neonates are at risk of intraventricular
haemorrhage due to friable, immature and poorly
supported vessels within the germinal matrix. Table 40.5  Summary of outcome and long-term progno-
sis for patients with IVH of prematurity. Progressive ven-
Both neural and glial elements originate from tricular dilatation equates with the need for CSF shunting
this region found in the subependymal region
Progressive Long-term
(adjacent to the lateral ventricle) medial to the Volpe Mortality ventricular neurological
caudate nucleus. The germinal matrix involutes grade (%) dilatation (%) sequelae (%)
progressively during the third trimester and is I 5 5 5
limited to the caudothalamic groove by 32 weeks. II 10 20 15
It has essentially disappeared by 35 weeks; III 20 55 35
IV 50 80 90
hence, the risk is inversely proportional to gesta-
tional age. Changes in cerebral blood flow and
hypoxia predispose these vessels to loss of integ- sion from grades I to IV more likely represents a
rity resulting in intracerebral haemorrhage continuum rather than distinct haemorrhagic
(ICH). ICH frequently ruptures into the ventri- entities. In addition, the periventricular echoden-
cles causing intraventricular haemorrhage sities often seen and formerly classified as ICH
(IVH). Blood within the CSF as discussed and thus grade Papile grade IV are, in fact, more
impairs CSF resorption and induces ventriculo- likely post-haemorrhagic venous infarctions
megaly and ­hydrocephalus. (Parodi et al. 2020). Subsequent scales have been
Unfused sutures and wide-open anterior and developed to account for these shortcomings
posterior fontanelles allow these children to which include the Volpe scale from 1986 which
accommodate for a large increase in intracerebral has largely superseded Papile (Table  40.4)
CSF volume. The open fontanelle also permits (Guzzetta et al. 1986).
transcranial ultrasound (see above) for diagnosis Higher IVH grading correlates with worse
and surveillance. A number of grading systems neurological outcome, higher mortality and pro-
have been devised to communicate the extent of gressive ventricular dilatation and as a result pro-
IVH and correlate with outcome. The first broadly gression to CSF shunt requirement (Table 40.5).
adopted was the Papile grade, first proposed in The practicalities of surgical management of
1978 (Table  40.4) (Papile et  al. 1978). There is PHHP is covered earlier in this chapter. The com-
some controversy over its reliability as progres- plex and multifaceted non-­surgical management of
522
these difficult patients is obviously beyond the
scope of this chapter. That said, serial head circum-
ference measurement and cranial US with accurate
VI trend recording constitute the cornerstone of
monitoring and are the minimum dataset a consult-
ing neurosurgeon will require. Typically, there will
be numerous interactions between neonatologists
and neurosurgeons over a period of weeks prior to
intervention so accurate record keeping and com-
munication within the teams is vital.
Approach to Newly Diagnosed
Hydrocephalus in Childhood (Non-neonates)
The aetiologies underlying childhood hydrocepha-
lus are closely age-associated. Management strate-
gies and approaches will therefore be nuanced.
Congenital causes of hydrocephalus, particularly
aqueduct stenosis and mass lesions (tumours)
obstructing CSF pathways, are the prime causes of
symptomatic hydrocephalus in this age group.
In circumstances of acute presentation in a
child ‘in extremis’, management should first fol-
low an APLS algorithm with appropriate airway
management if required being paramount.
Subsequent management is to make a timely
diagnosis, and in the context of a supportive his-
tory or neurologically impaired child, urgent
cross-sectional brain imaging is mandated. CT
scans are most appropriate in this circumstance
as they are almost universally available and can
be performed in a matter of minutes. If feasible
and safe, any suggestion of a mass lesion requires
further investigation with MRI. There is little role
for post-contrast CT, unless MRI is contraindi-
cated. In a life-threatening circumstance with a
neurologically compromised child, definitive
radiological assessment can be delayed until after
management of the hydrocephalus is instituted.
It is worth stating that such an urgent presenta-
tion seldom occurs. Hydrocephalus usually has a
more subacute presentation with a prodrome of
raised ICP features allowing earlier identification
and diagnosis. A common route of diagnosis in
fact involves the optician who notes papilloe-
dema when examining a child referred for inves-
tigation of headaches by their GP or at routine
eye test. Vomiting and subtle neurological fea-
tures may also feature in the history as can mul-
G. Sunderland et al.
tiple attendances to their GP.  There may be
features indicative of the underlying cause, i.e.
ataxia from cerebellar tumours, cranial nerve pal-
sies in brainstem or CP angle tumours, eye move-
ment disorders from midbrain/pineal region
tumours and visual failure from tumours near the
optic apparatus. Hydrocephalus may also present
secondary to infections such as meningitis or
bacterial abscess although with these latter
pathologies, there is usually a much more marked
systemic illness. An MRI brain rather than CT is
advised in these circumstances as a primary
investigation as it yields much more information
and permits prompt and appropriate decision-­
making without ionising radiation exposure to
vulnerable tissues. This should ideally be per-
formed at the referring unit, prior to referral to
neurosurgery in a stable child with suspected
hydrocephalus.
Approach to a Child with a Potentially
Blocked Shunt
Children presenting with acute symptoms of
shunt block or presenting in extremis with no his-
tory available can be difficult to diagnose.
Children still die from unrecognised shunt fail-
ure, though this is thankfully very rare, and a
high index of suspicion is vital to save lives.
The symptoms and signs of shunt failure are
primarily those of raised ICP and have been cov-
ered at length but bear repetition. Any child with
a VP shunt with a recent history of headaches,
nausea and vomiting and/or progressive cogni-
tive decline or neurological compromise should
receive a CT head without delay. Children with
prior episodes of shunt block are at higher risk of
future blockages, and presentation with similar
features, however apparently esoteric (behav-
ioural change, not eating, odd-posturing), should
raise suspicion of a further shunt block. Suffice it
to say if you or the parents think ‘could the shunt
be blocked?’ it needs investigating. These chil-
dren should not be sent home from A&E without
discussion with a neurosurgeon. The ‘query-­
blocked-­shunt’ call is the most common contact a
paediatric neurosurgeon on-call will receive and,
while a cause of occasional frustration, is a call
that should always be made.
40 Hydrocephalus 523

40.9.12 Follow-Up of the Patient quality-­

of-life metrics also show deficit com-
with Treated pared with a healthy background population.
Hydrocephalus

Patients with treated hydrocephalus should remain 40.10 Conclusion

under the care of a neurosurgeon. Previously, fol-
low-up was variable, some clinicians following all Hydrocephalus occurs as a consequence of a
patients and managing clinics full of well-shunt multitude of pathological processes and is inci-
patients, while others discharged after a short fol- dent at all stages of childhood. Once identified, it
low-up period, relying on the fact that if the patient is eminently treatable, but this treatment can be
has a problem with their shunt, they would present complex and is not infrequently complicated.
to the hospital. Neither of these scenarios is ideal, Treatment is surgical, but decisions regarding the
and the growth in the clinical nurse specialist and timing and modality of intervention are nuanced.
nurse practitioner field has led to the ‘delegation’ Follow-up is lifelong, for the most part, and pre-
of immediate responsibility for follow-up away sentation with a suspicion of shunt block will
from consultant-­led clinics. Open- or rapid-access occur in virtually all patients at some point dur-
face-to-face or telephone clinics with experienced ing this period. Up to 85% will require shunt
nurse specialists are a very effective method for revision surgery during their lifetime, and there
triaging concerned shunt patients and keeping in should be a high index of suspicion for shunt fail-
touch with less engaged shunt patients. Baseline ure when reviewing these patients acutely, espe-
imaging several weeks after insertion of a VP cially in the context of a supportive history.
shunt once recovered is useful to provide a ‘well
scan’ against which future imaging can be con-
trasted. For young children, this should be updated Further Reading
every 1–2 years to keep pace with their changing
head shape and size. Cinalli G, Maixner WJ, Sainte-Rose C (eds) (2012)
Pediatric hydrocephalus. Springer, New York
Ellenbogen JR, Waqar M, Pettorini B (2016) Management
of post-haemorrhagic hydrocephalus in prema-
40.9.13 Outcome of Treated ture infants. J Clin Neurosci 31:30–34. https://doi.
Paediatric Hydrocephalus org/10.1016/j.jocn.2016.02.026
International Society of Paediatric Neurosurgery Guide:
https://www.ispn.guide/. A freely available online edu-
A number of longitudinal studies have reported cational resource providing up to date and evidence-­
long-term outcomes in children treated for hydro- based information on all paediatric neurosurgical
cephalus (Hoppe-Hirsch et  al. 1998; Paulsen conditions. Compiled by internationally renowned
clinicians
et al. 2015; Gmeiner et al. 2017). Overall mortal- Liebenberg WA, Johnson RB (2018) Neurosurgery for
ity is reported between 5 and 39%, the large vari- surgical trainees. Hippocrates Books, Carnforth
ability due to differences in how mortality was Mallucci C, Sgouros S (eds) (2016) Cerebrospinal fluid
attributed and duration of follow-up. Shunt-­ disorders. CRC, Boca Raton, FL
related mortality is estimated at 4.6–17%, most
due to infection (approximately 2/3) followed by
mechanical failure (1/3). The majority (60–67%) References
of patients graduated from normal school; how-
Aschoff A, Kremer P, Hashemi B, Kunze S (1999) The
ever, significant impairment in intelligence quo- scientific history of hydrocephalus and its treatment.
tient has been noted compared with age-matched Neurosurg Rev 22(2–3):67–93. Discussion 4–5
counterparts. Cognitive impairment has been Barina AR, Virgo KS, Mushi E, Bahadursingh AM,
related to the number of shunt failures a patient Johnson FE (2007) Appendectomy for appendicitis in
patients with a prior ventriculoperitoneal shunt. J Surg
experiences in their lifetime. Self-reported Res 141(1):40–44
524
Bayston R (2018) Cerebrospinal fluid shunt infection:
microbiological basis, pp 1–19
Bouras T, Sgouros S (2011) Complications of endoscopic
third ventriculostomy. J Neurosurg Pediatr 7(6):643
Dandy WE, Blackfan KD (1914) An experimental, clini-
cal and pathological study: Part 1.—Experimental
studies. Am J Dis Child VIII(6):406–482
Demerdash A, Rocque BG, Johnston J, Rozzelle CJ,
Yalcin B, Oskouian RJ et al (2017) Endoscopic third
ventriculostomy: a historical review. Br J Neurosurg
31(1):28–32
Dewan MC, Rattani A, Mekary R, Glancz LJ, Yunusa
I, Baticulon RE et  al (2018a) Global hydrocephalus
epidemiology and incidence: systematic review and
meta-analysis. J Neurosurg 1–15
Dewan MC, Rattani A, Fieggen G, Arraez MA, Servadei
F, Boop FA et al (2018b) Global neurosurgery: the cur-
rent capacity and deficit in the provision of essential
neurosurgical care. Executive Summary of the Global
Neurosurgery Initiative at the Program in Global
Surgery and Social Change. J Neurosurg 1–10
Dickerman RD, McConathy WJ, Morgan J, Stevens QE,
Jolley JT, Schneider S et al (2005) Failure rate of fron-
tal versus parietal approaches for proximal catheter
placement in ventriculoperitoneal shunts: revisited. J
Clin Neurosci 12(7):781–783
Dinçer A, Özek MM (2011) Radiologic evalua-
tion of pediatric hydrocephalus. Childs Nerv Syst
27(10):1543–1562
Dinçer A, Yildiz E, Kohan S, Memet ÖM (2011) Analysis
of endoscopic third ventriculostomy patency by MRI:
value of different pulse sequences, the sequence
parameters, and the imaging planes for investigation
of flow void. Childs Nerv Syst 27(1):127–135
Dorner RA, Burton VJ, Allen MC, Robinson S, Soares BP
(2018) Preterm neuroimaging and neurodevelopmen-
tal outcome: a focus on intraventricular hemorrhage,
post-hemorrhagic hydrocephalus, and associated brain
injury. J Perinatol 38(11):1431–1443
Drake JM, Kestle JR, Milner R, Cinalli G, Boop F, Piatt
J Jr et  al (1998) Randomized trial of cerebrospinal
fluid shunt valve design in pediatric hydrocephalus.
Neurosurgery 43(2):294–303. Discussion 303–305
Eid S, Iwanaga J, Oskouian RJ, Loukas M, Jerry Oakes
W, Shane TR (2018) Ventriculosubgaleal shunting—a
comprehensive review and over two-decade surgical
experience. Childs Nerv Syst 34(9):1639–1642
Ein SH, Miller S, Rutka JT (2006) Appendicitis in the
child with a ventriculo-peritoneal shunt: a 30-year
review. J Pediatr Surg 41(7):1255–1258
Gmeiner M, Wagner H, Zacherl C, Polanski P, Auer
C, van Ouwerkerk WJ et  al (2017) Long-term
mortality rates in pediatric hydrocephalus—a ret-
rospective single-center study. Childs Nerv Syst
33(1):101–109
Guzzetta F, Shackelford GD, Volpe S, Perlman JM, Volpe
JJ (1986) Periventricular intraparenchymal echodensi-
ties in the premature newborn: critical determinant of
neurologic outcome. Pediatrics 78(6):995–1006
Harris CA, McAllister JP 2nd. (2012) What we should
know about the cellular and tissue response causing
G. Sunderland et al.
catheter obstruction in the treatment of hydrocephalus.
Neurosurgery 70(6):1589–1601. Discussion 601–602
Hayhurst C, Beems T, Jenkinson MD, Byrne P, Clark S,
Kandasamy J et al (2010) Effect of electromagnetic-­
navigated shunt placement on failure rates: a prospec-
tive multicenter study. J Neurosurg 113(6):1273–1278
Hoppe-Hirsch E, Laroussinie F, Brunet L, Sainte-Rose C,
Renier D, Cinalli G et al (1998) Late outcome of the
surgical treatment of hydrocephalus. Childs Nerv Syst
14(3):97–99
Jackman SV, Weingart JD, Kinsman SL, Docimo
SG (2000) Laparoscopic surgery in patients with
­ventriculoperitoneal shunts: safety and monitoring. J
Urol 164(4):1352–1354
Kestle J, Drake J, Milner R, Sainte-Rose C, Cinalli G,
Boop F et al (2000) Long-term follow-up data from the
Shunt Design Trial. Pediatr Neurosurg 33(5):230–236
Khan F, Shamim MS, Rehman A, Bari ME (2013) Analysis
of factors affecting ventriculoperitoneal shunt survival
in pediatric patients. Childs Nerv Syst 29(5):791–802
Khasawneh AH, Garling RJ, Harris CA (2018)
Cerebrospinal fluid circulation: What do we know and
how do we know it? Brain Circ 4:14–18
Kimelberg HK (2004) Water homeostasis in the brain:
basic concepts. Neuroscience 129(4):851–860
Kulkarni AV, Drake JM, Lamberti-Pasculli M (2001)
Cerebrospinal fluid shunt infection: a prospective
study of risk factors. J Neurosurg 94(2):195–201
Kulkarni AV, Drake JM, Kestle JR, Mallucci CL, Sgouros
S, Constantini S (2010) Predicting who will benefit
from endoscopic third ventriculostomy compared with
shunt insertion in childhood hydrocephalus using the
ETV Success Score. J Neurosurg Pediatr 6(4):310–315
Kulkarni AV, Sgouros S, Leitner Y, Constantini S (2018)
International Infant Hydrocephalus Study (IIHS):
5-year health outcome results of a prospective, mul-
ticenter comparison of endoscopic third ventricu-
lostomy (ETV) and shunt for infant hydrocephalus.
Childs Nerv Syst 34(12):2391–2397
Lifshutz JI, Johnson WD (2001) History of hydrocephalus
and its treatments. Neurosurg Focus 11(2):E1
Mallucci CL, Jenkinson MD, Conroy EJ, Hartley JC,
Brown M, Dalton J et al (2019) Antibiotic or silver ver-
sus standard ventriculoperitoneal shunts (BASICS): a
multicentre, single-blinded, randomised trial and eco-
nomic evaluation. Lancet 394(10208):1530–1539
McGirt MJ, Zaas A, Fuchs HE, George TM, Kaye K,
Sexton DJ (2003) Risk factors for pediatric ventriculo-
peritoneal shunt infection and predictors of infectious
pathogens. Clin Infect Dis 36(7):858–862
McGirt MJ, Buck DW 2nd, Sciubba D, Woodworth GF,
Carson B, Weingart J et al (2007) Adjustable vs set-­
pressure valves decrease the risk of proximal shunt
obstruction in the treatment of pediatric hydrocepha-
lus. Childs Nerv Syst 23(3):289–295
Neale ML, Falk GL (1999) In vitro assessment of back
pressure on ventriculoperitoneal shunt valves. Is lapa-
roscopy safe? Surg Endosc 13(5):512–515
Paff M, Alexandru-Abrams D, Muhonen M, Loudon W
(2018) Ventriculoperitoneal shunt complications: A
review. Interdiscip Neurosurg 13:66–70
40 Hydrocephalus
Papile LA, Burstein J, Burstein R, Koffler H (1978)
Incidence and evolution of subependymal and intra-
ventricular hemorrhage: a study of infants with birth
weights less than 1,500 gm. J Pediatr 92(4):529–534
Parodi A, Govaert P, Horsch S, Bravo MC, Ramenghi LA
(2020) Cranial ultrasound findings in preterm ger-
minal matrix haemorrhage, sequelae and outcome.
Pediatr Res 87(Suppl 1):13–24
Paulsen AH, Lundar T, Lindegaard KF (2015) Pediatric
hydrocephalus: 40-year outcomes in 128 hydroce-
phalic patients treated with shunts during childhood.
Assessment of surgical outcome, work participation,
and health-related quality of life. J Neurosurg Pediatr
16(6):633–641
Pople IK (2002) Hydrocephalus and shunts: what the neu-
rologist should know. J Neurol Neurosurg Psychiatry
73 Suppl 1(3):i22
Rachel RA (1999) Surgical treatment of hydrocepha-
lus: a historical perspective. Pediatr Neurosurg
30(6):296–304
Reddy GK, Bollam P, Caldito G (2014) Long-term out-
comes of ventriculoperitoneal shunt surgery in patients
with hydrocephalus. World Neurosurg 81(2):404–410
Rekate HL (2011) A consensus on the classification of
hydrocephalus: its utility in the assessment of abnor-
malities of cerebrospinal fluid dynamics. Childs Nerv
Syst 27(10):1535–1541
Rosenfeld EH, Mazzolini K, DeMello AS, Yu YR,
Karediya A, Nuchtern JG et  al (2019) Do ventricu-
loperitoneal shunts increase complications after
laparoscopic gastrostomy in children? J Surg Res
236:119–123
Rosenthal RJ, Hiatt JR, Phillips EH, Hewitt W, Demetriou
AA, Grode M (1997) Intracranial pressure. Effects
of pneumoperitoneum in a large-animal model. Surg
Endosc 11(4):376–380
Sakka L, Coll G, Chazal J (2011) Anatomy and physiol-
ogy of cerebrospinal fluid. Eur Ann Otorhinolaryngol
Head Neck Dis 128(6):309–316
Sankpal R, Chandavarkar A, Chandavarkar M (2011)
Safety of laparoscopy in ventriculoperitoneal shunt
patients. J Gynecol Endosc Surg 2:91–93
Schatlo B, Hamed M, Grote A, Schramm J, Neuloh
G (2013) Gravity assisted vs. medium pressure
valves for communicating hydrocephalus show
similar valve-revision rates. Acta Neurochir (Wien)
155(10):1987–1991
Serarslan Y, Yilmaz A, Çakır M, Güzel E, Akakin A,
Güzel A et al (2017) Use of programmable versus non-
525
programmable shunts in the management of normal
pressure hydrocephalus: A multicenter retrospective
study with cost–benefit analysis in Turkey. Medicine
(Baltimore) 96(39)
Simon TD, Hall M, Riva-Cambrin J, Albert JE, Jeffries
HE, LaFleur B et  al (2009) Infection rates follow-
ing initial cerebrospinal fluid shunt placement across
pediatric hospitals in the United States. J Neurosurg
Pediatr 4(2):156–165
Spader HS, Hertzler DA, Kestle JR, Riva-Cambrin J
(2015) Risk factors for infection and the effect of an
institutional shunt protocol on the incidence of ven-
tricular access device infections in preterm infants. J
Neurosurg Pediatr 15(2):156–160
Spector R, Keep RF, Robert Snodgrass S, Smith QR,
Johanson CE (2015) A balanced view of choroid
plexus structure and function: Focus on adult humans.
Exp Neurol 267:78–86
Stone SS, Warf BC (2014) Combined endoscopic third
ventriculostomy and choroid plexus cauterization
as primary treatment for infant hydrocephalus: a
prospective North American series. J Neurosurg
­
Pediatr 14(5):439–446
Stone JJ, Walker CT, Jacobson M, Phillips V, Silberstein
HJ (2013) Revision rate of pediatric ventriculoperi-
toneal shunts after 15 years. J Neurosurg Pediatr
11(1):15–19
Tomycz LD, Hale AT, George TM (2017) Emerging
insights and new perspectives on the nature of hydro-
cephalus. Pediatr Neurosurg 52(6):361–368
Walker DH, Langer JC (2000) Laparoscopic surgery in
children with ventriculoperitoneal shunts. J Pediatr
Surg 35(7):1104–1105
Wang JY, Amin AG, Jallo GI, Ahn ES (2014) Ventricular
reservoir versus ventriculosubgaleal shunt for post-
hemorrhagic hydrocephalus in preterm infants:
infection risks and ventriculoperitoneal shunt rate. J
Neurosurg Pediatr 14(5):447–454
Warf BC (2005) Comparison of endoscopic third ven-
triculostomy alone and combined with choroid plexus
cauterization in infants younger than 1 year of age: a
prospective study in 550 African children. J Neurosurg
103(6 Suppl):475–481
Wu Y, Green NL, Wrensch MR, Zhao S, Gupta N
(2007) Ventriculoperitoneal shunt complications in
California: 1990 to 2000. Neurosurgery 61(3):557–
562. Discussion 62–3
 Dermal Sinus Tract and Tethered
Cord Syndrome
41
Geraint Sunderland and Jonathan Ellenbogen

41.1 Introduction

Dermal sinus tracts (DST) and tethered spinal

cord (TSC) are two separate but ontogenically
related conditions, which occur as a result of an
interruption in the normal neurulation process
during embryogenesis and early foetal develop-
ment. A DST is a tract lined with stratified squa-
mous epithelium that is in continuity with the
skin surface and deeper structures originating
from the neuroectoderm (Fig. 41.1). Spinal cord
tethering occurs as a result of a multitude of dys-
raphic conditions, including dermal sinus tracts
(Fig.  41.2). While relatively rare, these condi-
tions have potentially serious neurological conse-
Fig. 41.1  Photograph of the lumbar region of a neonate.
Key features to note are the tiny punctum (arrowed) repre-
senting the dermal sinus tract ostium. Also visible are
characteristic cutaneous stigmata, haemangioma and a
hypertrichosis patch

G. Sunderland (*) · J. Ellenbogen

Department of Neurosurgery, Alder Hey Children’s
NHS Foundation Trust, Liverpool, UK
e-mail: geraintsunderland@doctors.org.uk;
jellenbogen@doctors.org.uk

© Springer Nature Switzerland AG 2023 527

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_41
528
Fig. 41.2  Sagittal T2-weighted sequence demonstrating
a dermal sinus tract penetrating from the skin surface to
the intradural space. Note the associated dorsal lipoma
and spinal cord tethering—low level of conus (L5 in this
patient) and dorsally displaced spinal cord
quences, and early recognition and appropriate
management ensure the best possible outcome.
41.2 Historical Overview
Earlier in the last century, little attention was paid
to the small pits and cutaneous stigmata (e.g.
hyper-/hypopigmentation, hypertrichosis, angio-
matous skin changes/haemangioma, lipoma, skin
tag) occasionally found in the midline overlying
the neuroaxis, which were often ignored. Patients
would present to medical attention only when man-
ifesting either neurological or infective sequelae.
The first documented surgery on a spinal dermoid
was recorded in 1875 by Eppinger, referred to as a
spinal cholesteatoma. In 1913, the dermal sinus
and its surgical excision were first described
(Ackerman and Menezes 2003).
Improvements in the understanding of
embryogenesis, specifically that relevant to the
nervous system, enhanced our understanding of
G. Sunderland and J. Ellenbogen
the congenital origin of these lesions, related to
in utero developmental failures.
TCS is a relatively novel pathological entity,
having been first described in terms we under-
stand today in 1953 by Garceau, who coined the
term ‘filum terminale syndrome’ (Garceau 1953).
Garceau also pioneered untethering surgery,
identifying on spinal exploration that the spinal
cords were held under tension by a t­hickened
filum terminale, which he then sectioned. The
term ‘tethered spinal cord’ was first used in a
seminal publication from 1976, which describes
improvement in neurological function in 31
patients following sectioning of a thickened filum
terminale that was causing elongation of the spi-
nal cords (Hoffman et al. 1976). The concept of a
‘normal’ adult spinal conus level was introduced
in 1944 by Reimann and Anson, who reviewed
800 adult cadaver spinal cords and concluded
that the average level of conus was the lower
third of the L1 vertebrae (Reimann and Anson
1944). The consensus now is that conus termina-
tion below the L2 inferior end plate qualifies as
‘low-lying’ and lower limb and autonomic symp-
toms in this context could be attributed to TCS
(Hertzler et al. 2010).
The definition of the normal conus level in
infants came later, and as recently as the 1990s,
there remained some controversy over what
should be considered normal. Ultrasound and,
subsequently, magnetic resonance imaging
(MRI) spinal studies have established that the
conus usually achieves its adult level by birth or
by 2 months of age, at the latest (Di Pietro 1993;
Kesler et al. 2007).
The treatment of DST, TCS and the rest of the
spectrum of spinal dysraphic conditions has been
impacted most profoundly by the advent of
MRI. This has revolutionised diagnosis and is the
gold standard investigation for pre-operative
assessment and planning, allowing surgeons to
carefully define anatomy and deeper relation-
ships of dysraphic lesions to neural tissue
(Fig.  41.2) (Radmanesh et  al. 2010). From an
operative standpoint, modern high-definition
operating microscopes allow greater precision
41  Dermal Sinus Tract and Tethered Cord Syndrome 529

when operating. In addition, intra-operative neu- The process of neurulation, wherein the ecto-
rophysiological monitoring (IOM) allows the derm of the trilaminar embryo folds in the mid-
surgeon to map in real-time functional neural tis- line (neuroectoderm) to form the neural tube,
sue using electrical stimulation. These tools per- occurs between post-ovulatory days (POD) 18
mit better clearance of abnormal tissue with and 28. The spectrum of cranial and spinal dys-
preservation of functional neural tissue. The raphic conditions exists as a result of failure of
result is improved long-term outcomes with far this process.
lower rates of neurological complication. Following folding along the neural groove
between POD 22 and 28, neural tube closure
begins at the cervicomedullary junction and con-
41.3 Incidence tinues in a discontinuous manner, both cephalad
towards the anterior neuropore and caudad
The quoted incidence of DST is estimated at 1 in towards the posterior neuropore, with the poste-
2500 births; however, this is likely an over-­ rior neuropore closing last on POD 27. This
estimation, as it is based on pre-MRI era studies requires neuroectoderm cells at the leading edge,
and thus a proportion of cases were likely coc- the neural folds, to fuse and detach from adjacent
cygeal pits, a much more common and benign cutaneous ectoderm cells. In the case of DST, a
entity present in as many as 4% of births focal failure of the cutaneous ectoderm to detach
(Ackerman and Menezes 2003; Singh et  al. from the neural tube results in a persistent chan-
2015). Between 10 and 60% of cases of DST will nel connecting the skin to deeper structures
present with an infective complication ranging within the neuroaxis. As a result of their embryo-
from simple superficial skin infection to more logic origin, they are almost always dorsal and in
serious CNS infections. All babies will have the midline (though ostia at the skin can be dis-
intact neurology at birth, but up to 90% will have placed laterally by underlying lipomas or inclu-
some deficit by the time of referral to neurosurgi- sion tumours). DSTs then extend deep through
cal services (Ackerman and Menezes 2003; the tissues of mesodermal origin (fascia, muscle,
Jindal and Mahapatra 2001;  Radmanesh et  al. bone). Most DSTs penetrate the dura (80–90%)
2010). and often attach to neural tissue (conus medul-
The finding of a low-lying conus on ultra- laris, cauda equina or filum terminale). They are
sound is as frequent as 50% at birth, improving lined with tissues of cutaneous ectodermal origin
to <20% by 2 months. The prevalence of TCS, (stratified squamous epithelium) and may also
however, is much lower, affecting approxi- contain related structures, such as hair, sebaceous
­
mately 0.1% of all children (Asil and glands and apocrine/eccrine sweat glands. The
Yaldiz 2018). It has a higher prevalence in those lumbosacral area is the most commonly affected
with bladder dysfunction and enuresis (1.4%) spinal region (32–54%), followed by the lumbar
(Bademci 2006). TCS and low-lying conus are (32–43%), the thoracic (3–14%) and least com-
closely associated with spina bifida occulta
which occurs in 12% of the general population Table 41.1  Summary of the key clinical differences to
(Cools 2014). distinguish between dermal sinus tract and benign coccy-
geal pit
Dermal sinus tract Coccygeal pit
41.4 Aetiopathogenesis Located above the natal Located within the natal cleft
cleft
Frequent cutaneous No associated cutaneous
A working knowledge of nervous system embry- stigmata stigmata
ology is required to appreciate the relationship Tract orientated Caudally orientated tract
between the DST, TCS and the myriad of other cranially
related dysraphic conditions. Base of tract not visible Typically shallow with
visible base
530 G. Sunderland and J. Ellenbogen

monly the cervical region (1–18%) (Ackerman physiological mechanism is unknown though a
and Menezes 2003;  Jindal and Mahapatra number of theories exist.
2001;  Singh et  al. 2015). DSTs are also fre-
quently found in association with other spinal
dysraphic abnormalities including lipoma, lipo- 41.5 Pathophysiology
meningocoele, lipomyelomeningocoele, myelo- and Pathology
meningocoele and split cord malformations.
Due to conus ascent during development, the DSTs are lined with stratified squamous epithe-
dermal sinus tract has a component that is crani- lium and may also contain dermal structures,
ally orientated—useful to help distinguish from such as hair follicles, fat, fibrous tissue and glan-
coccygeal pits, which can occasionally be deep dular structures. These tissues continue to func-
but are found within the natal cleft (DSTs are tion in the same way as they would on the skin
found above the natal cleft), and their tracts uni- surface, which can lead to an accumulation and
versally run straight or caudally (Table 41.1). sometimes discharge of sebaceous or keratinous
A number of pathological processes can occur debris via the ostia. A build-up of keratin within
during primary and secondary neurulation that the cul de sac of the tract will produce an epider-
lead to tethering of the spinal cord. The most fre- moid tumour (13%); alternatively, in the pres-
quent is a failure of regression of the caudal cell ence of dermal appendages, secreted sebum can
mass during secondary neurulation. This will accumulate to produce a dermoid tumour (87%),
result in a thickened or fibrous and inelastic filum which will frequently also contain hair but also
terminale. This is associated with a low-lying occasionally fat, blood, bone and even teeth
conus in the majority of cases; however, TCS can (Gupta et al 2005). These inclusion tumours can
exist in patients with a conus that terminates at a expand and exert mass effect on adjacent neural
normal level (~18% of cases) but is dorsally dis- tissue manifesting as neurological deficit, motor
placed due to an inelastic filum terminale (Warder weakness, sensory impairment, pain or bladder/
and Oakes 1993). Spinal lipomas, as well as lipo- bowel dysfunction. Furthermore, inclusion
matous tissue within the filum, occur as a result tumours can rupture or leak contents around and
of errors during secondary neurulation and also a
failure of regression of the caudal cell mass.
These can directly exert mass effect on neural tis-
sue but also adhere to and extend through mesen-
chymal structures tethering the spinal cord at that
level.
Tethering of the spinal cord is also associated
with other congenital dysraphic conditions, such
as myelomeningocoele and split cord malforma-
tion, as well as acquired causes, such as arach-
noiditis occurring secondary to infection or
trauma.
No specific genetic abnormality has been asso-
ciated with the development of DSTs, TCS nor
spinal dysraphism overall. The only undisputed
risk factor is relative maternal folate deficiency
based on the observation that supplementation
with 4 mg folate per day for 3 months prior to, and
Fig. 41.3  Sagittal T1-weighted MRI with gadolinium
following, conception is protective against the contrast demonstrating an extensive intradural abscess in
range of neural tube defects. The exact patho- association with a dorsal dermal sinus tract and dermoid
inclusion tumour
41  Dermal Sinus Tract and Tethered Cord Syndrome 531

into the central nervous system, exciting an exag-

gerated inflammatory reaction, known as chemi-
cal or aseptic meningitis. Arachnoiditis, clumping
of neural structures and tethering can ensue.
DSTs provide a route of ingress for bacteria
from the skin surface into deeper tissues, which A
can then become colonised and infected. This
may manifest as a local cutaneous infection or C
more serious deeper soft tissue infection and B
abscess. This deep communication extends to
the dura, and thus central nervous system infec-
tions, meningitis (often recurrent) and even
extradural, intradural or occasionally intramed-
ullary abscesses are a risk with potentially dev- Fig. 41.4  Histological section of an excised lumbar der-
astating consequences (Fig.  41.3) (Foster mal sinus tract. Note the keratin debris (A) contained
within the stratified squamous epithelium (B) lined lumen
et  al.  2019). Infection can spread via the CSF
complete with glandular dermal appendages (C)
circulation, and reports of cerebral abscesses as a
consequence of DST exist (Emami-Naeini
et al. 2008). strated an escalating risk of persistent neurological
Tethered spinal cord and TCS have been deficit between children presenting <1 year
defined thus: ‘a potentially reversible stretch-­ (50%) and >1 year (90%). Abnormal reflexes
induced functional disorder of the spinal cord, were the most common examination finding
with the cord’s caudal end anchored by inelastic (37.5% vs. 75%), followed by motor weakness
structures’. The underlying pathophysiology (19% vs. 67%), sensory change (0% vs. 58%)
leading to neurological dysfunction has been and sphincteric dysfunction (6% vs. 42%)
studied in animal models and in human subjects (Ackerman and Menezes 2003).
undergoing detethering surgery (Stetler
et  al.  2010). Animal models have yielded evi-
dence of relative ischaemia and impaired oxida- 41.6 Pathology
tive phosphorylation within spinal cords under
both acutely and chronically applied tension Histological examination of excised DSTs dem-
(Yamada et al. 2007). Blood flow studies in chil- onstrate a lumen lined with stratified squamous
dren undergoing detethering surgery demonstrate epithelium and surrounded by dermal tissue with
a greater than twofold increase in spinal cord or without associated appendages, hair follicles
blood flow following sectioning of the filum ter- and sweat/sebaceous glands (Fig. 41.4).
minale (Schneider et al. 1993). This hypoxic state Unsurprisingly, the most common infective
leads to dysfunction in neuronal mitochondria organisms are Staphylococcus species, given the
and ultimately neuronal dysfunction and cell route from the skin surface. Proximity of the
death. It is postulated that due to cumulative cell DST ostium to the natal cleft, however, results
loss over time, patients with longer duration of in the risk of faecal contamination and the intro-
symptoms have less prospect of recovering neu- duction of associated organisms (E. coli,
rological function following restorative surgery. Enterococcus, etc.) (21).
Non-infectious neurological sequelae of DST
and TCS tend to be slowly progressive, and
patients presenting at more advanced age are thus
more likely to have an established deficit.
Ackerman et  al. reported their 30-year experi-
ence of managing DSTs in 2002 and demon-
532
Fig. 41.5  Photographs demonstrating key features of
coccygeal pits: location within the natal cleft within
2.5  cm of the anus and absence of cutaneous stigmata.
Table 41.2  Cutaneous stigmata associated with spina
bifida occulta
Subcutaneous lipoma
Hypertrichosis
Haemangioma
Skin appendages/skin tags
Areas of abnormal pigmentation
41.7 Diagnosis
41.7.1 Clinical Presentation
In recent decades, there have been substantial
improvements made in the recognition of the
significance of dorsal midline skin lesions
amongst paediatricians, general practitioners
and the like. Furthermore, the performance of
routine screening physical examination means
that, increasingly, even subtle lesions are identi-
fied in infancy and referred prior to the onset of
symptoms or development of neurological or
infective complications.
DSTs can be difficult to identify, as the ostium
at the skin surface is often only a small dimple or
punctum and may be obscured by other cutane-
ous stigmata. Care should be taken to distinguish
DST dimples from the benign coccygeal pit.
G. Sunderland and J. Ellenbogen
While they can be deep, retraction on the buttocks will
often permit visualisation of the base as demonstrated
Coccygeal pits are found within the natal cleft
and have no communication with the nervous
system (Fig. 41.5). Location within 2.5 cm of the
anus also favours a diagnosis of coccygeal pit.
DSTs are invariably found cephalad to the natal
cleft. Skin lesions are frequently associated with
DSTs and include hypertrichosis (hairy patch),
areas of abnormal pigmentation, haemangiomas,
swelling due to subcutaneous lipomas and more
rarely skin tags or appendages (Table  41.2,
Fig.  41.6). The presence of such findings dor-
sally, on or near the midline, should alert clini-
cians to the potential of underlying spina bifida
occulta spectrum lesions, including DST, and
warrant further investigation. In addition, there
may be discharge from the ostium as a result of
build-up of keratinous material or sebaceous
secretions. Evidence of localised infection, ery-
thema or purulent discharge is a common finding
also. Patients may also primarily present with
more serious CNS infections, such as meningitis
or abscess (14–20%) (Ackerman and Menezes
2003; Radmanesh et al. 2010; Singh et al. 2015).
Recurrent episodes of bacterial meningitis in
particular are highly suspicious and mandate
close examination and investigation for occult
connections between the CNS and external
41  Dermal Sinus Tract and Tethered Cord Syndrome 533

a b

c d

Fig. 41.6  Photographs of common cutaneous stigmata associated with spina bifida occulta. (a) Subcutaneous lipoma.
(b) Hypertrichosis. (c) Haemangioma. (d) Skin appendage

structures, such as neurocutaneous fistulas

(DST), enterocutaneous fistulas or occult CSF
leak. It is advised that in the context of meningi-
tis with a suspected lumbar or lumbosacral DST,
diagnostic lumbar puncture should be delayed
until spinal imaging (MRI) is performed to avoid
iatrogenic injury to a tethered and low-lying spi-
nal cord. Empirical antibiotic treatment pending
these investigations would be appropriate in this
situation.
Investigations performed following identifica- Fig. 41.7 Transcutaneous ultrasound of the neonatal
tion of skin lesions (see above) will most fre- spine showing the level of conus termination (arrowed) at
quently precede the diagnosis of tethered spinal the lower end of the L2 vertebrae
cord. Occasionally, there may be development of
progressive neurological deterioration. Common regression to crawling in a previously ambulant
presenting features of TCS are lower limb weak- child. Sensory disturbances, such as paraesthesia
ness and related altered gait, including toe-­ or lower limb numbness, are less frequent though
walking, a failure to meet walking milestones or perineal sensory impairment is frequently found
534
in patients with urological dysfunction (see
below). Orthopaedic presentations comprise pro-
gressive scoliosis and foot and ankle deformities,
such as talipes equinovarus, pes cavus and claw-
ing of toes. These should prompt consideration of
potential neural tension. Urological dysfunction
is another cardinal feature of TCS; enuresis or
failure to achieve urinary continence is a com-
mon finding and progressive. Frequent urinary
tract infections (UTI) are associated with altered
or impaired voiding and may be the only urologi-
cal feature identified.
Obviously, there are a multitude of causes for
the disparate clinical features identified above, and
TCS would be low down the list of differentials for
a child presenting with UTI, for example.
Clustering of these symptoms, however, should
cue reflection over a potential unifying pathology.
41.7.2 Investigations
Ultrasound
Ultrasound scan (USS) evaluation of the lumbar
spine in the neonatal and infant population (<6
months) is an effective and readily accessible
method both for assessing soft tissues and also
for establishing the level of the conus medullaris
(Fig. 41.7) (Unsinn et al. 2000). USS can be used
to reliably assess conus level and the spinal cord
a b
Fig. 41.8 MR images demonstrating tethered spinal
cord. (a, b) Sagittal T1 and T2-weighted MRI showing
intradural lipoma causing spinal cord tethering. (c)
G. Sunderland and J. Ellenbogen
structure and thus may identify a spinal cord syr-
inx, cystic lesion (e.g. inclusion tumour) or
lipoma. In addition, a skilled operator may be
able to evaluate the filum to some extent but not
as reliably as other imaging methods. As a time-­
resolved study, USS has utility in evaluating
conus mobility—i.e. whether it pulsates or moves
with respiration or changes in position (prone-­
lateral) (Raghavendra et al. 1983).
USS may also identify a sinus tract, though it
can fail to establish the level of penetration or
relationship to deeper structures. Ultrasound is
limited in resolution of tissues and cannot be
used to reliably assess anatomical detail. As such,
it has utility as a screening tool, but any abnor-
malities identified will require further evaluation
with MRI (Fitzgerald 2011).
Plain X-Ray
Plain X-rays have limited utility in the routine
workup of DST or TCS; however, they are essen-
tial for evaluation of any scoliotic deformity. A
whole spine X-ray, for example, using the EOS
system, is the gold standard as it permits accurate
evaluation of the scoliotic deformity (structural
osseous abnormalities, major and minor curves,
apex and sagittal/coronal balance). There may be
dysraphic structural abnormalities evident; how-
ever, cross-sectional imaging modalities (CT/
MRI) are much more sensitive to these.
c d
Sagittal and (d) axial T1-weighted MRI sequences dem-
onstrating tethered cord due to a fatty and hypertrophic
filum terminale
41  Dermal Sinus Tract and Tethered Cord Syndrome
Computed Tomography (CT)
For evaluation of bony anatomy, CT scanning is
unrivalled; however, such detail is rarely required
in the management of DST and TCS, except in
cases of split cord malformation or in the global
evaluation of scoliosis related to tethered spinal
cord.
Identification of bony spicules within split
cord malformations is useful for pre-operative
planning and also permits confident radiological
classification (SCM type I, 2 separate cords with
duplicated dural sheath often associated with a
bony or fibrocartilaginous spur; SCM type II, 2
separate cords sharing a single dural sheath—not
associated with bony or fibrocartilaginous spur).
CT myelogram has a rare indication in cases
where MRI is contraindicated. Contrast injected
via LP outlines neural structures within the theca
and permits evaluation of conus lie and potentially
any tethering lesions and their relative anatomy.
Magnetic Resonance Imaging
MRI is the gold standard imaging modality for
the evaluation of DST and TCS. In addition to the
identification of DST and/or tethered spinal cord,
MRI provides a wealth of further information
(Hall et al. 1988). Sagittal lumbar spine sections
permit visualisation of the cord termination level
and position of the conus within the spinal canal.
In cases of TCS, axial lumbosacral images will
frequently demonstrate a thickened filum termi-
nale, again often containing fatty deposits or
bearing lipomatous tissue—readily identified on
T1 sequences where fat is bright white within the
dark CSF (Fig. 41.8) (Lane et al. 2017). In addi-
tion, MRI performed in the prone position further
reveals information on the degree of mobility of
the spinal cord and thus gives an indication of the
extent of tension on the cord (Stamates
et al. 2018).
Modern MRI (1.5T and above) returns high-­
resolution images of the subcutaneous soft tis-
sues, along with any contained anomaly, DST or
associated inclusion tumour. It allows the evalu-
ation of the extent of penetration of DSTs within
the spinal canal, i.e. extra-/intradural or intra-
medullary. Dysraphic lesions, such as lipomy-
535
elomeningocoele and lipoma, can be identified
and their relationship to neural structures evalu-
ated (Kumar et al. 2017).
On identification or clinical suspicion of a dys-
raphic lesion, and following diagnosis of a teth-
ered spinal cord, an MRI of the whole neuroaxis
(brain and whole spine including the sacrum)
should be performed. This is done to assess the
rest of the cord for the presence of associated syr-
inx, the craniocervical junction for evidence of
secondary Chiari malformation (descent of the
cerebellar tonsils >5mm below the level of the
foramen magnum) and intracranially for any evi-
dence of hydrocephalus (Mehta 2017).
There are limitations to MRI investigation in
these conditions. The intradural portion of DSTs
is often not well visualised, and the relationship
to neural structures is difficult to interpret, espe-
cially in the presence of an associated inclusion
tumour or infection. Post-meningitic changes
include adhesion and clumping of neural tissues,
which can further limit the interpretation of ana-
tomical relationships within the theca. Movement
artefacts within infants due to rapid pulse and
respiration compromise image integrity
(Barkovich 1991). Access to MR imaging may be
limited in lower-income countries and those
without equitable access to healthcare. In addi-
tion, children are required to remain still for the
duration of the scan in a hot and noisy scanner.
For infants <3 months, who will usually lie still
enough if fed immediately prior to scanning (the
so-called feed-and-wrap protocol), this is feasi-
ble. Any child older than 3 months and under 5
years of age will require general anaesthesia in
order to obtain useful images.
Urological Investigations
Evaluation of urological function is an essential
part of the workup for any child with a diagnosis
of spinal dysraphism spectrum disorder and, in
particular, with a history of TCS. Investigations
should include both structural and functional
assessment, as structural abnormalities are less
frequently associated. One published series of 35
surgically managed DSTs quotes a 25% inci-
dence of structural abnormalities identified on
536
USS but a 71% incidence of abnormal urody-
namic findings (Radmanesh et al. 2010). A renal
ultrasound will return information on the size and
structural integrity of the renal tract and will
identify cases of hydronephrosis/hydroureter
when present, in addition to any renal scarring or
atrophy. Urodynamic studies are immensely use-
ful in the baseline assessment and follow-up of
patients. Impaired urodynamics may precede the
diagnosis of TCS and may be the precipitant to
investigation and can help guide the requirement
for and timing of surgical intervention.
41.8 Differential Diagnosis
Benign congenital coccygeal dimples, also
referred to as coccygeal pits or sacral dimples,
represent the major differential diagnosis to
DSTs. Features to distinguish them clinically are
summarised in Table 41.1. Once a confident diag-
nosis of coccygeal pit is made, no further investi-
gation or intervention is required.
As discussed above, DST and TCS are condi-
tions on the spinal dysraphic spectrum, and they
may coexist with other conditions. DSTs are often
seen in conjunction with myelomeningocoele,
lipomyelomeningocoele, spinal lipomas and inclu-
sion tumours. TCS is a feature associated with the
full range of spinal dysraphic conditions.
41.9 Management
The primary management strategy for both DST
and TCS is surgical because of the potential for
both serious and disabling consequences. The main
goal of surgery is to protect and preserve neuro-
logical function. There is the potential for recovery
of impaired neurological function; however, this is
not guaranteed, and as discussed in scenarios of
chronic compromise of neural structures, the
chances of improvement are lower (Warder 2001).
In the scenario of asymptomatic tethered cord,
there is some controversy over whether prophylac-
tic surgery is justified or whether expectant conser-
vative management is more appropriate (Hertzler
et  al. 2010). Surgical intervention is clearly indi-
cated on development of neurological compromise,
G. Sunderland and J. Ellenbogen
however, to prevent further deterioration or to max-
imise chances of good outcome.
Dermal Sinus Tract
Dermal sinus tracts have a relatively high compli-
cation rate if left untreated and so, when identi-
fied, should be scheduled for surgical excision
once the necessary evaluation and investigation is
complete. In the circumstance of a presentation
with meningitis or superficial infection, it would
be sensible to treat the infection and allow any
associated inflammation to settle before planning
a surgical excision.
41.10 Surgical Technique
The patient is positioned prone on the operating
table with care paid to padding of pressure points.
Prophylactic antibiotics are administered on
induction of general anaesthesia.
Neurophysiological monitoring is performed in
the majority of surgeries involving manipulation
or direct dissection on the surface or within the
spinal cord. This may comprise somatosensory
evoked potentials (SSEPs) alone or with motor
evoked potentials (MEPs), which provide longitu-
dinal assessment of spinal cord function and act as
an ‘early warning system’ to help prevent irrevers-
ible neurological injury. In addition, they allow
‘mapping’ of neurological structures and identifi-
cation of non-functional structures, such as the
filum terminale, which may be safely sectioned.
Dermal Sinus Tract
The sinus ostium is sharply excised in an ellipse
of surrounding skin, and the tract is circumferen-
tially dissected from surrounding soft tissues, as
it penetrates though deeper layers.
Any inclusion tumours associated with the
DST should be entirely excised due to their pro-
pensity to rupture and excite marked local inflam-
matory response, in addition to harbouring
infective organisms. Care should be taken to
avoid rupturing inclusions adjacent to or inside
the dura, as spillage of contents can result in a
marked aseptic or ‘chemical’ meningitis, which
may cause significant morbidity, and is also a risk
factor for the development of hydrocephalus.
41  Dermal Sinus Tract and Tethered Cord Syndrome
A laminectomy cranial to the level of DST
entry into the spinal column is performed to per-
mit access to the neural structures within. This
may be extended cranially or caudally as required,
to resect any intraspinal component.
The majority of DSTs are intimately related to
the dura and frequently breach it. Complete exci-
sion of any intradural component and complete
detachment of neural structures is crucial to min-
imise the risk of retethering. The durotomy is
performed in an elliptical fashion encompassing
the tract. Some sinus tracts end abruptly with
dural attachment, which is readily apparent after
dural opening. In these cases, after confirming
normal intradural anatomy, the dura is closed,
and the wound can be closed in layers. If there is
any intradural component, this should be system-
atically and carefully dissected from neural tissue
and delivered en bloc.
Tethered Spinal Cord
Surgical detethering is most frequently per-
formed via the division of hypertrophic or fatty
filum terminale. The outline of the procedure is
described here. More complicated detethering
surgery related to spinal lipomas, lipomeningo-
coeles or myelomeningocoeles is beyond the
scope of this chapter.
Prior to surgery, the conus level is established,
and X-ray guidance is used to mark the spinal
level. A midline incision is performed, followed
by dissection through the soft tissues to the spinal
column. Subperiosteal dissection to expose the
vertebral lamina precedes a small laminotomy to
expose the dura. A dural opening is made in a lon-
gitudinal fashion, and the thickened filum is often
readily identified dorsal to the nerve roots. The
filum is typically paler than the adjacent nerve
roots, and there is a characteristic and prominent
tortuous blood vessel running along its length.
The filum is dissected using a nerve root probe
and suspended in a Cottonoid patty to isolate from
nearby structures. Electrical stimulation (as dis-
cussed above) confirms non-functional status, and
the filum can be confidently sectioned.
The durotomy is sutured closed aiming for a
watertight closure, which may require the appli-
cation of an artificial dural patch plus possibly
the use of a fibrin glue. The soft tissues are then
537
opposed in layers with absorbable sutures. The
skin is closed with absorbable sutures.
Outcomes
Given that the stated aim of surgery in both DST
and TCS is stabilisation of neurological function,
evaluation of outcomes is not straightforward,
especially in patients who are asymptomatic at
presentation. Further complicating evaluation is
the variability in patient status at presentation.
Likewise, summarising outcomes for untethering
surgery is complicated, as they are very much
dependent on the underlying pathology.
Sectioning a thickened and fatty filum, for exam-
ple, is an order of magnitude less complex and
potentially morbid than resecting a dorsal lipoma.
Nevertheless, published outcomes for both
straightforward DST and TCS surgery are very
good with 90–95% of those operated for DST
demonstrating improved or stable neurological
function on follow-up and the majority of
patients, undergoing detethering surgery of any
description, likewise demonstrating stabilisation
or improvement in symptoms (Ackerman and
Menezes 2003; Jindal and Mahapatra 2001;
Singh et al. 2015; Warder 2001). It follows that
patients with worse presenting neurological state
have poorer outcomes. DSTs associated with
inclusion tumours and CNS infection are also
associated with worse long-term neurological
outcome.
Complications
Watertight dural closure is critical following sur-
gery for both DSC and TCS, to prevent the post-­
operative morbidity associated with CSF leak.
Fastidious attention should be paid to dural clo-
sure, which may require patching with autolo-
gous or synthetic/animal-derived allograft if a
substantial section of dura is resected along with
the transgressing lesion. Following surgery, a
48-72 hour period of flat bed rest is advocated by
many to reduce hydrostatic pressure on the dural
repair, to allow initial healing of the durotomy to
establish.
Serious complications following surgery for
DST or untethering surgery are very rare.
Infections following excision of DST or deteth-
ering are fortunately rare and are almost univer-
538
sally superficial, localised and uncomplicated to
manage. In the era of intra-operative neurophysi-
ological monitoring, new and permanent post-­
operative neurological injury is very uncommon.
The surgeon’s main fear is that of a CSF leak,
resulting in prolonged hospital stay and the asso-
ciated physical and psychological morbidity.
CSF leaks can be difficult to manage. Simple
interventions, such as lumbar drainage, which
diverts the flow of CSF away from the nascent
dural repair may be effective, but occasionally, a
return to theatre and formal dural repair is
necessary.
Retethering, due to scarring from previous
surgery, is a well-established phenomenon,
affecting approximately 15–20% of all cases
(Warder 2001). Neurological deterioration fol-
lowing a stable period after index surgery strongly
suggests retethering and warrants investigation.
Retethering following simple detethering surgery
(sectioning thickened or fatty filum terminale or
uncomplicated dermal sinus tract) is relatively
uncommon (7.5%) (Vassilyadi et al. 2012).
41.11 Conclusion
Dermal sinus tracts are an uncommon, but impor-
tant, pathological entity with potentially serious
consequences (meningitis, spinal abscess, per-
manent neurological disability) if left untreated.
Identification or clinical suspicion of a dermal
sinus tract should prompt investigation with MRI
and, when the diagnosis is confirmed, followed
by surgical intervention. Risks associated with
surgery to excise these lesions and associated
inclusion tumours are low; thus, there is really no
justification for conservative management.
Spinal cord tethering is associated with the
whole spectrum of dysraphic spinal conditions
and may manifest as tethered cord syndrome.
Neurological deficit is progressive, and delay in
intervention is associated with poorer outcomes.
Spinal cord release should be performed early in
cases associated with spinal cord or subcutane-
ous lipomas. Expectant conservative manage-
ment may be pursued in asymptomatic cases, but
G. Sunderland and J. Ellenbogen
surgical detethering should be considered at the
earliest onset of symptoms.
Further Reading
Ackerman LL, Menezes AH (2003) Spinal congeni-
tal dermal sinuses: a 30-year experience. Pediatrics
112:641–647
International Society of Paediatric Neurosurgery Guide:
https://www.ispn.guide/. A freely available online
educational resource providing up to date and evi-
dence based information on all paediatric neurosurgi-
cal conditions. Compiled by internationally renowned
clinicians
Yamada S, Iacono RP, Andrade T, Mandybur G, Yamada
BS (1995) Pathophysiology of tethered cord syn-
drome. Neurosurg Clin N Am 6:311–323
References
Ackerman LL, Menezes AH (2003) Spinal congenital der-
mal sinuses: a 30-year experience. Pediatrics 112(3 Pt
1):641–647
Asil K, Yaldiz M (2018) Conus medullaris levels on
ultrasonography in term newborns: normal levels and
dermatological findings. J Korean Neurosurg Soc
61(6):731–736
Bademci G, Saygun M, Batay F, Cakmak A, Basar H,
Anbarci H et al (2006) Prevalence of primary tethered
cord syndrome associated with occult spinal dysra-
phism in primary school children in Turkey. Pediatr
Neurosurg 42(1):4–13
Barkovich AJ, Edwards MS, Cogen PH (1991) MR evalu-
ation of spinal dermal sinus tracts in children. Am J
Neuroradiol 12(1):123
Cools MJ, Al-Holou WN, Stetler WR, Wilson TJ,
Muraszko KM, Ibrahim M et al (2014) Filum termi-
nale lipomas: imaging prevalence, natural history, and
conus position. J Neurosurg Pediatr 13(5):559
DiPietro MA (1993) The conus medullaris: normal
US findings throughout childhood. Radiology
188(1):149–153
Emami-Naeini P, Mahdavi A, Ahmadi H, Baradaran N,
Nejat F (2008) Brain abscess as a manifestation of spi-
nal dermal sinus. Ther Clin Risk Manag 4:1143–1147
Fitzgerald K (2011) Ultrasound examination of the neona-
tal spine. Australas J Ultrasound Med 14:39–41
Foster MT, Moxon CA, Weir E, Sinha A (2019) Dermal
sinus tracts. BMJ 366:l5202
Garceau GJ (1953) The filum terminale syndrome (the
cord-traction syndrome). J Bone Joint Surg Am
35-a(3):711–716
Gupta DK, Shastank RR, Mahapatra AK (2005) An
unusual presentation of lumbosacral dermal sinus
41  Dermal Sinus Tract and Tethered Cord Syndrome
with CSF leak and meningitis. Pediatr Neurosurg
41(2):98–101
Hall WA, Albright AL, Brunberg JA (1988) Diagnosis of
tethered cords by magnetic resonance imaging. Surg
Neurol 30(1):60–64
Hertzler DA 2nd, DePowell JJ, Stevenson CB, Mangano
FT (2010) Tethered cord syndrome: a review of the
literature from embryology to adult presentation.
Neurosurg Focus 29(1):E1
Hoffman HJ, Hendrick EB, Humphreys RP (1976) The
tethered spinal cord: its protean manifestations,
diagnosis and surgical correction. Childs Brain
2(3):145–155
Jindal A, Mahapatra AK (2001) Spinal congenital dermal
sinus: an experience of 23 cases over 7 years. Neurol
India 49(3):243–246
Kesler H, Dias MS, Kalapos P (2007) Termination of the
normal conus medullaris in children: a whole-spine
magnetic resonance imaging study. Neurosurg Focus
23(2):E7
Kumar J, Afsal M, Garg A (2017) Imaging spectrum of
spinal dysraphism on magnetic resonance: A pictorial
review. World J Radiol 9(4):178–190
Lane J, Dias M, Iantosca M, Zacharia T, Tubbs RS, Rizk
E (2017) Dermal sinus tract of lumbosacral spine in
children: patterns on magnetic resonance imaging and
scoring system. Cureus 9(12)
Mehta DV (2017) Magnetic resonance imaging in pae-
diatric spinal dysraphism with comparative useful-
ness of various magnetic resonance sequences. J Clin
Diagn Res 11(8):TC17–TC22
Radmanesh F, Nejat F, El Khashab M (2010) Dermal sinus
tract of the spine. Childs Nerv Syst 26(3):349–357
Raghavendra BN, Epstein FJ, Pinto RS, Subramanyam
BR, Greenberg J, Mitnick JS (1983) The tethered spi-
539
nal cord: diagnosis by high-resolution real-time ultra-
sound. Radiology 149(1):123–128
Reimann AF, Anson BJ (1944) Vertebral level of termina-
tion of the spinal cord with report of a case of sacral
cord. Anat Rec 88(1):127–138
Schneider S, Rosenthal A, Greenberg B, Danto J (1993) A
preliminary report on the use of laser-Doppler flowm-
etry during tethered spinal cord release. Neurosurgery
32:214–218
Singh I, Rohilla S, Kumar P, Sharma S (2015) Spinal dor-
sal dermal sinus tract: An experience of 21 cases. Surg
Neurol Int 6:S429–S434
Stamates MM, Frim DM, Yang CW, Katzman GL, Ali
S (2018) Magnetic resonance imaging in the prone
position and the diagnosis of tethered spinal cord. J
Neurosurg Pediatr 21(1):4–10
Stetler WR Jr, Park P, Sullivan S (2010) Pathophysiology
of adult tethered cord syndrome: review of the litera-
ture. Neurosurg Focus 29(1):E2
Unsinn KM, Geley T, Freund MC, Gassner I (2000) US
of the spinal cord in newborns: spectrum of normal
findings, variants, congenital anomalies, and acquired
diseases. Radiographics 20(4):923–938
Vassilyadi M, Tataryn Z, Merziotis M (2012) Retethering
in children after sectioning of the filum terminale.
Pediatr Neurosurg 48(6):335–341
Warder DE (2001) Tethered cord syndrome and occult
spinal dysraphism. Neurosurg Focus 10(1):e1
Warder DE, Oakes WJ (1993) Tethered cord syndrome
and the conus in a normal position. Neurosurgery
33(3):374–378
Yamada S, Won DJ, Pezeshkpour G, Yamada BS, Yamada
SM, Siddiqi J et  al (2007) Pathophysiology of teth-
ered cord syndrome and similar complex disorders.
Neurosurg Focus 23(2):E6
 Part VI
Anterior Abdominal Wall Defects
 Omphalomesenteric Duct
Remnants
Ampaipan Boonthai, Dhanya Mullassery,
and Paul D. Losty
42.1 Introduction
The omphalomesenteric (or vitellointestinal) duct
is an embryonic communication between the
primitive yolk sac and the developing midgut.
During normal development at the sixth week of
embryogenesis, the midgut loop elongates and
herniates into the umbilical cord. Within the
‘physiological umbilical hernia’, the midgut
rotates 90° counterclockwise around the axis of
the superior mesenteric artery. At the same time,
as the midgut elongates, the lumen of the ompha-
lomesenteric duct begins a process of obliteration.
By the tenth week of early foetal development,
A. Boonthai
Paediatric Surgery and Transplantation Consultant
Surgeon, Division of Pediatric Surgery, Ramathibodi
Hospital, Mahidol University, Bangkok, Thailand
e-mail: ampaipan.nam@mahidol.edu
D. Mullassery
Consultant Paediatric Surgeon, Great Ormond Street
Hospital for Sick Children, University College
London, London, UK
e-mail: dhanya.mullassery@gosh.nhs.uk
P. D. Losty (*)
Alder Hey Children’s Hospital NHS Foundation
Trust, School of Health and Life Science, University
of Liverpool, Liverpool, UK
e-mail: paul.losty@liverpool.ac.uk
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_42
42
the midgut returns to the abdominal cavity and the
omphalomesenteric duct becomes a thin fibrous
band, which undergoes resorption. Persistence of
the duct leads to a spectrum of anomalies that can
present clinically in the newborn period, infancy,
or later childhood years.
42.2 Variant Pathology
of Omphalomesenteric Duct
Remnants
1. Meckel’s diverticulum (Fig.  42.1). The ileal
segment remnant of the duct remains patent
and usually contains heterotopic gastric
mucosal tissue. The diverticulum may be con-
nected to the umbilicus by a fibrous band, if
the obliterated duct fails to be fully resorbed.
2. Umbilical fistula (Fig. 42.2). Completely pat-
ent omphalomesenteric duct connects the ileal
segment of the small intestine to the anterior
abdominal wall.
3. Omphalomesenteric cyst (Fig. 42.3) develops
when a segment in the midportion of the duct
remains patent whilst each corresponding end
portion of the tract obliterates.
4. A persistent fibrous cord (Fig. 42.4) connects
the umbilicus to the small intestine where the
duct here obliterates but is not fully resorbed.
5. Umbilical polyp (Fig. 42.5). A bright red nod-
ule of sequestered ectopic gastrointestinal tis-
sue may reside in the umbilical dimple.
543
544 A. Boonthai et al.

Fig. 42.1  Meckel’s diverticulum Fig. 42.3  Omphalomesenteric cyst

Fig. 42.2  Umbilical fistula Fig. 42.4  Persistent fibrous cord

42  Omphalomesenteric Duct Remnants
Fig. 42.5  Umbilical polyp
42.3 Meckel’s Diverticulum
Originally described in 1809 by the German
anatomist, Johann Friedrich Meckel (1781–
1833), it is a true diverticulum composed of all
three layers of the intestinal tract. Frequently, it
contains heterotopic gastric, pancreatic and less
commonly duodenal, colonic, or biliary mucosa.
Meckel’s diverticulum is the most common of
the omphalomesenteric duct anomalies encoun-
tered in clinical practice with a 0.3–2.9% preva-
lence (Zani et  al. 2008; Hansen and Soreide
2018). It results from a patency of the intestinal
segment of the duct, with or without a fibrous
obliteration of the distal tract at the umbilicus.
According to a recent systematic review by
Hansen et  al. (Hansen and Soreide 2018), the
lesion is often located some 52 cm from the IC
valve (range 7 cm–200 cm) on the antimesenteric
border of the ileum. The diverticulum is nour-
ished by a rich blood supply from the vitelloin-
testinal vessels that lie within a fold of the gut
mesentery.
545
42.4 Clinical Presentation
Regarding prevalence, Meckel’s diverticulum is
rare and often clinically silent. Symptomatic
Meckel’s diverticulum accounts for only 4–16%
of all index cases (Zani et  al. 2008; Park et  al.
2005). It was estimated in two large published
series that the lifetime probability of a Meckel’s
diverticulum becoming symptomatic is 4.2–
6.4%. More than 75% of symptomatic MD occur
in children younger than 10 years of age with a
median age of some 3.5  years old (Keese et  al.
2019). A lifetime risk of developing
complication(s) from a Meckel’s diverticulum is
6.4% (possibly higher in males vs females). The
clinical presentation usually results from compli-
cations arising from the presence of the diverticu-
lum, which include the following:
Lower gastrointestinal haemorrhage is the
most common presentation in 25–56% of symp-
tomatic cases (Keese et al. 2019; Menezes et al.
2008; Lohsiriwat et  al. 2014) secondary to
active bleeding resultant from peptic ulceration
due to ectopic gastric mucosa. The incidence of
gastric mucosa is estimated at 16–24% in
asymptomatic case(s) and 24–71% in symptom-
atic Meckel’s diverticulum associated with
haemorrhage. Ectopic pancreatic tissue is found
in some 12% of Meckel’s diverticulum speci-
mens (Hansen and Soreide 2018). The ulcer
may be located in the diverticulum or adjacent
ileum. Bleeding is usually profuse and painless,
manifesting as bright red bloody stools with
clots and physiological instability hypovolae-
mic shock. Rarely, bleeding may be occult in
nature for weeks–months with patients featur-
ing a chronic anaemia.
Abdominal pain may occur secondary to
intestinal obstruction and diverticulitis.
Intestinal obstruction is the second common-
est complication accounting for 14–46% of
symptomatic cases (Hansen and Soreide 2018;
Keese et al. 2019; Menezes et al. 2008). It is more
often encountered in older children. Symptoms
include bile- stained vomiting, abdominal disten-
sion and colicky pain. Findings at operation may
546
reveal intussusception, volvulus, or an internal
visceral hernia from persistent internal attach-
ment of the diverticulum to the umbilicus.
Diverticulitis is more commonly encountered
in the adult surgical population compared to chil-
dren (29% vs 20%) (Hansen and Soreide 2018),
which may manifest as abdominal pain, fever and
vomiting. Clinical presentation may be indistin-
guishable from that of acute appendicitis.
42.5 Investigations and Diagnosis
Imaging studies to aid definitive diagnosis should
be tailored according to the varied clinical
presentation.
Haemorrhage is the major complication of
Meckel’s diverticulum in paediatric population.
Technetium-99  m pertechnetate scintigraphy
(Fig. 42.6) is commonly employed to aid diagno-
sis. The gastric mucosa accumulates and secretes
the pertechnetate isotope. After intravenous
injection, a focal area of increased isotope activ-
ity is often apparent in the right lower abdominal
quadrant within some 30  minutes. However,
visualization may take up to1 hour if there are
Fig. 42.6 Tc pertechnetate isotope scan showing
increaseds uptake in a Meckel‘’s diverticulum (M) bear-
ing gastric mucosa. Note the normal uptake in gastric
mucosa (G) and excretion through the urinary bladder
(UB)
A. Boonthai et al.
smaller amounts of heterotopic tissue. The sensi-
tivity of Tc-99  m pertechnetate scintigraphy is
estimated at approximately 85% with a specific-
ity of 95%. These figures may decline with
increasing age as seen in the poorer yield of posi-
tive studies in the adult surgical population.
False-positive studies may occur when there is a
gastric or small intestinal cystic or tubular
duplication(s). False-negative reports are also
seen with Meckel’s diverticula that do not con-
tain adequate amounts of heterotopic gastric
mucosa to sufficiently concentrate the Tc-99  m
isotope. A recent study from Canada (Vali et al.
2015) found 30% positivity on the second
‘repeated’ Tc-99  m scan after a first negative
study. The study authors encourage better prepa-
ration of patients before repeating the isotope
study to enhance the scan result. However, mini-
mally invasive surgery is an option here to deploy
or utilize an umbilical scar crease incision to
access the abdomen in clinically suspicious cases
with a negative scan.
Intestinal obstruction is readily diagnosed
from plain film radiography. Ultrasonography
may be of value in the further evaluation of chil-
dren with a suspected Meckel’s diverticulum and
a negative Tc-99  m pertechnetate scintigraphy
scan. Diverticula appear as round or tubular
‘cyst-like’ structures. Echogenic foci in the
lumen of the diverticulum may represent entero-
liths or inflammatory debris. Colour Doppler
may also demonstrate anomalous vessels. CT
scanning has been used in some centres. A recent
report by Almadi et  al. (Almadi and Aljohani
2020) interestingly showed complicated Meckel’s
diverticulum presenting with intestinal obstruc-
tion with Schistosoma parasites harbouring
within the diverticulum.
42.6 Differential Diagnosis
Differential diagnosis includes appendicitis,
bleeding peptic ulcer disease, inflammatory
bowel disorders, or pelvic inflammatory disease,
especially in teenage girls.
42  Omphalomesenteric Duct Remnants 547

42.7 Management a
The definitive treatment for symptomatic
Meckel’s diverticulum (Fig. 42.7) is open or lap-
aroscopic surgical resection. This can be achieved
either by diverticulectomy (Fig. 42.8a and b) or
by a segmental-limited small bowel resection and
then intestinal anastomosis (Fig. 42.9).
There has been ongoing debate about the mer-
its of excision of Meckel’s diverticulum when it
is incidentally found at operation. A review from
the Mayo Clinic, USA, in 2005, noted four fea-
tures commonly associated with a symptomatic
Meckel’s (1) age < 50 years, (2) male sex, (3) a
diverticulum exceeding more than 2 cm in length b
and (4) the presence of heterotopic tissue.
A subsequent review in 2018 by Slivova et al.
(2018) and then Sinopidis et al. (2019) also found
that the width of the Meckel’s diverticulum base
is thought to be a significant predictor for the
presence of heterotopic mucosal tissue(s) and
here recommend removing all incidental
Meckel’s diverticulum that also have any of the
four features listed above (Park et  al. 2005). It
must be acknowledged, however, that during any
operation, it is not always possible to clearly
determine by inspection or palpation whether an
incidentally found Meckel’s diverticulum is at an
increased risk of complications or not.
Fig. 42.8  Meckel’s Diverticulectomy (a and b)

Fig. 42.9  Resection of a Meckel’s diverticulum

Fig. 42.7  Operative appearance of a Meckel’s diverticulum
548 A. Boonthai et al.

A number of studies have advocated the resec- 42.8 Morbidity

tion of ‘all’ incidental Meckel’s diverticulum
based on the conclusion that incidental Meckel’s Early postoperative complications are estimated in
diverticulectomy is not associated with added around 10% patients following Meckel’s resection,
morbidity and mortality (Arnold and Pellicane including anastomotic leak, stricture, adhesions and
1997; Bani-Hani and Shatnawi 2004). Morbidity postoperative ileus. Late events may include intesti-
and mortality after operation for complicated nal obstruction from adhesions. Whilst enthusiasm
diverticulitis was seen in up to 12% of cases, for minimally invasive surgery in children has
which is some 10% higher compared to inciden- undoubtedly greatly advanced a large US national
tal diverticulectomy (Cullen et al. 1994). A sys- database study (Ezekian et  al. 2019) claimed to
tematic review study in 2008 did not find report no differences in outcome metrics in patients
compelling evidence to support routine excision having ‘open operation’ vs MIS for Meckel’s. A
of an asymptomatic Meckel’s diverticulum as study by Ruscher et  al. (2011), however, had a
incidental ‘unplanned’ diverticulectomy was shorter hospital stay in those having MIS resection.
noted to have higher postoperative complica-
tions. Moreover, it was estimated by the study
authors that some 758 patients would require 42.9 Umbilico-Ileal Fistula
incidentally detected Meckel’s diverticulum to be (Fig. 42.10)
resected to prevent a single fatality (Zani et  al.
2008). A persistent fistula here usually presents in the
Of interest, several neoplasms have been newborn period, with discharge of intestinal con-
reported arising in Meckel’s diverticulum, tent from the umbilicus with periumbilical exco-
which have clearly not received much attention riation (Fig. 42.10). Investigations (if indicated)
in the paediatric surgical literature. These may include a contrast fistulogram (or ultra-
lesions include notably carcinoid tumours, sound) to confirm aberrant anatomy. Management
GIST and signet ring cell intestinal adenocarci- includes resection via a cosmetic umbilical skin
nomas. Amongst the tumours arising from crease incision, identification of the fistula tract
Meckel’s diverticulum, GIST tumours are the and intestinal resection with anastomosis.
most common lesions, accounting for some
41% of cases with gut adenocarcinoma(s) the
least common (Kabir et al. 2019). Sakpal et al.
(2009) reported Krukenberg tumour metastases
arising from a Meckel’s diverticulum signet
ring cell adenocarcinoma in a 56-year-old
female with a family kindred cancer history.
Berry et  al. (2017) recorded another 48-year-
old female with a Meckel’s diverticulum har-
bouring GIST. Mora-­Guzman et al. (2018) also
described neuroendocrine tumours found in
three patients with Meckel’s diverticulum
(Sakpal et al. 2009).
Overall, although tumours arising within
Meckel’s diverticulum are extremely rare, the
benefits of incidental resection in ‘asymptomatic’
patients, with a strong cancer family history(s),
may merit some special consideration (Berry Fig. 42.10  Umbilical discharge from a patent vitelloin-
et al. 2017; Mora-Guzman et al. 2018). testinal tract. Note the excoriation of the skin
42  Omphalomesenteric Duct Remnants 549

a b

Fig. 42.11 (a) Prolapse of a patent’s vitellointestinal tract with the classic ‘‘double- horn’’ anomaly; (b) shows new-
born infant with a ‘stoma-like’ ‘prolapsed vitello-intestinal mucosa’

Prolapse of a large patent vitellointestinal tract 42.12 Persistent Fibrous Cord

at the umbilicus presents as a characteristic
‘double-­horn’ deformity with intestinal lumen
clearly evident in the anomaly (Fig. 42.11a and b).
42.10 Umbilical Sinus
Congenital fibrous bands are clinically signifi-
cant as they may lead to intestinal obstruction
or gut volvulus and must be resected when
symptomatic.

An umbilical sinus usually presents with a persis- 42.13 Umbilical Polyp

tent serous or serosanguinous discharge from the
navel area. When there is doubt about the nature
of the discharge (often because of its intermittent
nature), a radiology sinogram study may demon-
strate the extent of the tract before formal
resection.
42.11 Umbilical Cyst
(Omphalomesenteric Cyst or
Vitelline Cyst)
The cyst with the fibrous cord at either end of the
tract can present with features of intestinal
obstruction and is best managed by exploration
and resection.
Polyps related to the presence of an OMD rem-
nant can occasionally present as a red nodular
lesion in the umbilical dimple. They may contain
tiny fragments of intestinal or gastric mucosa.
One should also consider the more common
diagnosis notably umbilical granuloma.
Exploration of umbilical polyp lesions to exclude
underlying omphalomesenteric duct anomalies is
controversial. According to a recent useful study
by Pacilli et al. (2007), exploration in these cir-
cumstances revealed 46% negative findings, i.e.
no connection with the gastrointestinal intestinal
tract, whilst another 54% of cases did not develop
clinical symptoms after only simple polyp exci-
sion without exploring the abdominal cavity.
550
42.14 Conclusion
Omphalomesenteric duct remnants are fascinat-
ing developmental lesions that may present vari-
antly at any age range though most notably will
be encountered and managed by paediatric sur-
geons. It is therefore key that surgeons are wholly
familiar with care pathways to expedite success-
ful management and outcomes.
References
Almadi F, Aljohani E (2020) Rare cause of complicated
Meckel's with Schistosoma infection: an unusual
cause of acute intestinal obstruction in adults. Int J
Surg Case Rep 66:257–259
Arnold JF, Pellicane JV (1997) Meckel's diverticulum: a
ten-year experience. Am Surg 63(4):354–355
Bani-Hani KE, Shatnawi NJ (2004) Meckel's diverticu-
lum: comparison of incidental and symptomatic cases.
World J Surg 28(9):917–920
Berry AC, Nakshabendi R, Kanar O, Hamer S (2017)
Meckel diverticulum harboring a rare gastrointestinal
stromal tumor. Ochsner J 17(1):121–123
Cullen JJ, Kelly KA, Moir CR, Hodge DO, Zinsmeister
AR, Melton LJ 3rd. (1994) Surgical management of
Meckel's diverticulum. An epidemiologic population-­
based study. Ann Surg 220(4):564–568
Ezekian B, Leraas HJ, Englum BR, Gilmore BF, Reed C,
Fitzgerald TN et al (2019) Outcomes of laparoscopic
resection of Meckel's diverticulum are equivalent to
open laparotomy. J Pediatr Surg 54(3):507–510
Hansen CC, Soreide K (2018) Systematic review of the
epidemiology, presentation, and management of
Meckel's diverticulum in the 21st century. Medicine
(Baltimore) 97(35):e12154
Kabir SA, Raza SA, Kabir SI (2019) Malignant neo-
plasms of Meckel's diverticulum: an evidence based
review. Ann Med Surg (Lond) 43:75–81
Keese D, Rolle U, Gfroerer S, Fiegel H (2019)
Symptomatic Meckel's diverticulum in pediatric
A. Boonthai et al.
patients  - case reports and systematic review of the
literature. Front Pediatr 7:267
Lohsiriwat V, Sirivech T, Laohapensang M, Pongpaibul
A (2014) Comparative study on the characteristics of
Meckel's diverticulum removal from asymptomatic
and symptomatic patients: 18-year experience from
Thailand's largest university hospital. J Med Assoc
Thail 97(5):506–512
Menezes M, Tareen F, Saeed A, Khan N, Puri P (2008)
Symptomatic Meckel's diverticulum in children: a
16-year review. Pediatr Surg Int 24(5):575–577
Mora-Guzman I, Munoz de Nova JL, Martin-Perez E
(2018) Neuroendocrine tumours within a Meckel's
diverticulum. Ann R Coll Surg Engl 100(1):e10
Pacilli M, Sebire NJ, Maritsi D, Kiely EM, Drake DP,
Curry JI et  al (2007) Umbilical polyp in infants and
children. Eur J Pediatr Surg 17(6):397–399
Park JJ, Wolff BG, Tollefson MK, Walsh EE, Larson DR
(2005) Meckel diverticulum: the Mayo Clinic expe-
rience with 1476 patients (1950-2002). Ann Surg
241(3):529–533
Ruscher KA, Fisher JN, Hughes CD, Neff S, Lerer TJ,
Hight DW et al (2011) National trends in the surgical
management of Meckel's diverticulum. J Pediatr Surg
46(5):893–896
Sakpal SV, Babel N, Pulinthanathu R, Denehy TR,
Chamberlain RS (2009) Krukenberg tumor metastasis
of Meckel's diverticulum adenocarcinoma to the ova-
ries. J Nippon Med Sch 76(2):96–102
Sinopidis X, Fouzas S, Kambouri K, Panagidis A,
Alexopoulos V, Karatza A et  al (2019) Predictive
model of heterotopy in Meckel's diverticulum in chil-
dren. ANZ J Surg 89(6):e241–e2e5
Slivova I, Vavrova Z, Tomaskova H, Okantey O, Penka
I, Ihnat P (2018) Meckel's diverticulum in children -
parameters predicting the presence of gastric heteroto-
pia. World J Surg 42(11):3779–3784
Vali R, Daneman A, McQuattie S, Shammas A (2015)
The value of repeat scintigraphy in patients with a
high clinical suspicion for Meckel diverticulum after a
negative or equivocal first Meckel scan. Pediatr Radiol
45(10):1506–1514
Zani A, Eaton S, Rees CM, Pierro A (2008) Incidentally
detected Meckel diverticulum: to resect or not to
resect? Ann Surg 247(2):276–281
 Omphalocele and Gastroschisis
43
Leah M. Sieren, Duane S. Duke,
and Marshall Z. Schwartz

43.1 Introduction 43.2 Differential Diagnosis: Types

of Anterior Abdominal Wall
Gastroschisis and omphalocele represent two dis- Defects
tinct congenital abnormalities of the anterior
abdominal wall. Dudrick’s development of total Congenital abdominal wall defects include gas-
parenteral nutrition in the late 1960s, and troschisis, omphalocele, hernia of the umbilical
Schuster’s successful application of extraabdom- cord, cloacal exstrophy (infraumbilical omphalo-
inal housing (silo) for eviscerated bowel in 1967, cele, exstrophy of the bladder), and Cantrell’s
provided surgeons with much needed tools to Pentalogy (supraumbilical omphalocele, sternal
enhance the treatment and improve the survival cleft, intracardiac anomaly, diaphragmatic her-
of infants with gastroschisis and omphalocele. nia, and pericardial defect). This chapter will spe-
Advancements in prenatal ultrasonography, neo- cifically address gastroschisis and omphalocele
natal monitoring, and ventilator management (Table 43.1).
continue to improve outcomes for critically ill
infants with abdominal wall defects. Table 43.1  Clinical features: Gastroschisis and ompha-
locele contrasted
Gastroschisis Omphalocele
Location To the right of Umbilical ring
intact umbilical
cord
Defect size Small: 2–4 cm Large: 2–10 cm
Covering sac Absent Present
L. M. Sieren Intestinal Nonrotation Nonrotation
Wake Forest Baptist Health, rotation
Winston-Salem, NC, USA GI motility Adynamic ileus Normal
e-mail: lsieren@wakehealth.edu Associated Intestinal atresia Common and often
D. S. Duke anomalies (10–15%) major
Children’s Hospital of the King’s Daughter, Associated None Beckwith-­Wiedemann
Norfolk, VA, USA syndromes
Trisomy 13–15, 16–18, 21
M. Z. Schwartz (*)
Lower midline syndrome
Department of Surgery and Urology, Wake Forest
Cantrell’s pentalogy
University School of Medicine, Wake Forest Institute
for Regenerative Medicine, Wake Forest Baptist Survival 90% 30–85% largely
Health, Winston-Salem, NC, USA dependent on associated
e-mail: mzschwar@wakehealth.edu anomalies

© Springer Nature Switzerland AG 2023 551

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_43
552 L. M. Sieren et al.

43.3 Gastroschisis

Gastroschisis (Gr. gaster-, belly + − schisis, fis-

sure) is a congenital defect of the anterior
abdominal wall most often to the right of the
umbilicus. The abdominal wall defect typically
ranges in size from 2 to 4  cm, notably smaller
than the defect often encountered with omphalo-
cele. All patients with gastroschisis have intesti-
nal nonrotation. The stomach, small and large
intestines, and occasionally the gonads, can be
found outside of the abdominal wall (Figs. 43.1
and 43.2). The liver is almost never eviscerated.
There is no peritoneal sac over the herniated
contents. Therefore, the viscera is exposed to
amniotic fluid (pH 7.0), which later in pregnancy
becomes irritating and results in a thick fibrotic
inflammatory peel over the eviscerated bowel.
Another proposed cause of the thickened,
inflamed bowel is the constriction of the abdomi-
nal wall defect late in gestation leading to venous
congestion. Although the etiology is unclear,
Fig. 43.2  The eviscerated bowel typically shows evi-
patients with gastroschisis sustain intrauterine
dence of a fibrin “peel,” bowel wall thickening, bowel
growth retardation. The typical birth weight for dilatation, and mesenteric foreshortening
these infants is between 2000 and 2500  g.
Associated anomalies are uncommon; intestinal
atresia can complicate gastroschisis in 10–15% 43.3.1 Etiology and Incidence
of cases. With modern neonatal intensive care,
nutritional supplementation, and appropriate The abdominal wall forms from infolding of the
surgical care, overall survival is excellent and cranial, caudal, and two lateral embryonic folds.
surpasses 90%. The rapid growth of the intestine at this time
causes it to migrate external to the forming
abdominal wall. By the 10–12th gestational
week, the abdominal wall is well-formed and the
intestinal tract has returned to its intraperitoneal
location with normal rotation and fixation.
The etiology of gastroschisis has not been
fully elicited. One hypothesis suggests that the
abdominal wall defect results from an ischemic
event during the formation of the abdominal wall.
The right paraumbilical area is vulnerable
because it is supplied by the right umbilical vein
and right omphalomesenteric artery until they
involute. If their development and/or involution
is disturbed, ischemia of the right paraumbilical
Fig. 43.1  Gastroschisis defect with eviscerated bowel abdominal wall may result, leading to an abdomi-
43  Omphalocele and Gastroschisis
nal wall defect. An alternative hypothesis sug-
gests that gastroschisis results from an intrauterine
rupture of an omphalocele, with subsequent
resorption of the omphalocele sac. This hypothe-
sis is less likely because of the other clear differ-
ences between these two abdominal wall defects,
such as the position of the defect and the marked
increase in associated anomalies in infants with
omphalocele.
The incidence of gastroschisis worldwide is
increasing. Presently, gastroschisis occurs in
roughly 0.4–3 per 10,000 live births. Gastroschisis
is commonly seen with young mothers, and 25%
of cases involve teenage mothers. Maternal expo-
sure to cigarette smoking, illicit drugs, vasoactive
over-the-counter drugs (pseudoephedrine, aspi-
rin, acetaminophen), and environmental toxins
have been linked to gastroschisis.
43.3.2 Associated Anomalies
Associated anomalies are uncommon with gas-
troschisis. Intestinal atresia/stenosis is present in
10–15% of cases. The areas of atresia/stenosis are
attributed to intrauterine volvulus, intussuscep-
tion, or secondary to the bowel’s blood supply
being impeded by a narrow abdominal wall defect.
Bowel perforation occurs in 5% of patients. Other
less common anomalies include: undescended tes-
tes, hypoplastic gallbladder, hydro-nephrosis,
Meckel’s diverticulum, and intestinal duplications.
The incidence of chromosomal abnormalities is
not higher than the normal population.
43.3.3 Prenatal Diagnosis
Abdominal wall defects can be detected with
abdominal ultrasonography as early as the
10–12th week of gestation. This modality is lim-
ited by timing and objective of ultrasonography,
fetal position, and experience of the sonographer.
Ultrasound diagnosis has a specificity of 95%
and a sensitivity of 60–75% for identifying
abdominal wall defects. The prenatal differential
diagnosis based on ultrasound includes bladder
exstrophy, cystic cord lesion, urachal cysts, as
553
well as omphalocele and gastroschisis. Ultraso-
nography cannot reliably detect whether gastros-
chisis is complicated by areas of intestinal atresia.
However, serial ultrasonography, especially in
the third trimester, can detect increasing bowel
diameter and thickening suggestive of vascular
obstruction. These findings have been used as an
indication for urgent delivery to avoid midgut
infarction, a devastating complication in gastros-
chisis patients.
Maternal serum levels of alpha fetoprotein
(AFP) can be elevated up to nine times the mean
in gastroschisis. Elevation of AFP is less dra-
matic in omphalocele (3–4 times the mean).
When diagnosed prenatally, gastroschisis
patients should be referred to a high-risk obstet-
ric unit with neonatal intensive care and pediatric
surgery capabilities. Serial ultrasonography
should be performed looking for changes in the
size and thickness of the bowel, as well as the
diameter of the abdominal wall defect. Significant
bowel wall thickening and bowel dilatation
accompanied by a decrease in defect diameter
may be indications for earlier delivery to avoid
bowel infarction. When compared to normal vag-
inal delivery, preterm elective delivery and term
cesarean section delivery have not been shown to
improve outcomes. Vaginal delivery is an accept-
able birth route for gastroschisis patients. The
survival rate for babies with gastroschisis is uni-
formly greater than 90%.
43.3.4 Perinatal Care
As a result of the absence of a hernia sac, the
exposure of the eviscerated contents to fluid, and
evaporative heat loss, infants with gastroschisis
will have profound fluid and temperature regula-
tion difficulties upon delivery. After addressing
any airway, breathing, or circulation issues,
immediate placement of the lower half of the
infant into a sterile bowel bag will ameliorate the
fluid and heat losses. The bowel bag should
extend from the axilla to the feet and must include
the herniated viscera. A nasogastric tube should
be placed to facilitate the decompression of the
gastrointestinal tract.
554
Empiric parenteral antibiotics with broad
spectrum coverage are administered. Ampicillin
100 mg/kg/day and gentamicin 7.5 mg/kg/day in
combination are commonly used. Maintenance
of normothermia is critical to facilitate an opti-
mum outcome.
Fluid resuscitation should be guided by hemo-
dynamic status, urine output, tissue perfusion,
and correction of metabolic acidosis (if present).
An initial fluid bolus of 20 ml/kg with 10% dex-
trose in lactated Ringer’s solution should be
administered, followed by 125–150% mainte-
nance intravenous fluid.
Once adequate urine output is obtained and
any acidosis is corrected, the newborn can
undergo either primary closure of the abdominal
defect or temporary silo placement as described
below.
43.3.5 Operative Management
Primary closure should be attempted whenever
feasible. The abdominal wall defect can be
enlarged 1–2 cm cephalad if needed to facilitate
reduction. Caudal extension of the abdominal
wall defect is not recommended due to the risk of
injuring the bladder. Peak inspiratory pressures
greater than 25 mmHg, increases in central venous
pressure of 4 mmHg or more, or elevated bladder
pressures greater than 20  mmHg upon fascial
apposition should steer the surgeon away from
performing primary closure. The consequences of
creating an abdominal compartment syndrome
with an ambitious primary closure may be cata-
strophic—there should be a low threshold to con-
vert to silo placement with delayed closure.
The temptation to debride the fibrotic mem-
branous peel or to search for areas of intestinal
atresia should be avoided. These maneuvers
only serve to prolong the procedure and
increase the likelihood of iatrogenic bowel
injury. Additionally, some pediatric surgeons
have employed abdominal wall stretching and
efforts to manually decompress the luminal con-
tents from the thickened, dilated intestine in an
effort to accomplish primary closure. It is the
view of the authors that these maneuvers are not
L. M. Sieren et al.
Fig. 43.3  A preformed spring loaded silo bag in place in
an infant with gastroschisis
safe and the risks outweigh the benefits. A gentle
finger-­sweep should be performed along the pos-
terior aspect of the abdominal wall circumferen-
tially around the defect in order to ensure that any
congenital bands are divided.
Preformed spring-loaded silos are universally
used for visceral reduction and delayed fascial
closure when primary reduction and closure are
not possible (Figs.  43.3, 43.4, and 43.5). These
silos can often be placed in the NICU without
anesthesia. The goal is to progressively reduce
the eviscerated bowel over the next 3–7  days.
Daily reductions of the silo contents are usually
possible until the viscera is flush with the abdom-
inal wall. Once the bowel has been completely
reduced into the peritoneal cavity, the silo is
removed and primary fascial/skin closure is per-
formed (Fig. 43.6a, b).
The abdominal silo can also be constructed
from rectangular-shaped Dacron reinforced
Silastic sheeting. The silastic pieces are attached
to each side of the abdominal wall fascia with
interrupted mattress sutures of 3–0 or 2–0 silk
(Fig. 43.7a, b). The top can be narrowed at each
side thus creating a trapezoid with a wide base.
This modification will protect the bowel from
potential strangulation from a narrow silo base
during reduction. The apex of the silo is usually
closed with a running monofilament suture, but
other techniques have been employed (Fig. 43.8).
The incidence of an umbilical defect follow-
ing gastroschisis closure is high. To avoid this, it
43  Omphalocele and Gastroschisis
is important to release the subcutaneous tissue off
the fascia circumferentially to ensure that good,
strong fascia is approximated and no residual
hernia exists. This includes separating the base of
the umbilical cord off the underlying fascia to
clearly identify the fascia lateral to the umbilical
Fig. 43.4  The viscera are progressively reduced using
gentle pressure from the apex of the silo. Gradual reduc-
tion can be carried out in the NICU with minimal to no
anesthesia
a b
Fig. 43.6  (a) An infant status post gastroschisis closure. (b) Close up of a recent primary closure of a gastroschisis
555
ring for inclusion in the closure. The fascial
defect may then be closed transversely or longi-
tudinally, depending on how the edges best come
together. The choice of suture material and the
technique for placement of sutures, whether
interrupted, figure-of-eight, or running, is per-
sonal preference. We prefer interrupted figure-
of-­eight sutures of 2–0 or 3–0 absorbable suture
(vicryl or PDS). It is preferable to place all of
the sutures prior to tying them to ensure that the
underlying viscera is not injured. A thin ribbon
retractor placed under the fascia is advantageous
while tying the sutures in sequence to avoid
trapping the bowel during tying of the sutures. It
is important at the time of fascial closure to
monitor airway pressure, as a drastic increase
could indicate abdominal compartment syn-
Fig. 43.5  Once the bowel has been completely reduced
into the peritoneal cavity, the infant is returned to the OR
where the silo is removed and the fascia and skin are
closed primarily
556
a b
Fig. 43.7 (a) In infants who cannot undergo complete
reduction of herniated intestinal contents, an abdominal
wall silo is created by suturing a Dacron reinforced
Silastic sheet to the fascial edges using horizontal mat-
tress, 2–0 silk sutures placed through the fascia to anchor
Fig. 43.8  A completed abdominal wall “silo.” The single
monofilament suture is used to prevent the skin from
retracting
drome. Postoperatively, urine output and lower
extremity perfusion should be monitored for the
same concern.
Once the fascia is closed, any ischemic skin
should be trimmed. The skin is then closed using
5–0 monofilament absorbable suture. Creating a
L. M. Sieren et al.
the Silastic mesh. (b) Close-up view of Silastic mesh
anchored to the fascia and also showing several suture
lines in the mesh indicating serial reductions of the herni-
ated contents
purse string stitch circumferentially along the
skin edge, including the skin lateral to the umbili-
cal stalk, creates a natural-appearing umbilicus
upon closure with the umbilical stalk in the cen-
ter of the skin closure. The stalk will dessicate
and fall off naturally, leaving an esthetically-­
pleasing umbilicus.
In 2004, Sandler et  al. described a sutureless
closure technique for gastroschisis where the rem-
nant umbilical cord is used to cover the umbilical
defect after visceral reduction. A dressing is
applied over the site and left in place for two
weeks. After the dressing is removed, an estheti-
cally normal umbilicus remains. This technique
can be performed at the bedside without the need
for anesthesia. If the fascia does not close sponta-
neously over the subsequent months/years, then
the residual umbilical hernia should be repaired
per usual practice when the child is older.
Gastric decompression and parenteral nutri-
tion should be continued until return of bowel
function. If an ileus persists beyond 3–4 weeks,
contrast studies of the gastrointestinal tract
should be considered to rule out mechanical
obstruction due to atresia, stenosis, or adhesive
bands.
Complicated gastroschisis are cases with an
associated atresia, necrosis, perforation, or volvu-
lus. These account for 10–15% of live-born cases.
Surgical care in these instances is customized to the
needs of the patient taking into consideration the
severity of the bowel complication. Atresias repre-
43  Omphalocele and Gastroschisis
sent up to 30% of complicated gastroschisis cases.
Simple atresias can often be managed with gastro-
intestinal low tube sunction decompression.
Abdominal exploration 4–6 weeks to correct the
atresia can be performed following initial abdomi-
nal wall closure once the bowel edema and inflam-
matory peel have resolved. Recent reports have
described the management of atresias as early as
2–3 weeks after abdominal wall closure, or imme-
diately at the initial operative intervention follow-
ing birth. However, a higher complication rate can
occur. Necrotic bowel and perforations require
immediate management with resection and imme-
diate or delayed anastomosis, temporary ostomy,
or tube enterostomy to control the perforation. In
cases of vanishing gastroschisis where the herni-
ated contents infarct in utero secondary to ischemia
caused by a closing defect, short bowel syndrome
oftentimes results and the insertion of a gastros-
tomy tube should be considered for initial bowel
decompression and subsequent enteral feeding.
Patients with gastroschisis should be monitored
closely for necrotizing enterocolitis, as the inci-
dence of n­ecrotizing enterocolitis is reported as
high as 8–9% following abdominal wall closure.
43.3.6 Complications
All gastroschisis patients will have some degree
of postoperative ileus. While awaiting the return
of bowel function, gastroschisis patients will
a b
Fig. 43.9 (a) A typical omphalocele protruding through
a large central defect of the umbilical ring. The omphalo-
cele contains the liver (seen) and small intestine. (b) An
557
require parenteral nutrition via central venous
access, thus subjecting them to the possible risk of
central line-associated blood stream infection.
Ten percent of patients with gastroschisis will
have persistent bowel hypoperistalsis and some
infants will require long-term parenteral nutrition.
Gastroschisis complicated by intestinal atresia or
perforation can result in short bowel syndrome.
Other postoperative complications include surgi-
cal site infection, sepsis, aspiration pneumonia,
necrotizing enterocolitis, and complications
related to increased abdominal pressure (gastro-
esophageal reflux and inguinal hernia).
Most postnatal deaths are attributed to prema-
turity, overwhelming sepsis, or complications of
short bowel syndrome such as liver failure sec-
ondary to parenteral nutrition induced cholestasis
and hepatic fibrosis.
43.4 Omphalocele
Omphalocele (Gr. omphalos—umbilicus + Gr.
kele—swelling; hernia; syn. exomphalos) is a
congenital anterior abdominal wall defect present
at the umbilical ring. It represents a wide spec-
trum of pathology ranging from an uncompli-
cated hernia of the umbilical cord to a large
defect with the extrusion of all abdominal viscera
(giant omphalocele) and life-threatening associ-
ated anomalies (Fig.  43.9a, b). The abdominal
wall defect can range in size from 2–10  cm.
omphalocele sac containing several loops of intestine. An
incision has been made at the junction of the omphalocele
sac and skin
558
Virtually all intraabdominal structures can be
eviscerated including the liver. The omphalocele
is covered by an outer layer of amnion, a middle
layer of Wharton’s jelly, and an inner layer of
peritoneum. This robust covering protects the
gastrointestinal tract during intrauterine life and
bowel function is usually normal at birth.
Associated anomalies can involve the cardiovas-
cular, alimentary, genitourinary, musculoskeletal,
and nervous systems. The presence of such sig-
nificant associated anomalies suggests that the
omphalocele defect is simply one element of a
global intrauterine developmental disruption.
The severity of the associated anomalies ulti-
mately determines the morbidity and mortality of
infants with omphalocele.
43.4.1 Etiology and Incidence
The etiology of omphalocele is not fully under-
stood. It is believed that an omphalocele results
from failure in the fusion of the three embryonic
folds. The infolding process is a delicate balance
of cell proliferation and cell apoptosis. When the
abdominal wall cranial embryonic infolding is
unsuccessful, epigastric omphalocele results and
can be accompanied by diaphragmatic hernias,
sternal clefts, pericardial defects, and intracardiac
defects as in the Pentalogy of Cantrell (Fig. 43.10a,
b). When the caudal embryonic fold has unsuc-
cessful infolding, a hypogastric omphalocele
a b
Fig. 43.10 (a) A giant omphalocele with ectopia cordis (see b) (b) Omphalocele with accompanying diaphragmatic
hernia. The heart is visible in the cephalad part of the defect through large diaphragmatic and pericardial defects
L. M. Sieren et al.
results and can be accompanied by or cloaca,
vesicointestinal fissure, colon atresia, imperforate
anus, sacral vertebral defects, or meningomyelo-
cele as in the Lower Midline Syndrome. The clas-
sic central omphalocele is thought to result from
an infolding error of the lateral embryonic folds.
Omphalocele is less common than gastros-
chisis with an incidence of 1–2 per 10,000 live
births. Males are more affected than females.
Whereas gastroschisis is seen in young moth-
ers, omphalocele is more frequently associated
with maternal age over 30 years old. No terato-
gens have been linked to the formation of an
omphalocele.
43.4.2 Associated Anomalies
The survival of an infant with an omphalocele
depends more on the severity of its associated
anomalies than the abdominal wall defect itself.
Multiple anomalies are seen in 40–70% of
infants with omphaloceles. Cardiac anomalies
are present in 50% of cases (septal defects,
tetralogy of Fallot, ectopia cordis). Gastrointes-
tinal anomalies can occur (e.g., duodenal atre-
sia). Malignant tumors such as Wilms’ tumor,
neuroblastoma, and adrenocortical tumors have
an increased incidence with omphalocele.
Ambiguous genitalia can also accompany
omphalocele, providing the additional clinical
hurdle of sex determination.
43  Omphalocele and Gastroschisis
Chromosomal anomalies are seen in 30–40%
of omphalocele cases (trisomy 13–15, 16–18, 21,
Turner’s syndrome, Klinefelter’s syndrome, and
triploidy). Omphalocele is part of the Beckwith-­
Wiedemann chromosomal syndrome (umbilical
defect, gigantism, macroglossia, visceromegaly,
and pancreatic islet cell hyperplasia). Chromo-
somal anomalies are seen more frequently when
there is a small defect and no liver is present in
the omphalocele hernia. There is a reported
higher incidence of Wilms’ tumor associated
with Beckwith-Weidmann Syndrome.
43.4.3 Prenatal Diagnosis
Abdominal viscera in the base of the umbilical
cord can be visualized with ultrasonography as
early as the tenth gestational week. Ultrasound
can also determine if the liver has herniated out-
side the abdominal wall. Maternal alpha fetopro-
tein is elevated 3–4 times the normal mean in
omphalocele. If omphalocele is detected, amnio-
centesis with chromosomal analysis should be
offered to screen for lethal anomalies.
For small omphaloceles without lethal associ-
ated anomalies, serial ultrasonography should be
continued and term vaginal delivery pursued.
There are no absolute indications for cesarean
section, however for large omphaloceles with
eviscerated liver, most centers will recommend
term cesarean section to avoid liver injury with
hemorrhage upon parturition.
The main advantage of prenatal ultrasonogra-
phy and chromosomal analysis is that it allows
for more informed prenatal counseling. If severe
or lethal anomalies are detected, the option of ter-
minating the pregnancy can be discussed with the
parents.
43.4.4 Perinatal Care
After addressing any airway, breathing, or circu-
lation issues, the assessment of the neonate with
omphalocele should include a complete physical
examination, chest radiograph, and an echocar-
diogram of the heart and ultrasonography of the
559
kidneys to identify associated anomalies.
Baseline laboratory values should be obtained
with attention to immediate serum glucose analy-
sis. Omphalocele complicated by Beckwith-­
Wiedemann syndrome can present with profound
hypoglycemia resulting in brain injury and men-
tal retardation if not rapidly corrected. If not
already established, karyotype analysis also
should be performed. A nasogastric tube should
be inserted to decompress the gastrointestinal
tract. Empirical broad-spectrum antibiotics
should be initiated as with gastroschisis.
Intravenous fluids using 10% dextrose and 0.25–
0.50% normal saline solution should be run at
125% maintenance rate. Potassium chloride
should be added once the infant has voided to
avoid hypokalemia. A non-adherent plastic sheet
may be placed over the omphalocele sac to pro-
tect the sac from inadvertent tear during patient
transport. The neonate should be maintained in a
thermoneutral environment while in transport or
awaiting further care.
A rupture of the sac occurs in 10% of infants
with omphalocele. If the sac has a minor tear, it
may be carefully sutured closed at the bedside.
Fluid resuscitation for a neonate with a ruptured
omphalocele sac should be approached as
described above for gastroschisis.
43.4.5 Operative Management
The operative strategy for the repair of omphalo-
cele depends on the size of the defect, the extent
of visceroabdominal disproportion (VAD), and
the severity of the associated anomalies.
Primary closure of the omphalocele can be
accomplished 60–70% of the time (generally
2–5 cm in diameter). In large omphaloceles with
the liver herniated, care must be taken in the
reduction of the liver to avoid torsion of the
hepatic veins, obstruction of portal vein inflow,
obstruction of the inferior vena cava, or injury to
the hepatic capsule. Resection of the sac prior to
the reduction of the herniated contents is not nec-
essary, but in small defects is recommended. In
very large defects that require staged reduction,
failure of serial reduction allows the omphalocele
560
sac to still be present. Regardless, if a portion of
the sac overlying the liver is densely adherent, it
should be left in place as attempts at debridement
can cause difficult-to-control liver hemorrhage.
When there is significant VAD and loss of
abdominal domain, the reduction of the viscera
can be approached via a staged silo technique as
described for gastroschisis (Fig.  43.11). The
abdominal wall can then be closed in layers result-
ing in an intact abdominal wall (Fig.  43.12a, b).
Alternatively, the surgeon can create an intentional
ventral hernia by elevating skin flaps for skin-only
Fig. 43.11  Appearance of the Silastic silo after complete
reduction of the herniated contents. Note the six horizon-
tal suture lines indicating the number of sequential reduc-
tions. This sac has been “painted” with Betadine antiseptic
ointment
a b
Fig. 43.12 (a) Infant born with an omphalocele shown
immediately following removal of the Silastic silo and
closure of the fascia and skin. Note the long steri strips
L. M. Sieren et al.
closure or inserting a prosthetic mesh. This
planned hernia can be repaired at 1 year of age.
For neonates with giant omphaloceles
(>8–10  cm), significant pulmonary or cardiac
abnormalities, or significant chromosomal abnor-
malities, a non-operative approach can be
employed by the induction of epithelialization of
the omphalocele sac. Topical eschar forming
agents are painted on the sac twice a day. Silver
sulfadiazine is presently the most commonly
used eschar forming agent in the United States.
When using this agent, sodium levels should be
monitored because this compound is hypotonic
and can cause sodium loss through the sac.
43.4.6 Complications
Early post repair complications include surgical
site infection, sepsis, dehiscence, ventral hernia,
bowel obstruction, bowel perforation, necrotiz-
ing enterocolitis, fistulas, renal failure, and respi-
ratory distress. Iatrogenic injury to the eviscerated
liver can result in hepatic hematoma and hepatic
infarction.
A significant increase in intraabdominal pres-
sure following abdominal wall closure can induce
respiratory distress and require long-term venti-
latory support. The increase in intraabdominal
that are placed to distribute the tension over a wide dis-
tance (b) A well healed omphalocele repair
43  Omphalocele and Gastroschisis
pressure can also lead to abdminal compartment
syndrome.
Antenatal mortality ranges from 30–60% and
largely depends on the severity of associated
anomalies.
43.5 Conclusion
The outcome for patients with gastroschisis has
dramatically improved over the past few decades.
Whereas the mortality in western countries was
approximately 50% five decades ago, the survival
rate is now uniformly greater than 90%. This
improvement in outcome is related to the availabil-
ity of intravenous nutrition and the use of staged
closure when indicated. Late complications and
mortality are related to sepsis from an intraabdom-
inal or wound complication, central line-associ-
ated blood stream infection, or a complex bowel
complication, such as necrosis or volvulus.
The outcome for patients with omphalocele
greatly depends on their associated comorbidi-
ties. Mortality rates in omphalocele patients were
as high as 80% fifty years ago. As with gastros-
chisis, the development of parenteral nutrition
and staged-closure methods has greatly improved
survival rates over the recent decades. Mortality
is mostly related to major other congenital anom-
alies (e.g. cardiac and renal).
Further Reading
Baerg J, Kaban G, Tonita J et al (2003) Gastroschisis: a
sixteen year review. J Pediatr Surg 38:771–774
Baerg J, Munoz A (2019) Long term complications
and outcomes in omphalocele. Sem Pediatr Surg
28:118–121
561
Driver CP, Bruce J, Bianchi A et  al (2000) The con-
temporary outcome of gastroschisis. J Pediatr Surg
35:1719–1723
Emil S (2018) Surgical strategies in complex gastroschi-
sis. Sem Pediatr Surg 27:309–315
Laje P, Fraga MV, Peranteau WH et  al (2018) Complex
gastroschisis: clinical spectrum and neonatal outcomes
at a referral center. J Pediatr Surg 53:1904–1907
Ledbetter DJ (2006) Gastroschisis and omphalocele. Surg
Clin N Am 86:249–260
Puri P, Höllwarth ME (eds) (2006) Pediatric surgery.
Springer, Heidelberg
Sandler A, Lawrence J, Meehan J et al (2004) A “plastic”
sutureless abdominal wall closure in gastroschisis. J
Pediatr Surg 39:738–741
Schlatter M, Norris K, Uitvlugt N et al (2003) Improved
outcomes in the treatment of gastroschisis using a
preformed silo and delayed repair approach. J Pediatr
Surg 38:459–464
Schuster SR (1967) A new method for the staged
repair of large omphaloceles. Surg Gynecol Obstet
125:837–850
Schwartz MZ, Tyson KR, Milliorn K et al (1983) Staged
reduction using a silastic sac is the treatment of choice
for large congenital abdominal wall defects. J Pediatr
Surg 18:713–719
Weber TR, Au-Fliegner M, Downward CD, Fishman SJ
(2002) Abdominal wall defects. Curr Opin Pediatr
14:491–497
Wilson DR, Johnson MP (2004) Congenital abdominal
wall defects: an update. Fetal Diag Ther 19:385–398
Yokomori K, Olikura M, Kitano Y, Hori T, Nakajo T
(1992) Advantages and pitfalls of amnion inversion
repair for the treatment of large unruptured omphalo-
cele: results of 22 cases. J Pediatr Surg 27:882
 Conjoined Twins
Juan A. Tovar and Leopoldo Martinez
44.1 Introduction
Since conjoined twinning is an obvious condition
that often poses obstetric problems, it has been
known from ancient times. Two-faced deities
(Jano) or multiple-headed creatures (Hydra) were
likely introduced into mythology after observa-
tion of such twins. There are pictures and carv-
ings depicting conjoined twins in all cultures
from the European, Asiatic, African, and
American continents (Tovar 2009). Some cases
represented moral dilemmas and prompted risky
definitions (for instance, how many souls they
have) by religious authorities. One of the first
descriptions of esophageal atresia was based on
the autopsy of a set of conjoined twins. This con-
dition became popular because the twins were
often exhibited as freaks or circus attractions.
This was the case of Chang and Eng Bunker, the
original Siamese twins, who were taken for this
purpose from Siam to the United States, where
they eventually settled in North Carolina and
married Sarah and Adelaide, two sisters with
whom they had over 20 children; the two couples
lived for more than 50  years (Kaufman 2004;
Mian et  al. 2017). Many examples have been
J. A. Tovar · L. Martinez (*)
Departamento de Cirugía Pediátrica, Hospital
Universitario La Paz, Madrid, Spain
e-mail: juan.tovar@salud.madrid.org;
leopoldo.martinez@salud.madrid.org
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_44
44
publicized ever since, and they are often reported,
particularly in the current media era.
Due to the complexity of the technical prob-
lems involved, it is understandable that the first
separations were relatively recent (seventeenth
century). However, many unsuccessful separa-
tions were never reported, and the high mortality
of this condition is still hidden in part.
44.2 Etiopathogenesis
and Incidence
Twin pregnancy is not a rare event in humans.
Although the incidence of dizygotic twinning is
widely variable around the world, that of mono-
zygotic twinning is constant, at about 1:286 preg-
nancies. Some factors have been associated with
an increase in the number of this kind of twin-
ning: infertility treatments, pregnancies within
6 months of stopping oral contraceptives or after
assisted reproduction techniques. In an experi-
mental setting, monozygotic twinning has been
reported in the mouse, after using agents interfer-
ing with cell division, such as ethanol, urethan, or
vincristine sulfate (Kaufman 2004).
Conjoined twins are always monochorionic
and monoamniotic (Mian et al. 2017). The only
apparently possible mechanism for conjoined
twinning is an early failure of the division of a
primitive embryonal disk destined to produce
identical, monozygotic, monochorionic, isosex-
563
564
ual twins that share the same genome and have
similar fingerprints. In the human, the division of
the embryonal disc can occur at one of three
stages of development. The earliest stage will
raise a dichorionic and diamniotic placentation,
the middle one a monochorial diamniotic one
and, finally, the later stage (at about 13–14 days
of development), a monochorionic monoamni-
otic variety. In these cases, for unknown reasons,
only partial separation (fission) of the embryos
may occur resulting in conjoined twinning.
R.  Spencer, an acknowledged authority in this
field, pointed out that the twins are always
homologous, in the sense that they never have the
head or the lower limbs on opposite sides
(Spencer 2000). This suggests that the original
mechanism is the missed cleavage of the primi-
tive embryonal plate along the longitudinal axis
(Kaufman 2004).
Some ancient experiments in amphibians and
a few modern molecular genetic observations
suggest that fusion of two originally separated
embryos may be the explanation for some rare
cases, in which there is genetic sex discordance,
and some asymmetric heteropagus that otherwise
has to be interpreted as resulting from a partial
atrophy of one of two originally symmetric joined
individuals.
Each theory, fission or fusion, has its own
strengths and limitations. Unfortunately, a defi-
nite answer is hard to reach due to the lack of
experimental models of conjoined twinning and
the rarity of human cases. In fact, the true inci-
dence, after in utero deaths and stillborn are dis-
carded, is around 1 in 250,000 live births, with an
unclearly explained female predominance of 3:1
(Endres and Wilkins 2005; Martinez-Frias et al.
2009; Mutchinick et al. 2011).
44.3 Classification
There are multiple classifications based on vari-
ous criteria. Spencer proposed to divide them
into ventrally and dorsally joined and to subdi-
vide them within the groups according to the
level of fusion (Pierro et  al. 2015; Boer et  al.
2019). We preferred to adhere to a wider classifi-
J. A. Tovar and L. Martinez
Table 44.1  Classification of conjoined twins
Asymmetric
 fetus in feto
 Acardius acephalus
 Heteropagus
Symmetric
 Craniopagus
 Thoracopagus
 Omphalopagus
 Parapagus
 Rachiopagus
 Ischiopagus
 Pygopagus
cation, which takes into account their asymmetric
or symmetric nature and the anatomic point of
fusion. This is defined in each case by the suffix
pagus, the Greek term for fixed (Table 44.1).
Among the asymmetric twins, we include the
organoid teratomas and fetus in fetu, the acardius
acephalus, and the heteropagus parasitic twins.
The first ones are not included in the family of
conjoined twins and, except for some very
“organoid” ones in which a more or less rudi-
mentary spine is present, they are rather consid-
ered as teratomas. Acardius acephalus are
connected by marginal placental vessels that
allow the healthy twin (the autositus) to maintain
circulation and feeding of the incomplete para-
sitic one, characteristically devoid of a heart and
head. When they come to term, only the autositus
survives after umbilical cord division. However,
nowadays, prenatal diagnosis is possible, and
intrauterine coagulation of the umbilical vessels
of the parasitus is an alternative to postnatal
demise. Heteropagus twins (Fig. 44.1) are gener-
ally attached to the umbilical region of an ana-
tomically normal autositus twin, with or without
exomphalus, as organoid parasitic masses con-
taining various organs and limbs unable to sus-
tain independent circulation by them and are
therefore unviable if separated.
Symmetric conjoined twins may be joined by
the head (craniopagus), the thorax (thoracopa-
gus), the abdomen (omphalopagus), the spine
(rachiopagus), or the caudal pole (ischiopagus and
pygopagus). Occasionally, they are laterally fused
along the body axis (parapagus) (Table  44.1).
Thoracopagus are by far the most frequent type
and account for 60% of all cases (Tovar 2009).
44  Conjoined Twins
Fig. 44.1  Heteropagus parasitic twin joined to the abdo-
men of the autositus. Exomphalos is frequent in this par-
ticular form of conjoined twinning
44.4 Ethical Issues
Conjoined twinning is a unique situation in terms
of ethical dilemmas (Kallberg 2018; Thomas
et al. 2018). The principles that regulate the exer-
cise of our profession are particularly difficult to
respect in these cases:
The principle of autonomy (we shall respect the
decisions of the patient), that in children is usually
exerted by proxy by the parents and only very
rarely by others, may generate conflicts among par-
ents, doctors, or even the courts when unanimity
regarding decisions is not reached. When unsepa-
rated twins come into adolescent or adult age, they
may be involved in the conflict themselves (for
instance, if only one would accept separation).
The principle of justice (similar chances for
both patients) is obviously at risk when it comes
to separation that may involve mutilation or shar-
ing of organs.
565
The principles of beneficiency and non-­
maleficiency (we shall act on the benefit of our
patients and we shall not harm them) that are
considered the ethical backbone of medical
decision-­making are also difficult to apply if sep-
aration is necessary for the survival of only one
twin, if distribution of organs is uneven, and if
separation involves, as is usually the case, loss of
some functions that might be preserved without
separation.
All these considerations are not philosophical
but rather very down-to-earth (Annas 2001).
When separation of conjoined twins is consid-
ered, the patients are usually too young to decide
by themselves, the parents are heavily influenced
by information delivered by doctors, and the
team involved is usually so large and often ethi-
cally discordant that keeping a unified line of
decision becomes difficult (Wasserman 2001).
Acknowledgement of a strong moral leadership
is required after open discussion of every issue,
before passing on information to the parents
regarding the chances and the consequences of
separation. In the case of serious discrepancies,
among all participants in the process of decision,
the courts might be involved (Spitz 2015).
Furthermore, the ethical difficulties may be
seriously aggravated by broadcasting the infor-
mation in the media (this is rarely avoidable now-
adays, due to the large number of people involved
in these cases). The twins and their family should
be protected from the media, and, if possible, the
entire process of decision-making and even the
separation should be kept in the shade. However,
as we said, this is rarely, if ever, possible nowa-
days (Carreker et al. 2018; Frawley 2020).
44.5 Prenatal Management
Every twin pregnancy with a single placenta and
no visible separating amniotic membrane should
be considered as highly suspicious of conjoined
twinning. Other prenatal sonographic findings,
some of them visible from the early stages of
development, are polyhydramnios (up to 50% of
all conjoined twins pregnancies), inseparable
566
fetal bodies and skin contour, unchanged relative
position of the fetuses, both fetal heads always on
the same side, and, finally, a single umbilical
cord containing more than three vessels (Mian
et al. 2017). The use of fetal echocardiography is
mandatory around 20 weeks of gestation, because
twins with shared hearts have a poor prognosis.
Maternofetal MRI also provide useful i­ nformation
about the nature of the joining, but invariably, it
should be completed with a more detailed imag-
ing study after birth (O’Brien et al. 2015).
In cases of complex cardiac fusion or exten-
sive cerebral fusion, termination of pregnancy is
usually recommended (Spitz 2005). When the
decision is to continue with the pregnancy,
C-section should be scheduled at 36–38 weeks of
gestation, ideally at the surgical unit where sepa-
ration will be performed (Mian et al. 2017).
44.6 Imaging
Modern imaging considerably facilitates the
understanding of the often unusual anatomy of
these twins, which is necessary for planning sep-
aration strategies when possible, or to rule them
out when unviable. Standard radiology, including
plain X-rays, GI tract, or urogenital contrast stud-
ies may depict the points of junction and other
features of the corresponding organs, but due to
the atypical anatomy, they rarely produce a fully
accurate picture of its details (Martínez et  al.
2003).
Ultrasonography is used advantageously at
every diagnostic stage. Angiography that was the
only way of detecting the nature of the blood
inflow and outflow of the shared organs has been
largely replaced in the last few years by MR angi-
ography. MRI is the best way to depict the anat-
omy of the fused neural and meningeal tissues in
the heads or spines of craniopagus, rachiopagus,
ischiopagus, pygopagus, or parapagus. It is also
crucial for understanding the anatomy of con-
joined hearts. Helical CT reconstruction of the
bony junctions may be crucial for preparing sur-
gical strategies. Nuclear imaging may help to
define the functional anatomy of the liver, kidney,
or other organs (Watson and McHugh 2015).
J. A. Tovar and L. Martinez
Three-dimensional images and printed models
are useful for surgical guidance (Fallon and
Olutoye 2018).
A close collaboration between radiologists
and the surgical team is the base for a successful
separation of the twins. Every study should be
reviewed and confronted with the anatomy and
physical exploration, in order to anticipate the
surgical procedure (Pierro et al. 2015). The goal
should be to identify all the relevant anatomical
details before separation in order to limit intraop-
erative surprises. However, the team facing sepa-
ration of one of these sets must also be prepared
for unexpected anatomical findings that may
change the order or the nature of the participation
of the different specialists involved. In most
cases, the expected anatomy does not totally cor-
respond with the surgical findings, and a lot of
ingenuity is often required on site for improvis-
ing solutions and adapting the operative proce-
dures accordingly (Kiely and Spitz 2015; Fallon
and Olutoye 2018).
44.7 Postnatal Management:
Technical Issues
of Separation
The decision to separate should always consider
the potential for mortality, for significant disabil-
ity or unacceptable quality of life and, finally, the
potential for the need of an urgent separation in
case of one twin adversely affects the other after
birth (Kiely and Spitz 2015). This phase can be
anticipated prenatally based on sonographic and
MRI findings.
Once the decision for separation has been
accepted by the team and the family, the time of
surgery should be scheduled, ideally after an ini-
tial period of stabilization, growth, imaging stud-
ies, and tissue expansion when needed.
Emergency separation should be attempted when
one twin is dead or dying or where a life-­
threatening correctable congenital malformation
is present in one or both twins. This approach
has, obviously, a higher mortality compared to
elective procedures (Walton et  al. 1991; Nejat
et al. 2017).
44  Conjoined Twins
Some weeks before separation, the leader of
the team should schedule one or more meetings
with scrub nurses, nurses, anesthesiologists, and
surgeons that may involve several specialties
(general pediatric, orthopedic, plastic, urologic,
neurologic and cardiovascular surgery) (Spitz
2005). All technical aspects should be discussed
and eventually rehearsed during this planning
period. Simulation of the transport, manipula-
tion, and placement in the radiological unit and in
the operating room is highly recommended (Liu
et  al. 2018). Comfortably positioning the set of
twins on the table, skin prep and draping, and
moving one twin with the corresponding anes-
thetic equipment to another table for reconstruc-
tion after separation are problems to be solved
according to a previously established plan
(Parmekar et  al. 2018). The expected order and
extent of the participation of each specialist team
in the separation, as well as the critical steps dur-
ing the operation, should be scheduled and antici-
pated. The surgeon in charge of the direction of
the operation acts as an orchestral conductor, and
his/her coordinating activity extends well beyond
the end of the separation itself. Surgical separa-
tion is an exciting moment during the procedure,
but it is only a moment that initiates a complex
reconstruction (Spitz 2005; Fallon and Olutoye
2018).
Anesthesia for separation is already a great
challenge, not only because of the obvious ana-
tomical difficulties for intubation, insertion of
lines, and invasive monitoring but mainly because
both twins are, to a variable extent, in a situation
of cross-circulation in which they share more or
less directly one single internal environment
(Frawley 2020). Thus, the drugs administered to
one pass into the other one and biochemical and
gas monitoring may be confusing. It is a good
policy to take the patients to the anesthetic or
ICU area the day before surgery in order to per-
form these tasks that may last hours, without
interfering with the already tightly scheduled sur-
gical time.
In general, asymmetrical conjoined twins do
not represent surgical challenges beyond those
that are usual in our specialty and we are not
going to address these in the present chapter.
567
The separation of craniopagus is usually a dif-
ficult task, given the complexity of the neural,
arterial and venous connections involved. The
face, base of the skull, or the foramen magnum is
not usually involved in this type of twinning.
Modern imaging is particularly useful in these
cases. Brain arterial supply is usually separated,
but the venous drainage may be shared with
abnormal sinuses, making separation extremely
difficult or even impossible. The final results
depend, to a great extent, on the amount and
nature of the brain tissue and the vascular net-
work shared by the twins (Staffenberg and
Goodrich 2012).
Separation of omphalopagus twins involves
variable difficulties depending on the extent of
organ sharing (Fig. 44.2). These twins more often
have fused livers (80%) and GI tracts (30%)
(Fallon and Olutoye 2018). The degree of fusion
of these two organs will be the main determinant
of morbidity and mortality after separation. A
small liver bridge without major vascular con-
nections is relatively easy to take down, but a
large mass of anatomically atypical liver with
wide arterial, venous, and biliary connections
may be a serious undertaking. When livers are
fused, a preoperative sonography should docu-
ment the presence of one or two gallbladders and
the anatomic disposition of hepatic veins. These
hepatobiliary anomalies should be extensively
studied before separation, but an intraoperative
cholangiogram is often needed (Meyers and
Matlak 2002; Al-Rabeeah et  al. 2011; Barletta
et  al. 2019). Perioperative ultrasonography and
parenchyma dividing devices used for liver resec-
tion are very useful for this purpose. The most
common form of GI tract connection involves
fusion of the small bowel from the upper jejunum
down and divergence near the distal ileum, at the
expected location of a Meckel’s diverticulum, but
in cases of complex hepatobiliary malformations,
the intestines are fused at the second portion of
duodenum. In most cases, separation consists of
allocating half the available gut to each twin.
Additional problems may be found when atresia
of one of the tracts or a common cystic dilatation
of the mid bowel are present. Due to the nature of
these malformations, this type of conjoined twins
568
Fig. 44.2  Set of omphalopagus twins. Severe brain haemorrhage in Twin B after vaginal delivery prompted neonatal
separation. Twin A survived
can often be separated and may have a good long-­
term prognosis.
Thoracopagus are the most common con-
joined twins (40–60%). Cardiac involvement is
present in up to 90% of them, making separation
a rare event, because sharing cardiac tissue makes
the operation impossible in most cases (Fig. 44.3)
(Hamdan et  al. 2010). Only when small atrial
bridges or minimal myocardial fusion exist, the
operation can be realistically faced with all the
facilities for cardiopulmonary bypass available.
In addition, most of these twins have malforma-
tions of the connected hearts and the great vessels
that may further complicate or preclude the sepa-
ration. Sometimes, the aorta and the pulmonary
arteries are hypoplastic, and the infradiaphrag-
matic aortas are often largely connected by thick
trunks that compensate for other circulatory defi-
cits (Chiu et  al. 1994; McMahon and Spencer
2006; Hamdan et  al. 2010). If separation is
decided, surgical planning should include a strat-
J. A. Tovar and L. Martinez
egy for thoracic closure as these twins have an
abnormal disposition of the chest wall, with mal-
formations of the ribs and sternum. Moreover, the
relocation of the hearts into the chest cavity that
can interfere with cardiac inflow or outflow may
become a challenging surgical issue (Fallon and
Olutoye 2018). Only a few survivors after separa-
tion of thoracopagus with common cardiac tissue
have been reported and most involve survival of
only one of the twins. Of those who cannot be
separated, most die of the associated heart dys-
functions in the first months or years of life.
Rachiopagus, ischiopagus, pygopagus, and
parapagus twins share, to a different extent, parts
of the spine, central nervous system, and GI and
GU tracts, and they may represent major chal-
lenges (Figs.  44.4 and 44.5). The separation of
the bony parts requires highly skilled orthopedic
surgeons. In some cases, the reconstruction of the
pelvic rim requires bilateral iliac osteotomies and
pubic fixation. In others, even refashioning a
44  Conjoined Twins
Fig. 44.3  Thoracopagus twins. They shared ventricular
muscle as depicted in MRI. Angio MRI showed that, in
addition, they had pulmonary stenoses, aortic stenosis
bony pelvis is impossible, and the subsequent
prosthetic treatment is difficult. The spine has
often some malformations at other levels, and
scoliosis has to be taken into account during fol-
low-­up. If separation is considered, an extensive
preoperative study should include imaging of
skeletal and spinal cord fusion, vascular anatomy,
genitourinary structures, pelvic and sacral
­anatomy, and also intestinal evaluation (Shapiro
et al. 1991; Jones 2015).
Neurosurgical separation may involve divid-
ing a common spinal cord with reconstruction of
the dural sacs on each side. Since fusion of neural
tissue is usually distal, the motor and sensitive
effects tend to be limited (Fieggen et al. 2004).
The distribution of a common lower GI tract
between both twins entails the loss of continence
for one or both of them. In frontally united twins,
there is usually ileal confluence near the ileocecal
569
(twin B), and a thick infradiaphragmatic arterial trunk
connecting both aortas at that level. Separation was con-
sidered impossible
valve and a single colon. The functional recon-
struction of the pelvic organs is therefore rarely
possible. Seldom, the rectal function can be pre-
served in one twin, but more often, this is impos-
sible, and ostomies have to be fashioned at some
stage. All refinements of advanced bowel man-
agement are necessary to obtain subsequent
adaptation of these patients to a more or less nor-
mal life.
The same can be said about distributing the
urogenital tract structures between the twins
(Cuckow and Mishra 2015). Keeping a bladder
and urethra for one twin is rarely possible in most
frontally united sets (Fig. 44.6). Again, all refine-
ments of reconstructive urology, bladder aug-
mentation, clean intermittent catheterization, and
continent urinary diversion may help to readapt
these patients. The native genital tract can be
reconstructed if duplicated, but in some cases
570
Fig. 44.4  Ischiopagus tetrapus (four legs) twins. The
spines and the spinal cords were joined at the caudal end
as shown by helicoidal CT reconstruction. During separa-
tion, the spines were divided, the meningeal sacs were
reconstructed, a quadruple iliac osteotomy was performed
J. A. Tovar and L. Martinez
for joining both pubic bones in each twin, the urogenital
system was reconstructed, and colostomies were fash-
ioned. Both patients deambulate normally and enjoy rela-
tively normal lives
44  Conjoined Twins
Fig. 44.5  Caudal parapagus twins with an extra thoracic
limb irrigated from the abdominal aorta of twin
A. Separation involved two surgical steps. First, the spinal
cords and meningeal sacs were separated, and subcutane-
ous expanders were inserted. Secondly, the sacrum, the
g.i. and g.u. tracts were divided, and the parietal defects
571
were closed. In twin A, the skin and muscle of the addi-
tional limb were used as a vascularized flap. In twin B, a
synthetic mesh was used for this purpose. Colostomies
were fashioned. Both twins are able to deambulate with
braces
572
Fig. 44.6  Omphalopagus twins sharing a common cloa-
cal exstrophy. Urogenital systems and a common diges-
tive orifice were open at the cloacal plate. They were
successfully separated and, after many operative proce-
dures, fully reconstructed
vaginal replacement is necessary. Genital and
urological orifices should be preoperatively
planned for each twin, but sometimes, they can
only be located during the separation process
(Cuckow and Mishra 2015; Elizondo et al. 2017).
One of the major technical problems posed by
the separation of conjoined twins is the coverage
of the huge parietal defects left. When only one
of the component of the set survives, part of the
wall of the other one can be used to bridge the
defects, but in other cases, all sorts of plastic pro-
cedures have to be put into practice. The skin can
be expanded with subcutaneous expanders prior
to separation (Fig.  44.5). The use of synthetic
patches may help, but often they have to be
inserted in contaminated operative fields that
increase the risk of bacterial colonization and
infection. Occasionally, vascularized musculocu-
taneous material from an extra limb can be used
for coverage, and this is very convenient
(Yokomori et al. 1993).
J. A. Tovar and L. Martinez
44.8 Results
Mortality in conjoined twins is high. When a
diagnosis is firm in early pregnancy, interruption
of gestation is common practice in developed
countries, particularly for the forms with bad
prognosis. Fetal mortality or stillbirth is also fre-
quent. Obstetric mortality or severe birth trauma
remains a real risk when prenatal diagnosis has
not been made, and this happens more often in
less sophisticated health environments in which
pregnancies are not monitored. A considerable
proportion of twins have multiple malformations
that cause demise in the first hours or days of life
(O’Brien et al. 2015).
When separation is deemed possible, it must
be remembered that neonatal operations involve
higher mortality, not only because they are better
performed later when most anatomical and func-
tional features of the set have been ascertained
but also because neonatal separation in the new-
born period often has to be undertaken for life-­
threatening reasons (for instance, one twin may
be very ill or develop intestinal obstruction)
(Fig. 44.7) (Pierro et al. 2015).
Thoracopagus twins rarely survive because
most have severe malformations, in addition to
the common heart. There are only a few reports
of survival of one of the twins after separation
and practically none of survival of both
(Saranrittichai et al. 2007).
Most omphalopagus twins can survive if no
obstetric trauma or severe associated malforma-
tion are present. All other forms of conjoined
twinning may allow survival of both twins but
usually with more or less extensive deficits that
require follow-up for life and often additional
operations.
Short- and long-term sequelae are constant
after separation of conjoined twins. These sequelae
are the consequence of sharing unique or incom-
pletely duplicated organs, tissues, and systems that
belongs to two different individuals. In our experi-
ence, those twins who share axial or pelvic bone
structures or with complex organ common uro-
genital anatomy will need many more operations
44  Conjoined Twins
Fig. 44.7  Omphalopagus twins. Twin A (left) suffer from
a transposition of the great arteries with ventricular septal
defect diagnosed prenatally. A balloon atrial septostomy
was performed at 9  hours of life. Her cardiac function
and, even in some cases, kidney transplantation.
Fecal and urinary continence problems become
dominant with the passage of time, and orthopedic
and motor deficits may require prosthetic appli-
ances (Hoyle and Thomas Jr. 1989; Shapiro et al.
1991; Lazarus et al. 2011; Jones 2015).
It is particularly discouraging that the sophis-
ticated medical and social environments in which
these children have better chances of maintaining
a good quality of life produce the minority of
these cases, probably because such pregnancies
are terminated. In contrast, patients that come for
separation are often from less privileged coun-
tries, where they lack facilities for long-term fol-
low-­ up and all additional support, including
complementary operations, rehabilitation, and
social integration (Tovar 2009).
44.9 Conclusion
Separation of conjoined twins is a major test for
the quality of pediatric surgical care. Only insti-
tutions which possess a higher level of sophisti-
573
deteriorates rapidly so an urgent separation was scheduled
after only two weeks. Twin A died one month later due to
cardiac and intestinal complications
cation for all pediatric surgical subspecialties can
undertake these operations with any chance of
overall success.
References
Al-Rabeeah A, Zamakhshary M, Al-Namshan M,
Al-Jadaan S, Alshaalan H, Al-Qahtani A, Alassiri I
(2011) Hepatobiliary anomalies in conjoined twins. J
Pediatr Surg 46:888–892
Annas GJ (2001) Conjoined twins--the limits of law at the
limits of life. N Engl J Med 344:1104–1108
Barletta A, Pace M, D’Antiga L, Cheli M, Mangili G,
Sironi S (2019) Imaging of liver fusion and vascular
intra-hepatic shunts in omphalopagus conjoined twins:
the role of a double CT study. Clin Imaging 54:25–30
Boer LL, Schepens-Franke AN, Oostra RJ (2019) Two is
a crowd: two is a crowd: on the enigmatic etiopatho-
genesis of conjoined twinning. Clin Anat 32:722–741
Carreker C, De Jong H, Javor V, Fox L, Jacome J, Johns C,
Springmeyer K, Davies J (2018) Family support and
media considerations with conjoined twins. Semin
Perinatol 42:393–401
Chiu CT, Hou SH, Lai HS, Lee PH, Lin FY, Chen WJ,
Chen MT, Lin TW, Chu SH (1994) Separation of tho-
racopagus conjoined twins. A case report. J Cardiovasc
Surg 35:459–462
574
Cuckow P, Mishra P (2015) Urological management.
Semin Pediatr Surg 24:237–240
Elizondo RA, Au JK, Roth DR, Olutoye OO, Cass DL,
Koh CJ (2017) Urologic considerations in the sepa-
ration of Thoracoomphalopagus conjoined twins.
Urology 99:231–233
Endres L, Wilkins I (2005) Epidemiology and biology of
multiple gestations. Clin Perinatol 32:301–314, v
Fallon SC, Olutoye OO (2018) The surgical princi-
ples of conjoined twin separation. Semin Perinatol
42:386–392
Fieggen AG, Dunn RN, Pitcher RD, Millar AJ, Rode H,
Peter JC (2004) Ischiopagus and pygopagus conjoined
twins: neurosurgical considerations. Childs Nerv Syst
20:640–651
Frawley G (2020) Conjoined twins in 2020- state of the
art and future directions. Curr Opin Anaesthesiol
33:381–387
Hamdan MA, Begam M, Mirghani HM (2010)
Inseparable thoraco-omphalopagus twins. Pediatr
Cardiol 31:1252–1253
Hoyle RM, Thomas CG Jr (1989) Twenty-three-year
follow-up of separated ischiopagus tetrapus conjoined
twins. Ann Surg 210:673–679
Jones D (2015) Orthopedic aspects of separation. Semin
Pediatr Surg 24:249–251
Kallberg L (2018) The ethics of separating conjoined
twins: two arguments against. Theor Med Bioeth
39:27–56
Kaufman MH (2004) The embryology of conjoined twins.
Childs Nerv Syst 20:508–525
Kiely EM, Spitz L (2015) Planning the operation. Semin
Pediatr Surg 24:221–223
Lazarus J, Raad J, Rode H, Millar A (2011) Long-term
urological outcomes in six sets of conjoined twins. J
Pediatr Urol 7:520–525
Liu X, Dong K, Zheng S, Xiao X, Shen C, Dong C, Zhu
H, Li H, Bi Y, Ma R (2018) Separation of pygopagus,
omphalopagus, and ischiopagus with the aid of three-­
dimensional models. J Pediatr Surg 53:682–687
Martínez L, Fernández J, Pastor I, García-Guereta L,
Lassaletta L, Tovar JA (2003) The contribution of
modern imaging to planning separation strategies in
conjoined twins. Eur J Pediatr Surg 13:120–124
Martinez-Frias ML, Bermejo E, Mendioroz J, Rodriguez-­
Pinilla E, Blanco M, Egues J, Felix V, Garcia A,
Huertas H, Nieto C, Lopez JA, Lopez S, Paisan L,
Rosa A, Vazquez MS (2009) Epidemiological and
clinical analysis of a consecutive series of conjoined
twins in Spain. J Pediatr Surg 44:811–820
McMahon CJ, Spencer R (2006) Congenital heart defects
in conjoined twins: outcome after surgical separation
of thoracopagus. Pediatr Cardiol 27:1–12
Meyers RL, Matlak ME (2002) Biliary tract anomalies
in thoraco-omphalopagus conjoined twins. J Pediatr
Surg 37:1716–1719
Mian A, Gabra NI, Sharma T, Topale N, Gielecki J, Tubbs
RS, Loukas M (2017) Conjoined twins: from concep-
tion to separation, a review. Clin Anat 30:385–396
Mutchinick OM, Luna-Muñoz L, Amar E, Bakker MK,
Clementi M, Cocchi G, da Graça DM, Feldkamp
J. A. Tovar and L. Martinez
ML, Landau D, Leoncini E, Li Z, Lowry B, Marengo
LK, Martínez-Frías ML, Mastroiacovo P, Métneki J,
Morgan M, Pierini A, Rissman A, Ritvanen A, Scarano
G, Siffel C, Szabova E, Arteaga-Vázquez J (2011)
Conjoined twins: a worldwide collaborative epide-
miological study of the international clearinghouse
for birth defects surveillance and research. Am J Med
Genet C Semin Med Genet 157c:274–287
Nejat F, Habibi Z, Goudarzi M, Azad MS, Moradi E,
Heidari V, Kadivar M, Soltani ZE, Kouchakzadeh L
(2017) Emergency separation of craniopagus twins:
case report. J Neurosurg Pediatr 20:307–313
O’Brien P, Nugent M, Khalil A (2015) Prenatal diagno-
sis and obstetric management. Semin Pediatr Surg
24:203–206
Parmekar S, McMullen L, Washington C, Arnold JL
(2018) Role of simulation in preparation for the care
of conjoined twins-prenatal preparation to separation.
Semin Perinatol 42:329–339
Pierro A, Kiely EM, Spitz L (2015) Classification and
clinical evaluation. Semin Pediatr Surg 24:207–211
Saranrittichai S, Niramis R, Siripornadulsilp S,
Thianjaruwatthana W, Hompleum T, Tipsunthonsak
N, Pulnitiporn A, Vasanawathana S, Jarupoonphol V
(2007) Successful separation of thoracopagus con-
joined twins with a single extra-hepatic biliary system.
J Med Assoc Thail 90:994–1000
Shapiro E, Fair WR, Ternberg JL, Siegel MJ, Bell MJ,
Manley CB (1991) Ischiopagus tetrapus twins: uro-
logical aspects of separation and 10-year followup. J
Urol 145:120–125
Spencer R (2000) Theoretical and analytical embryology
of conjoined twins: part I: embryogenesis. Clin Anat
13:36–53
Spitz L (2005) Conjoined twins. Prenat Diagn 25:814–819
Spitz L (2015) Ethics in the management of conjoined
twins. Semin Pediatr Surg 24:263–264
Staffenberg DA, Goodrich JT (2012) Separation of cra-
niopagus conjoined twins with a staged approach. J
Craniofac Surg 23:2004–2010
Thomas A, Johnson K, Placencia FX (2018) An ethically-­
justifiable, practical approach to decision-making sur-
rounding conjoined-twin separation. Semin Perinatol
42:381–385
Tovar JA (2009) Conjoined twins. In: Puri P, Höllwarth M
(eds) Pediatric surgery. Springer, Berlin, pp 646–655
Walton JM, Gillis DA, Giacomantonio JM, Hayashi AH,
Lau HY (1991) Emergency separation of conjoined
twins. J Pediatr Surg 26:1337–1340
Wasserman D (2001) Killing Mary to save Jodie: con-
joined twins and individual rights. Philos Public
Policy Q 21:9–14
Watson SG, McHugh K (2015) Conjoined twins: radio-
logical experience. Semin Pediatr Surg 24:212–216
Yokomori K, Ohkura M, Kitano Y, Nakajo T, Harii K,
Tanikaze S (1993) Comprehensive planning of opera-
tive strategy for separation of ischiopagus tripus twins
with particular reference to quality of life. J Pediatr
Surg 28:833–837
Part VII
Tumors
 Vascular Anomalies
Anna McGuire, Steven J. Fishman,
and Belinda Hsi Dickie
45.1 Introduction
The term vascular anomalies encompasses two
distinct categories—vascular tumors and vascu-
lar malformations. Vascular tumors can be
benign, locally aggressive, malignant, or inter-
mediate. Vascular malformations are benign
lesions that are present at birth. They are classi-
fied based on the type, or types, of vessels
involved. Combined malformations are com-
posed of two or more types of vessels. There is a
great deal of variability in the natural history of
these lesions, so proper diagnosis is key.
Treatment and management strategies are often
specific to the type of lesion. Interdisciplinary
care at a center with experience in vascular
anomalies helps to ensure the best possible out-
comes for these patients.
Special Mention: this chapter was updated and modified
from its original version by Emily Christison-Lagay, MD
and Steven J. Fishman, MD
A. McGuire · S. J. Fishman · B. H. Dickie (*)
Vascular Anomalies Center, Department of Surgery,
Boston Children’s Hospital, Boston, MA, USA
e-mail: Anna.McGuire@childrens.harvard.edu;
Steven.Fishman@childrens.harvard.edu;
Belinda.Dickie@childrens.harvard.edu
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_45
45
45.2 Historical Overview
Most vascular anomalies involve the skin, the
largest, most visible organ of the body, and there-
fore are noted at birth. For centuries, vascular
birthmarks were referred to by vernacular names
derived from folk beliefs that a mother’s emo-
tions or patterns of ingestion could indelibly
imprint her unborn fetus. The present day use of
such terms as “cherry,” “port wine stain,” and
“strawberry” can be referenced to this false doc-
trine of maternal impressions. Virchow may be
credited with the first effort to categorize vascular
anomalies on the basis of histologic features,
breaking them down into three principal catego-
ries: angioma simplex, angioma cavernosum, or
angioma racemosum. While these more formal
terms attempted a more microanatomic-based
classification of vascular lesions, the lack of
specificity or identifying features perpetuated the
confusion over vascular anomalies well into the
late twentieth century (Mulliken and Young
1988). Today, vascular anomalies may be sepa-
rated into two major categories: tumors and mal-
formations. While this distinction is clinically
and heuristically useful, there are some anoma-
lies that appear to span both categories. Progress
in understanding the biology, genetics, and
pathogenesis of these lesions has led to a more
comprehensive molecular classification of this
diverse collection of lesions. The International
Society for the Study of Vascular Anomalies
577
578
(ISSVA) maintains a current classification
schema with dynamic links to mutations identi-
fied in this field burgeoning with new understand-
ing (issva.org).
45.3 Hemangiomas and Other
Vascular Tumors
45.3.1 Incidence
Hemangiomas represent the most common tumor
of any type seen in infancy with a documented
perinatal incidence of 1.0–2.6% but increasing in
incidence over the first year of life. They are
speculated to affect 4% of Caucasian children.
The incidence appears to be lower in Asian
infants and is very low in children of African
descent. Preterm infants with low birth weight
(<1000 g) have an increased incidence of heman-
gioma (Anderson et al. 2016). A female to male
ratio of 3:1 to 5:1 has been observed. There is no
clear genetic tendency toward hemangioma for-
mation. A family history can be elicited in
approximately 10% of infants, and although stud-
ies have demonstrated no difference in the fre-
quency of co-expression in monozygotic and
dizygotic twins, a subset of patients with heman-
giomas have been identified as having a missense
mutation in the genes encoding vascular endothe-
lial growth factor receptor-2 (VEGFR2) or tumor
endothelial marker-8 (TEM8) (Jinnin et al. 2008).
45.3.2 Etiopathogenesis
The pathogenesis of infantile hemangiomas is
not completely known, but evidence supports the
development of hemangiomas from clonal expan-
sion of endothelial cells subject to either abnor-
mal local cellular signals or an initial somatic
mutation favoring rapid expansion. The tissue of
origin of these endothelial progenitors, however,
remains elusive. Some studies suggest that a pop-
ulation of resident angioblasts, arrested in an
early stage of vascular development, give rise to
A. McGuire et al.
these endothelial cells. A second theory suggests
that these endothelial cells are derived from a dis-
tant population of endothelial precursors carried
by existing vascular pathways to a receptive envi-
ronment. One such possibility is the bone mar-
row; another suggested alternative is the placenta.
A small embolic nidus of placental endothelial
cells could reach fetal tissues through the permis-
sive right to left shunt of fetal circulation. This
could, in part, explain the threefold increased risk
of hemangiomas observed in infants subjected in
utero to chorionic villus sampling, as local pla-
cental injury might predispose the shedding of
cells into the fetal circulation. Moreover, at least
five markers of hemangiomas are uniquely co-­
expressed in the placenta: GLUT1, merosin,
Lewis Y antigen, Fcγ-RIIb, and type III iodothy-
ronine deiodinase (DIO3). A comparison of the
transcriptomes of human placenta and infantile
hemangioma supported a placental origin of the
tumor. This evidence would also support a com-
mon precursor cell of the placental and heman-
gioma endothelium. A placental stem cell with
the same properties has yet to be identified and it
remains possible that hemangioma and placenta
share a common transcriptional profile but not a
common ontogeny (Greenberger and Bischoff
2013).
45.3.3 Diagnosis
Hemangiomas typically appear in the neonatal
period, generally around the first or second week
of life. Approximately one third of hemangiomas
are nascent at birth, presenting as a premonitory
pink macular stain, pale spot, telangiectasia, or
purplish ecchymotic patch. The typical cutaneous
hemangioma permeates the dermis so that the
skin becomes raised, bosselated, and vivid crim-
son in color (Fig.  45.1). This knobby, scarlet
appearance of the superficial hemangioma his-
torically provided inspiration for its common
name of “strawberry hemangioma.” Deeper hem-
angiomas located in the lower dermis, subcutis,
or muscle may present as raised bluish lesions
45  Vascular Anomalies
Fig. 45.1  Occipital hemangioma in a one-year-old child. surrounded by a pale halo. Occasionally, a con-
The center of the lesion has begun to involute into fibro-­ genital hemangioma exhibits an area of central
fatty tissue, while the borders are more indicative of pro-
liferating endothelium
with indistinct borders manifesting at 2–3 months
of life or later.
Hemangiomas are most commonly located
in the head and neck region (60%), followed in
frequency by the trunk (25%), and extremities
(15%). Approximately 80% of hemangiomas
are isolated to a single location, but nearly one
fifth proliferate in multiple sites. Infants with
multiple skin lesions may also have visceral
organ involvement. Gastrointestinal hemangio-
mas may present with anemia or rectal bleeding
(McGuire and Dickie 2020). Hepatic hemangi-
omas are broadly broken down into three main
categories—focal, multifocal, and diffuse.
Focal hepatic hemangiomas are typically con-
genital hemangiomas, specifically rapidly invo-
luting congenital hemangiomas (RICH), and
are fully formed at birth. Multifocal hepatic
hemangiomas are infantile hemangiomas which
proliferate postnatally. Multifocal hepatic hem-
angiomas can progress to a diffuse pattern
where lesions replace the hepatic parenchyma.
579
While most hepatic hemangiomas are probably
clinically silent, a subset becomes symptom-
atic, manifesting as cardiac failure secondary to
high volume shunting, hypothyroidism second-
ary to overproduction of type III iodothyronine
deiodinase, fulminant hepatic failure, and/or
abdominal compartment syndrome. Five or
more cutaneous hemangiomas warrant a screen-
ing hepatic ultrasound, as many of these infants
will have hepatic hemangiomas as well. These
hepatic lesions should be followed with serial
ultrasounds until they have involuted (McGuire
et al. 2020).
In addition to the typical postnatal (“infan-
tile”) hemangioma, tumors, frequently referred to
as congenital hemangiomas, evolve in utero and
present fully grown at birth. The hemangiomas
can vary in appearance but are most commonly
present as raised dome-like purple masses, often
necrosis.
45.3.4 Clinical Course
Infantile hemangiomas exhibit unique biologic
behavior: they grow rapidly during the first
6–12 months of life (proliferative phase), enter a
second stage of growth proportionate with that of
the child, and finally enter a phase of slow regres-
sion (involuting phase) lasting 1–7 years during
which the endothelial matrix of hemangiomas is
replaced by loose fibrous or fibro-fatty tissue.
Regression is complete in half of the children by
age 5, in 70% of children by 7 years, and in the
remainder by age 10–12  years. In the involuted
phase, nearly normal skin is restored in approxi-
mately 50% of children. Otherwise, the involved
skin is damaged with telangiectasias, laxity, and
yellowish discoloration or scarred patches. In
rare cases, a large facial tumor can be associated
with cartilaginous or bony overgrowth (presum-
ably secondary to a local environment of
increased blood flow) or produce a mass effect on
the local facial skeleton.
580 A. McGuire et al.

Congenital hemangiomas, defined as a tumor can be accompanied by ocular abnormalities

present at birth, occur in two forms. The rapidly
involuting congenital hemangioma (RICH)
begins to regress during early infancy and is fully
involuted by age 12–14 months. The noninvolut-
ing congenital hemangioma (NICH) demon-
strates proportional growth to that of the child
and never undergoes regression. These tumors
may be distinguished from the more common
infantile hemangiomas by a variety of molecular
and histopathologic markers. More recently,
there are partially involuted lesions now that have
been labeled partially involuting congenital hem-
angioma (PICH).
45.3.5 Associated Malformative
Anomalies
Historical association of hemangiomas with a
wide variety of syndromes is most likely second-
ary to miscategorization of another vascular
lesion. However, certain true hemangiomas do
occasionally occur in association with other mal-
formations. A large cervicofacial hemangioma
(e.g., microphthalmia, congenital cataract, or
optic nerve hypoplasia), sternal clefting, supra-
umbilical raphe, persistent intracranial and extra-
cranial embryonic arteries, absence of ipsilateral
carotid or vertebral vessels, coarctation of a right
sided aortic arch, and Dandy–Walker cystic mal-
formation or other posterior fossa malformations
(PHACES syndrome) (Hartemink et  al. 2008).
Lumbosacral hemangioma is one of several ecto-
dermal lesions, such as hypertrichosis, capillary
malformation (port-wine stain), achordoma, and
sacral dimple that are known to signal underlying
occult spinal dysraphism (e.g., lipomeningocele,
tethered spinal cord). In patients with sacral hem-
angioma, ultrasound can be used to screen infants
less than 4 months of age for occult spinal dysra-
phism, whereas MRI is usually necessary to iden-
tify spinal cord abnormalities in older children.
There are rare reported incidences in which pel-
vic and perineal hemangioma is associated with
urogenital and anorectal anomalies. The constel-
lation of these lesions has been given the acro-
nym PELVIS or LUMBAR syndrome (Table 45.1)
(Golabi et al. 2014; Girard et al. 2006).

Table 45.1  PHACES () and LUMBAR/PELVIS (Golabi et al. 2014; Girard et al. 2006) syndromes and their associated
anomalies
Syndrome Associated Anomalies
PHACES  • Posterior fossa (Dandy–Walker malformation)
 • Hemangioma (most commonly on the face).
 • Arterial/Cerebrovascular (abnormalities of the carotid, vertebral, or trigeminal arteries)
 • Cardiac (congenital heart defects, aortic coarctation)
 • Eye (Microphthalmos, Horner’s syndrome)
 • Sternal (Sternal clefting)
PELVIS/LUMBAR  • Lower body infantile hemangioma
 • Urogenital anomalies/Ulceration (Bifid scrotum/ulceration of the hemangioma)
 • Myelopathy (tethered cord)
 • Bony deformities (dysplasia)
 • Anorectal malformations/Arterial anomalies (imperforate anus, fistulas, and anterior
displacement/arterial hypoplasia)
 • Renal anomalies (solitary kidney, hypoplasia, and hydronephrosis)
 • Perineal hemangioma
 • External genitalia malformations (bifid scrotum, ambiguous genitalia)
 • Lipomyelomeningocele
 • Vesicorenal abnormalities (bladder exstrophy)
 • Imperforate anus
 • Skin tag
45  Vascular Anomalies
45.3.6 Differential Diagnosis
Kaposiform hemangioendothelioma (KHE) and
tufted angioma (TA) should be in the differential
for vascular tumors in infant patients. Both types
of tumor are typically present at birth. Unlike
infantile hemangioma, KHE and TA affect both
genders equally, are unifocal, and generally
involve the trunk, shoulder, thigh, or retroperito-
neum. The overlying skin is deep red-purple in
color, tense, and shiny (Fig.  45.2). Ecchymosis
appears over and around the tumor in association
with generalized petechiae and may falsely raise
concern for child abuse. Common infantile hem-
angioma markers Glut1 and Lewis Antigen Y are
absent in these vascular tumors.
These tumors are often associated with
Kasabach–Merritt phenomenon. In 1940, radiolo-
gist Kasabach and pediatrician Merritt reported a
child with profound thrombocytopenia, petechiae,
and bleeding in the presence of a “giant heman-
gioma.” More recently, it became clear that per-
sistent, profound thrombocytopenia is never
associated with common hemangioma of infancy.
Thrombocytopenia unresponsive to platelet trans-
fusion can be profound (<10,000 mm3), but coag-
ulation values (partial thromboplastin time and
prothrombin time) are normal to mildly elevated.
The child with Kasabach–Merritt thrombocytope-
Fig. 45.2  An infant with kaposiform hemangioendothe-
lioma involving the right arm, shoulder, and trunk. The
overlying skin is crimson-purple and shiny with signifi-
cant associated edema. A separate lesion lower in the fore-
arm is not uncommon
581
nia is at risk for intracranial, pleural-pulmonic,
intraperitoneal, or gastrointestinal hemorrhage
with an associated mortality of 20–30%.
Diagnosis can be aided, if necessary, by MRI,
which demonstrates enhanced signal on
T2-weighted images and dilated feeding/drain
vessels like other vascular tumors but, specific to
KHE, demonstrates a poorly defined tumor mar-
gin with extension across tissues, prominent sub-
cutaneous stranding, and small vessels relative to
tumor size. Histopathologic examination differs
between KHE and TA.  KHE demonstrates an
aggressive cellular pattern of infiltrating sheets or
nodules of slender epithelial cells, slit-like vascu-
lar spaces filled with hemosiderin and fragments
of red blood cells, and coexistent dilated lym-
phatic spaces. TA, while macroscopically similar
to KHE, is histologically composed of small tufts
of capillaries (cannonballs) in the middle to lower
dermis with lymphatics present at the periphery
(Singh et al. 2019).
Several dermatologic entities of early infancy
are sometimes confused with an infantile heman-
gioma. The nevus flammeus neonatorum, known
by lay terms as “angel’s kiss,” “stork bite,” or
“salmon patch,” is a non-evolving macular stain
typically vanishing by the first year of life. A
deep hemangioma can be confused with either a
localized lymphatic or venous malformation
causing the overlying skin to assume a bluish
tinge with a few telangiectasias or draining veins.
Diagnosis may be aided by palpation of the
lesion: just as with a superficial hemangioma,
those of the deeper tissues are rubbery in consis-
tency. Venous and lymphatic malformations are
usually quite soft and compressible unless intral-
esional bleeding or thrombosis has occurred.
Hemangiomas can also imitate port-wine stains
(capillary malformations) and the blush of arte-
riovenous malformations (AVMs).
Hemangiomas can be confused with another
common cutaneous vascular tumor known as a
pyogenic granuloma. These are small lesions
rarely appearing before 6 months of age (average
age 6.7  years). There is an association of pyo-
genic granulomas with port wine stains. The
lesions grow rapidly, erupting through the skin on
582
a stalk or pedicle. Epidermal breakdown with
crusting is the norm associated with recurrent,
often copious, bleeding. Deep pyogenic granulo-
mas can also imitate deep hemangiomas. The
best treatment is curettage or excision.
45.3.7 Management
Most hemangiomas are small, and will regress
without need for intervention. They should be
allowed to undergo proliferation and involution
under the careful observation by a pediatrician
with gentle and sympathetic parental education
and reassurance. Referral to a specialty center
should occur in the event of equivocal diagnosis,
at risk locations, large size, rapidity of growth, or
potential for other complications.
Spontaneous epithelial breakdown, crusting,
ulceration, and necrosis occur in 5% of cutaneous
hemangiomas, and are most common in mucosal
hemangiomas of the lips or anogenital/perineal
region. Initial treatment should be the application
of a petroleum based antibiotic salve along with
viscous lidocaine to assist with pain control.
Petroleum-based salve can also be used
­prophylactically to prevent ulceration and skin
break down. Silicone-based dressing can also be
used as local protection. If there is an eschar on
the surface of the tumor, sharp debridement and
wet-­to-­dry dressing changes are used to stimulate
granulation tissue. Ulceration of larger than sev-
eral millimeters is an indication for referral.
Pharmacologic treatment with propranolol can
accelerate healing, expedite regression, and mini-
mize recurrence. Since the success of propranolol
reported in 2008 (Léauté-Labrèze et  al. 2008),
this medication has become the mainstay of ther-
apy and has nearly entirely replaced the prior
widespread use of corticosteroid. Flashlamp
pulsed-dye laser is also reported to aid healing
and alleviate pain. Total resection of an ulcerated
hemangioma should be considered if the resul-
tant scar would be the same as if the regressing
tumor were removed later in childhood.
Punctate bleeding can occasionally compli-
cate a bosselated, protuberant hemangioma.
Parents should be instructed to apply a full ten
A. McGuire et al.
minutes of pressure to the area with a clean pad.
In very rare instances, a suture is required for
control of a local bleeding site.
45.3.8 Endangering Complications
Advances in pharmacotherapy have reduced the
complications related to hemangiomas if they are
properly diagnosed and treated in a timely fash-
ion. Cervicofacial hemangiomas may cause
destruction, distortion, or obstruction, if untreated
or treated late. A large hemangioma can cause a
mass effect and expansion of tissue. A periorbital
hemangioma can block the visual axis and cause
deprivation amblyopia or extend into the retro-
bulbar space causing ocular proptosis. Similarly,
a hemangioma of the upper eyelid can distort the
growing cornea, producing astigmatic amblyo-
pia. Such cases should be referred for immediate
evaluation by a pediatric ophthalmologist.
Even a small hemangioma can obstruct the
subglottis; hence any cervical hemangioma
should be treated with a high index of suspicion.
Symptoms include hoarseness and, later, typi-
cally around 6–8 weeks of age, biphasic stridor
and may be confused with croup. Approximately
one half of these infants have cutaneous cervical
hemangioma, often in the “beard distribution.”
Other rare (approximately 1%) complications
include high-output congestive heart failure in
association with a hepatic hemangioma and gas-
trointestinal bleeding from mucosal hemangio-
mas of the bowel. Many GI hemangiomas may be
symptomatically controlled with pharmacologic
treatment. Rarely endoscopic or operative inter-
vention is required if unresponsive to medical
therapy if the lesions are focal; however, diffuse,
infiltrating hemangiomas are unamenable to
resection and management is dependent on trans-
fusion, parenteral nutrition, and maximal phar-
macologic therapy.
Hepatic hemangiomas may rarely present
with high-output cardiac failure or massive hepa-
tomegaly and abdominal compartment syndrome.
Infantile hepatic hemangiomas may also cause
profound hypothyroidism due to the expression
of a deiodinase that inactivates thyroid hormones.
45  Vascular Anomalies
Infants with symptomatic hepatic hemangiomas
should be referred to specialty centers.
Pharmacologic therapy should be used to accel-
erate hemangioma regression, embolization may
be employed to correct high-output arteriovenous
or portohepatic shunting, and aggressive thyroid
replacement therapy should be instituted in those
infants with documented hypothyroidism. In rare
cases, liver transplantation may be indicated
(McGuire et al. 2020).
45.3.9 Pharmacologic Therapy
The use of propranolol has dramatically changed
the treatment and course of infantile hemangio-
mas. Propranolol has been found to reduce prolif-
eration and expedite regression of these lesions.
A group of physicians from Bordeaux Children’s
Hospital in France were the first to note and pub-
lish this phenomenon. They observed patients
treated with propranolol for cardiac issues had
regression of their hemangiomas while on the
medication (Léauté-Labrèze et  al. 2008).
Propranolol is now the first line treatment modal-
ity with multiple studies demonstrating its
­effectiveness. It is important to note, that it is
only therapeutic for infantile hemangiomas as
congenital hemangiomas are fully formed at birth
and thus are not responsive to the therapy.
Propranolol should be considered for large or
ulcerated cutaneous hemangiomas, as well as
hemangiomas in sensitive areas (near the eye,
around the airway, etc.). It should be initiated for
patients with hepatic hemangiomas who have a
large or increasing tumor burden and those with
high output cardiac failure, shunting, or hypothy-
roidism. Patients with bleeding or multiple hem-
angiomas of the GI tract should be started on
therapy. The propranolol helps with the cardiac
failure, hypothyroidism, and GI bleeding by
reducing the tumor burden. Patients on proprano-
lol should be monitored for hypotension and
hypoglycemia. Topical timolol can be used for
small or medium sized singular, cutaneous hem-
angiomas. The timolol aids in regression and
while avoiding systemic therapy.
583
Systemic corticosteroids with or without vin-
cristine are occasionally used for disfigurement,
bleeding, or cardiac instability refractory to pro-
pranolol (Zhang et al. 2016). Interferon alpha was
previously used for these lesions but the unfortu-
nate occurrence of irreversible spastic diplegia
observed in patients has rendered it unusable.
45.3.10 Interventional Therapy
With the use of propranolol, the need for opera-
tive resection has become less frequent but may
be indicated in any stage of its life cycle. A well-­
localized, pedunculated hemangioma, particu-
larly one demonstrating ulceration or subject to
recurrent episodes of bleeding, may be removed
early in infancy during the tumor’s proliferative
phase. Hemangiomas with bleeding or repeated
ulceration despite propranolol can be considered
for resection. Additionally, hemangiomas may
not fully regress and can leave staining, scarring,
expanded skin, or fibrofatty tissue. The residual
lesion can be resected for cosmetic purposes, if
desired. If it is only residual staining that is both-
ersome, flashlamp pulsed dye laser can be used.
Focal or multifocal GI hemangiomas, which per-
sist in bleeding despite pharmacotherapy, may be
considered for removal, although one cannot
always assume that an endoscopically visualized
lesion is the only source for bleeding.
As children progress through their preschool
years and develop a sense of physical awareness,
consideration should be given to staged or total
excision of a large or protuberant involuting
phase hemangioma, if the lesion risks signifi-
cantly compromising the child’s body image.
Most commonly, it is preferable to wait until late
childhood to remove the hemangiomatous resid-
uum of the involuted phase. Protrusive hemangi-
omas frequently leave in their wake unsightly
expanded skin and fibrofatty tissue. Staged resec-
tion may be indicated to minimize distortion and
have an acceptable cosmetic outcome. Small,
well-localized cutaneous or mucosal hemangio-
mas in critical locations may be considered for
intralesional injection of corticosteroid.
584
45.4 Vascular Malformations
Vascular malformations are localized or diffuse
errors of embryonic development which may
affect any segment of the vascular tree, including
arterial, venous, capillary, and lymphatic vessels.
It is useful to subcategorize vascular malforma-
tions on the predominant type of channel abnor-
mality and flow characteristics. According to this
distinction, two major categories exist: (1) slow-­
flow anomalies (capillary malformations, lym-
phatic malformations, and venous malformations)
and (2) fast-flow anomalies (arterial malforma-
tions [e.g., aneurysm, coarctation, ectasia, and ste-
nosis], AVMs, and arteriovenous fistulas).
Complex, combined vascular malformations also
exist including slow-flow (capillary-lymphatic
[CLM], capillary-lymphaticovenous [CLVM], and
lymphaticovenous [LVM]) malformations and
fast-slow (capillary-lymphatic AVMs [CLAVMs]
and capillary-lymphatic AVFs [CLAVFs]).
Congenital vascular malformations occur at a
rate of approximately 1.2–1.5%, making them a
relatively common defect in embryogenesis. Most
vascular malformations are sporadic (i.e., nonfa-
milial), but some exhibit classic Mendelian inheri-
tance. Over the past decade, great advances have
been made in understanding the pathogenesis and
genetic mutations of vascular malformations.
45.4.1 Capillary Malformation
Still commonly referred to as port-wine stains,
capillary malformations (CMs) are dermal vascu-
lar anomalies that are reported to occur in 0.3%
of newborns with an even gender distribution
(Fig.  45.3). They can be localized or extensive,
occurring anywhere on the body, but are rarely
multiple. CM must be differentiated from the fad-
ing macular stain nevus flammeus neonatorum,
the most common vascular birthmark. These lat-
ter represent a minor transient dilation of dermal
vessels and must be relabeled if they persist into
childhood. CMs are composed of dilated, ectatic
capillary-to-venule sized vessels in the superfi-
cial dermis. Immunohistochemical studies dem-
onstrate normal endothelial and smooth muscle
A. McGuire et al.
Fig. 45.3  Capillary malformation on a young infant.
Facial lesions can darken with age and the skin can
become thickened and cobblestoned
cell morphology and mitotic index, but a paucity
of surrounding normal nerve fibers. With age, the
vessels gradually dilate, probably accounting for
the darkening color and tendency to nodular ecta-
sias. Although CMs are usually sporadic, a famil-
ial pattern of autosomal dominant inheritance
with incomplete penetration has been reported.
CMs are often associated with hypertrophy of
the soft tissue and underlying skeleton. They can
also signal underlying structural abnormality. A
midline occipital CM can overlie an encephalo-
cele or ectopic meninges. A CM over the cervical
or lumbosacral spine can be a clue to occult spi-
nal dysraphism. A child with a CM of the first or
first-second trigeminal nerve distribution should
be evaluated for Sturge–Weber syndrome, an
associated vascular anomaly of the ipsilateral
choroid and leptomeninges. This is secondary to
a somatic activating mutation in the GNAQ gene
(Shirley et  al. 2013). Clinical manifestations
include seizures, contralateral hemiplegia, and
variable developmental motor and cognitive
delay. Choroid involvement leads to increased
risk for retinal detachment, glaucoma, and blind-
ness. Fundoscopic examination and tonometry
are essential and should be performed twice
annually for two years and yearly thereafter for
life. Early diagnosis can be suggested by MRI
45  Vascular Anomalies
demonstrating pial vascular enhancement.
Angiographic findings include parenchymal
blush and apparent cortical venous occlusions
and collaterals. Other syndromes associated with
capillary malformation include Klippel–
Trenaunay syndrome, a combined slow-flow
CLVM with axial elongation and overgrowth in
girth involving one or more extremities, and
Parkes Weber syndrome, a fast-flow CALVM
vascular anomaly comprising a capillary stain
with AVM or AVFs, as well as lymphatic anoma-
lies of a limb. While most CMs are sporadic, a
mutation in the gene RA5AI is associated with
hereditary CMs, sometimes in combination with
arteriovenous malformations.
Flashlamp pulsed dye laser is currently the
treatment of choice for selective photothermoly-
sis of CM. The optimal timing is controversial. In
general, significant lightening is observed in
approximately 70% of patients with better out-
comes observed on the face than on the trunk and
limbs. Children with Sturge–Weber syndrome
and seizures refractory to pharmacologic treat-
ment may require neurosurgical resection of the
involved brain.
45.4.2 Telangiectasias
Tiny acquired capillary vascular marks, com-
monly known as spider nevus or spider telangiec-
tasis typically appear on children in the preschool
and school-aged years. Epidemiological studies
suggest they may be present in nearly half of all
children with an equal gender distribution.
Spontaneous disappearance is possible, but
pulsed dye laser successfully removes the lesion.
Cutaneous marbling of the skin of Caucasian
infants placed at a low temperature, so-called,
cutis marmorata or livido reticularis is an accen-
tuated pattern of normal cutaneous vascularity
which improves with age as the skin thickens. In
one rare congenital pathologic disorder, the new-
born has a distinctive deep purple, serpiginous,
and reticulated vascular staining pattern, called
cutis marmorata telangiectatica congenita
(CMTC). This vascular birthmark occurs in a
localized, segmental distribution, usually involv-
585
ing the trunk and extremities. Neonatal ulceration
of the depressed purple areas can occur; some-
times there is hypoplasia of the affected limb and
subcutaneous tissues. Almost all infants with
CMTC demonstrate steady improvement during
the first year of life. In time, venous dilation
becomes more prominent and persists into adult-
hood, together with residual cutaneous atrophy
and staining.
Hereditary hemorrhagic telangiectasia (HHT;
Osler–Weber–Rendu syndrome) is an autosomal
dominant disorder with high penetrance and vari-
able phenotype estimated to occur at a frequency
of 1–2 per 100,000 Caucasian populations.
Patients with HHT have mucocutaneous telangi-
ectasias, cerebral and pulmonary AVMs, and
hepatic vascular anomalies. Eighty-five percent
of HHT patients have a mutation in ENG or
ACVRL1; both are associated with loss of func-
tion of TGF-β (Pawlikowska et al. 2015).
Ataxia-telangiectasia (Louis–Bar syndrome)
is an autosomal recessive neurovascular disorder
that appears at 3–6 years of life. Bright red telan-
giectasias are first noted on the nasal and tempo-
ral area of the bulbar conjunctiva and subsequently
manifest on the face, neck, upper chest, and
flexor surfaces of the forearms. Cerebellar ataxia
begins nearly synchronously, followed by pro-
gressive motor neuron dysfunction. Endocrine
and immunologic deficiencies manifest and death
usually occurs in the second decade from recur-
rent sinopulmonary infections or from lymphore-
ticular malignancy. The disease is the result of
homozygous or compound heterozygous muta-
tions of the ATM gene, which is involved in DNA
repair.
45.4.3 Lymphatic Malformation
Historically termed lymphangioma or cystic
hygroma, slow-flow vascular anomalies of the
lymphatic system consist of localized or diffuse
malformations of lymphatic channels best charac-
terized as microcystic, macrocystic or both.
Lymphatic malformations (LMs) most commonly
appear as ballotable masses with normal overlying
skin, although a blue hue may result if large under-
586
Fig. 45.4  Truncal lymphatic mutation on an infant. In
this instance, the overlying skin is unaffected and the mal-
formation presents as a soft, ballotable mass
lying cysts are present (Fig. 45.4). Less commonly,
dermal involvement manifests as puckering or
deep cutaneous dimpling (peau d’orange appear-
ance). LMs in the subcutis or submucosa manifest
as tiny vesicles. Intravesicular bleeding is evi-
denced by tiny, dark red dome-shaped nodules.
Prenatal ultrasound can detect macrocystic
LM in the late first trimester. Those LMs not
diagnosed prenatally are generally evident at
birth; however, occasionally they can manifest
suddenly in older children and adults. Radiologic
documentation is best performed by MRI,
although ultrasound is a useful auxiliary agent to
confirm the presence of LM. With the advent of
intranodal puncture technique, conventional con-
trast lymphangiography has experienced a resur-
gence. It may be useful for determining the
precise location of lymphatic or chylous leaks in
a patient with a diffuse thoracic lymphatic
anomaly.
LMs are most commonly located in the axilla/
chest, cervicofacial region (70–80%), mediasti-
num, retroperitoneum, buttock, and anogenital
areas. Facial LM is the most common basis for
macrocheilia, macroglossia, macrotia, and mac-
romala. Cervicofacial LM is associated with the
overgrowth of the mandibular body, causing an
open bite and underbite. LMs in the floor of the
mouth and tongue are characterized by vesicles,
intermittent swelling, bleeding, and the possibil-
A. McGuire et al.
ity of oropharyngeal obstruction. Cervical LMs
involve the supraglottic airway and tracheostomy
may be necessary early in infancy. Diffuse tho-
racic lymphatic anomalies or rare abnormalities
of the thoracic duct or cisterna chyli can manifest
as recurrent pleural and pericardial chylous effu-
sion or chylous ascites. Anomalous lymphatics in
the GI tract can cause hypoalbuminemia as the
result of chronic protein-losing enteropathy. LMs
in an extremity cause diffuse or localized swell-
ing or gigantism with both soft tissue and skeletal
overgrowth. Progressive osteolysis, caused by
diffuse soft tissue and skeletal LM, is called
Gorham–Stout syndrome and is known also as
“disappearing bone disease” or “phantom bone
disease.”
Lymphedema should also be included as a
type of LM.  Type I hereditary lymphedema
(Milroy disease) is an autosomal dominant disor-
der presenting early in life with localized areas of
edema. The initial superficial lymphatics of these
areas are thought to be hypoplastic or absent,
although superficial lymphatics are observed in
non-edematous areas. Type II hereditary lymph-
edema (Meige disease) is a late-onset autosomal
dominant disorder with variable penetrance and
phenotype. Associated features include distichia-
sis (a double row of eyelashes), ptosis, cleft pal-
ate, yellow nails, and congenital heart disease.
The disorder is thought to arise from an impair-
ment of lymphatic drainage and lymphoscintigra-
phy demonstrates numerous dilated lymphatic
vessels (Connell et al. 2008).
The two main complications of LMs are intra-
lesional bleeding and infection. Bleeding, spon-
taneous or the result of local trauma, causes
rapid, painful enlargement of an LM.  The LM
becomes firm and ecchymotic. Analgesia, rest,
and time are generally sufficient. Hemorrhage
and infection can transform a macrocystic lesion
into a microcystic and scarred lesion.
LMs often swell in the event of a viral or bac-
terial infection. Most often this is a harmless
event likely related to change in lymphatic flow.
Bacterial cellulitis, however, is a more dangerous
condition. Prolonged intravenous antibiotic ther-
apy is frequently indicated, with choice of antibi-
45  Vascular Anomalies
otic agents based upon the presumptive pathogens
of the affected area.
The treatment of LMs, is often a combination
of interventional, surgical, and pharmacologic
therapy. LMs are the result of localized, somatic,
activating mutation in PIK3CA. As such, siroli-
mus, an mTOR inhibitor, has been used to treat
these lesions by inhibiting upstream from
PIK3CA. Sirolimus inhibits growth of the lesion
and can even induce regression. It also reduces
infection rate and leakage from vesicles.
The two strategies for interventional treatment
of lymphatic anomalies are sclerotherapy and
operative resection. Sclerotherapy works through
obliteration of the lymphatic lumen by endothe-
lial destruction with subsequent sclerosis/fibrosis.
Success depends upon the sclerotic agent selected
and the damage inflicted upon the e­ ndothelial and
deeper muscular and connective tissue layers.
Macrocystic LM is more likely than microcystic
tissue to shrink after an injection of sclerosant.
Typically, doxycycline is the most commonly
used agents. Ethanol was widely considered to be
the most effective sclerosing agent for low-flow
malformations, but side effects are potentially
increased and use is not as common. Injection
with sclerosants is quite painful, often requiring
general anesthesia and subsequent pharmacologic
pain relief. Edema following sclerosant injection
is associated with prolonged recovery and
increased therapeutic effect. Side effects include
local necrosis and blistering as well as local neu-
ropathy. Systemic absorption of ethanol may lead
to cardiac arrest, pulmonary vasoconstriction, or
systemic hypotension. Bleomycin is sometimes
used in the microcystic lesions. There has been
reported pulmonary fibrosis as a potential compli-
cation of bleomycin. Other sclerosant agents
include sodium tetradecyl sulfate if there is a
venous component.
Resection is the only way to potentially “cure”
an LM.  Often staged excision is necessary and
total excision is possible. In each resection, a sur-
geon should focus on a defined anatomic region,
attempt to limit blood loss, perform as thorough a
dissection as possible, and to be prepared to oper-
ate as long as necessary. Even with such an inten-
587
sive approach to resection, the “recurrence” rate
is reported to be 40% after an incomplete exci-
sion and 17% after a macroscopically complete
excision.
In recent years, the genetics of LMs have led
to more effective pharmacologic agents. LMs are
the result of localized, somatic, mutations and
most commonly in the PIK3CA pathway. As
such, sirolimus, an mTOR inhibitor, has been
used to treat these lesions by inhibiting upstream
from PIK3CA (Wiegand et al. 2018). Sirolimus
inhibits growth of the lesion and can even induce
regression. It also reduces infection rate and leak-
age from vesicles. Sirolimus can be started pre-
operatively and continued throughout the
postoperative period to reduce lymphatic leakage
after resection. For disease refractory to siroli-
mus, BYL719 (PIQRAY®, Cambridge, MA), a
direct PI3Kα inhibitor, is available for compas-
sionate use for patients with a documented
PIK3CA mutation.
45.4.4 Venous Malformation
Venous malformations are the most common of
all vascular anomalies and are frequently misdi-
agnosed as hemangiomas or mislabeled “cavern-
ous hemangiomas.” While present at birth, they
are not always immediately evident. The typical
description of a VM is of a blue, soft, and com-
pressible mass (Fig. 45.5). VM can vary greatly
in size, shape, and degree of associated deforma-
tion. Venous malformations demonstrate propor-
tional growth with the growth of the child.
Histologically, VMs are composed of thin-­
walled, dilated, sponge-like abnormal channels.
The normal architecture of vascular smooth
muscle is distorted into clumps. This mural mus-
cular abnormality is likely responsible for the
tendency of VMs toward gradual expansion.
Microscopy often reveals evidence of clot for-
mation, fibrovascular ingrowth, and phleboliths.
Phlebothrombosis is common and can be pain-
ful. They are the result of localized, somatic
mutations in the TEK gene. TEK encodes the
TIE2. Most cases are sporadic.
588
Fig. 45.5  Characteristic venous malformation on the
back of a school age boy. Dilated venous channels con-
tribute to a pale blue appearance. The lesion itself is soft
and compressible
Most VMs are solitary, but multiple cutane-
ous or visceral lesions can occur. There are
well-­documented pedigrees of familial, usu-
ally multifocal, VMs. Blue-rubber bleb nevus
syndrome is a rare, sporadic disorder com-
posed of cutaneous and gastrointestinal VMs.
The gastrointestinal lesions often cause
chronic gastrointestinal bleeding. Cutaneous
lesions can occur anywhere on the body, but
there is a predilection for the trunk, palms, and
soles of the feet. The lesions increase in size
and number, and become more apparent with
age. Gastrointestinal lesions are located
throughout the GI tract, but are most fre-
quently located in the small bowel (Fig. 45.6).
In addition to bleeding, these lesions can pro-
vide a lead point for intussusception or
volvulus.
A. McGuire et al.
Fig. 45.6  Intraoperative findings in a patient with blue-­
rubber bleb nevus syndrome. Hundreds of millimeter- to
multi-centimeter-sized lesions throughout the small and
large bowel and mesentery can be responsible for gastro-
intestinal hemorrhage
Complications of venous malformations
depend upon their location. VMs of the head
and neck can cause progressive distortion of
facial features, exophthalmia, dental malalign-
ment, and obstructive sleep apnea. VMs of the
extremities can cause limb length discrepancies,
painful hemarthrosis, and degenerative arthritis.
Intraosseous VMs can cause structural weaken-
ing of the bone shaft and pathologic fracture.
VMs of the gastrointestinal tract can manifest
with chronic bleeding and anemia. While VMs
may be localized throughout the length of the
bowel, they are most commonly found encom-
passing the entire left colon and rectum and sur-
rounding pelvic and retroperitoneal structures.
Lesions that involve the foregut can be associ-
ated with central mesenteric and portal venous
anomalies.
Like other vascular malformations, venous
malformations are best imaged by MRI which
demonstrates T2 enhancement and differ from
45  Vascular Anomalies
lymphatic slow-flow lesions by the presence of
contrast enhancement of the vascular spaces.
Sclerotherapy is the first line therapy for many
VMs. For small cutaneous or oromucosal VMs,
injection with 1–3% sodium tetradecyl sulfate is
often successful. Venous anomalies have a pro-
pensity for recanalization and recurrence.
Operative excision of a VM is usually successful
for small, well-localized lesions. In some loca-
tions, staged subtotal surgical removal can be
accomplished without preoperative sclerotherapy.
However, sclerotherapy may be used as a first line
approach to attempt to shrink the VM prior to sur-
gical resection. Complete resection of large, focal
GI VMs is occasionally necessary if chronic
bleeding, anemia, and transfusion requirements
are refractory to sirolimus. Multifocal GI lesions
are best treated by multiple excisions (sometimes
numbering in the hundreds). Bowel resections
should be minimized and used only in segments
in which there is a high density of VMs.
Diffuse VMs of the colorectum and surround-
ing pelvic structures can be left alone if bleeding
does not necessitate blood transfusions. When
bleeding is more severe, control may occasion-
ally be established with sclerotherapy. The defin-
itive surgical alternatives are to divert the fecal
stream by colostomy, or, preferably, to perform a
colectomy with endorectal mucosectomy and
coloanal endorectal pull-through. This procedure
entails a risk of pulmonary embolism because of
the manipulation of abnormal pelvic veins.
Elastic support stockings are indispensable in
the treatment of venous malformation of the
extremity. Additionally, low dose aspirin (81 mg/
day or every other day) helps to minimize
phlebothrombosis.
As with lymphatic malformations, sirolimus
has been used for extensive venous malforma-
tions. It can help with pain secondary to the
lesion as well as overlying skin breakdown,
though will only rarely visibly decrease the size.
It is also helpful in decreasing GI bleeding sec-
ondary to venous lesions of the GI tract. (McGuire
and Dickie 2020).
589
45.4.5 Arteriovenous
Malformations
Most often arteriovenous malformations are
latent during infancy and childhood and expand
during adolescence, manifesting as a warm, pink
patch in the skin and an underlying thrill or bruit.
Later, cutaneous consequences may include isch-
emic changes, ulceration, pain, and intermittent
bleeding (Fig.  45.7). The hormonal changes of
puberty or local trauma seem to trigger expan-
sion. Nonetheless, there are rare examples of an
AVM or AVF presenting at birth with life-­
threatening high-output heart failure. The natural
history of AVMs can be documented by a clinical
staging system introduced by Schobinger
(Table 45.2). Like other vascular malformations,
AVMs are best imaged by MRI and MRA.
Prompt embolization may be necessary in the
uncommon occurrence of postnatal congestive
heart failure caused by an AVF or AVM. Treatment
is rarely indicated during infancy or early child-
hood for a quiescent (stage I) AVM; however, the
child should be re-examined annually for signs of
expansion. Conventional dogma dictates that
intervention should be delayed until there are
Fig. 45.7 Ulceration associated with a lower leg
AVM. Such lesions can be quite painful and, due to their
high flow, present with difficult to control bleeding
590
Table 45.2  Schobinger clinical staging system for AVMs
Stage Description
I (Quiescence) Pink-bluish stain, warmth and
arteriovascular shunting by
continuous Doppler scanning or
20 MHz color Doppler Scanning
II (Expansion) Same as stage I plus enlargement,
pulsations, thrill, bruit, and
tortuous/tense veins
III (Destruction) Same as stage II plus either
dystrophic skin changes,
ulceration, bleeding, persistent
pain, or tissue necrosis
IV Same as stage III plus cardiac
(Decompensation) failure
symptoms or endangering signs (e.g., recurrent
ulceration refractory to treatment, pain, bleeding,
increased cardiac output, or Schobinger stage
III–IV); however, some advocate for early inter-
vention to potentially prevent or minimize
growth. Advances in the field of interventional
radiology have led to more treatment options,
including embolization and sclerotherapy. These
interventions have allowed for the delay and
potential avoidance of operative resection for
some patient groups. Embolization and the use of
coils is a delicate art based off the inflow and out-
flow patterns seen on angiography. Understanding
the drainage pattern is critical for choosing the
proper technique. Traditional teaching has been
to embolize from the arterial side; however, this
can lead to numerous complications. It limits
access for future treatments and can lead to
necrosis of the overlying skin. Additionally, the
AVM will continue to recruit flow from nearby
arteries to supply the malformation. These proce-
dures are best performed by providers with
expertise in interpreting angiography and per-
forming these complex procedures.
For operative resection, the usual strategy is
arterial embolization for the temporary occlusion
of the nidus 24–72  h prior to the operation.
Whenever possible, the lesion should be resected
completely. Extensive preoperative planning is
often necessary to determine the exact extent of
resection necessary. Intraoperative frozen sec-
tioning of the resection margins can be helpful,
but the most accurate way to determine whether
A. McGuire et al.
resection is complete is by observation of the pat-
tern of bleeding from the wound edges.
Unfortunately, many AVMs are not localized and
may permeate throughout the deep craniofacial
structures or the soft and/or skeletal tissues of an
extremity. In these instances, embolization is
usually palliative and surgical resection is rarely
indicated.
45.5 Combined (Eponymous)
Vascular Malformations
Combined (or complex) vascular malformations
are associated with the overgrowth of soft tissue
and the skeleton. Many are named after the phy-
sicians who are credited with the most memora-
ble description of the condition.
45.5.1 Slow-Flow Anomalies
Klippel–Trenaunay syndrome (KTS) and
Congenital Lipomatous Overgrowth-Vascular
malformation (CLOVES) are both disorders on
the PIK3CA-related overgrowth spectrum
(PROS). They are caused by activating, somatic
PIK3CA mutations. KTS is a well-described,
combined CLVM associated with soft tissue and
skeletal hypertrophy of a limb or trunk. Most
cases are unilateral, but it can be bilateral. The
capillary malformations are multiple and typi-
cally arranged in a geographic pattern over the
lateral side of the extremity or buttock (Fig. 45.8).
The CM component is macular in the newborn,
but becomes studded with lymphatic vesicles as
the child ages. Anomalous lateral veins become
prominent because of incompetent valves and
deep venous abnormalities. Lymphatic hypopla-
sia is present in more than 50% of patients with
associated lymphedema or isolated lymphatic
microcysts. Thrombophlebitis occurs in 20–45%
of patients and pulmonary embolism can occur.
CLOVES is also characterized by combined
slow-flow vascular anomalies, primarily abnor-
malities of the venous and lymphatic vessels. It is
notable for fatty overgrowth that can affect any
45  Vascular Anomalies
Fig. 45.8  Klippel–Trenaunay syndrome presenting in
the bilateral lower extremities of a young child. Note the
lateral CMs with vesicular studding. Medially, the bluish
hues of the venolymphatic component are apparent.
Lymphedema is present bilaterally, but is more marked in
the right lower extremity and associated with foot
deformity
region of the body. Patients tend to have over-
growth of limbs as well as truncal involvement.
Management of both diseases is aimed at
reducing morbidity, and requires a multidisci-
plinary team. The use of sirolimus has been revo-
lutionary for patients with this disease. It helps to
slow and reduce overgrowth, as well as improve
function. It is helpful for pain, leakage from ves-
icles, and skin breakdown. A direct PI3Kα inhibi-
tor, BYL719 (PIQRAY®, Cambridge, MA), has
been shown to be effective in severe cases of
PROS (Venot et al. 2018). It is currently available
for compassionate use. An AKT inhibitor is also
under investigation. Interventional therapy also
plays a role in management of these complex dis-
eases. This includes closure of large, ectatic veins
with central connection to reduce the risk of pul-
monary embolism. In selected patients, sclero-
591
therapy can be used to obliterate incompetent
superficial veins and to shrink focal VMs or lym-
phatic cysts. Debulking procedures can be very
effective in selected patients. Laser photocoagu-
lation or injection with doxycycline can provide
temporary control over intermittent lymphatic
oozing or bleeding from the lymphatic vesicles.
Children should be seen at least annually and
limb length followed by serial radiographs. If a
limb length discrepancy of greater than 1.5  cm
develops, a shoe-lift is prescribed to prevent
limping and secondary scoliosis. Elastic com-
pression stockings are recommended. Grotesque
enlargement of the foot requires selective abla-
tive procedures (i.e., ray, midfoot, or Syme ampu-
tation) to allow the child to wear proper
footwear.
Proteus syndrome is a sporadic vascular, skel-
etal, and soft tissue disorder of asymmetric over-
growth and gigantism. Its salient clinical features
include verrucous nevus, lipomas and lipomato-
sis, macrocephaly, asymmetric limbs with partial
gigantism of the hands and feet or both, and cere-
briform plantar thickening (“moccasin” feet). It
has been suggested that Joseph Merrick, the
unfortunately entitled “Elephant Man” suffered
from Proteus syndrome.
Maffucci syndrome, a variant of Ollier’s dis-
ease, is a low-flow malformation of exophytic vas-
cular anomalies associated with multiple
enchondromas and bony exostoses. The children
appear normal at birth with osseous lesions becom-
ing manifest as small nodules on a finger or toe
during childhood and the vascular lesions appear-
ing later. The vascular anomalies are predomi-
nantly venous and occur in the subcutaneous tissue
and bones of the extremities. Early recognition is
important as greater than half of these patients
develop spindle cell tumors with potential for
malignant transformation into chondrosarcoma.
45.5.2 Fast-Flow Anomalies
The Parkes Weber syndrome shares many similari-
ties with Klippel–Trenaunay syndrome, but should
be distinguished by a component of an additional
592
capillary-arteriovenous malformation/fistula. The
lesions are obvious at birth and the involved,
asymmetrically enlarged limb is covered by a geo-
graphic pink, warm, macular stain with an under-
lying bruit or thrill. There may be associated
lymphatic abnormalities. Large lesions may be
associated with high output cardiac failure. MRI/
MRA, complemented by angiography, is impor-
tant for diagnosis and anatomic delineation.
45.6 Conclusions
The last decade has witnessed remarkable forays
into understanding the pathogenesis of vascular
anomalies. Improved definitions based upon this
genetic–anatomic–histologic classification have
allowed the development of multidisciplinary
approaches toward disease treatment and manage-
ment. As our appreciation of the embryonic and
developmental contributions to disease increases,
so does our ability to develop novel strategies for
management of previously insurmountably com-
plex lesions. Molecular and pharmacologic manip-
ulation of vascular anomalies holds great promise.
References
Anderson KR, Schoch JJ, Lohse CM, Hand JL, Davis
DM, Tollefson MM (2016) Increasing incidence of
infantile hemangiomas (IH) over the past 35 years:
correlation with decreasing gestational age at birth
and birth weight. J Am Acad Dermatol 74(1):120–126
Connell F, Brice G, Mortimer P (2008) Phenotypic char-
acterization of primary lymphedema. Ann N Y Acad
Sci 1131(1):140–146
Girard C, Bigorre M, Guillot B, Bessis D (2006) PELVIS
Syndrome. Arch Dermatol (1960) 142(7):884–888
Golabi M, An AC, Lopez C, Lee L, Kwong M, Hall BD
(2014) A new case of a LUMBAR syndrome. Am J
Med Genet A 164(1):204–207
A. McGuire et al.
Greenberger S, Bischoff J (2013) Pathogenesis of infan-
tile haemangioma. Br J Dermatol 169(1):12–19
Hartemink DA, Chiu YE, Drolet BA, Kerschner JE
(2008) PHACES syndrome: a review. Int J Pediatr
Otorhinolaryngol 73(2):181–187
Jinnin M, Medici D, Park L, Limaye N, Liu Y, Boscolo
E et al (2008) Suppressed NFAT-dependent VEGFR1
expression and constitutive VEGFR2 signaling in
infantile hemangioma. Nat Med 14(11):1236–1246
Léauté-Labrèze C, la Roque D, Dumas E, Hubiche T,
Boralevi F, Thambo J-B, Taïeb A (2008) Propranolol
for severe Hemangiomas of infancy. N Engl J Med
358(24):2649–2651
McGuire A, Dickie BH (2020) Colorectal Vascular
Anomalies. Eur J Pediatr Surg 30(5):406–412
McGuire A, Fernandez-Pineda I, Fishman SJ, Dickie
BH (2020) Pediatric hepatic vascular tumors. Semin
Pediatr Surg 29(5):150970
Mulliken J, Young A (1988) Vascular birthmarks:
Hemangiomas and malformations. Saunders,
Philadelphia
Pawlikowska L, Nelson J, Guo DE, McCulloch CE,
Lawton MT, Young WL et  al (2015) The ACVRL1
c.314—35A>G polymorphism is associated with
organ vascular malformations in hereditary hemor-
rhagic telangiectasia patients with ENG mutations, but
not in patients with ACVRL1 mutations. Am J Med
Genet A 167(6):1262–1267
Shirley MD, Tang H, Gallione CJ, Baugher JD, Frelin
LP, Cohen B et  al (2013) Sturge–Weber syndrome
and port-wine stains caused by somatic mutation in
GNAQ. N Engl J Med 368(21):1971–1979
Singh S, Bhari N, Jassi R (2019) Kasabach–Merritt
phenomenon. Indian J Paediatr Dermatol 20(4):
295–301
Venot Q, Blanc T, Rabia SH, Berteloot L, Ladraa S,
Duong J-P et  al (2018) Targeted therapy in patients
with PIK3CA-related overgrowth syndrome. Nature
(London) 558(7711):540–546
Wiegand S, Wichmann G, Dietz A (2018) Treatment of
lymphatic malformations with the mTOR inhibitor
sirolimus: a systematic review. Lymphat Res Biol
16(4):330–339
Zhang L, Yuan W-E, Zheng J-W (2016) Pharmacological
therapies for infantile hemangiomas: a clinical study
in 853 consecutive patients using a standard treatment
algorithm. Sci Rep 6(1):21670
 Congenital Nevi
Hannes Prescher, Julia F. Corcoran,
and Bruce S. Bauer
46.1 Introduction
Congenital melanocytic nevi (CMN) are com-
posed of clusters of nevo-melanocytes that are
generally present at birth but occasionally arise
as late as several years. These lesions arise from
melanocytic stem cells that migrate from the neu-
ral crest to the embryonic dermis and upward into
the epidermis. They may also migrate into the
leptomeninges.
Although the bulk of these lesions are small
and benign, some cover large portions of the
body or can be in conspicuous locations, present-
ing challenging reconstructive problems.
Furthermore, their potential for malignant degen-
eration causes anxiety for the parent, primary
care physician, and surgeon alike. Although small
pigmented nevi are present in 1 out of 100 births,
large nevi are present in only 1 in 20,000 births,
H. Prescher
University of Chicago Medicine & Biological
Sciences, Chicago, IL, USA
e-mail: hannes.prescher@uchospitals.edu
J. F. Corcoran
Section of Plastic and Reconstructive Surgery,
University of Illinois Chicago, Chicago, IL, USA
e-mail: juliav@mac.com
B. S. Bauer (*)
Section of Plastic and Reconstructive Surgery,
University of Chicago, Pritzker School of Medicine,
Chicago, IL, USA
e-mail: bbauer@surgery.bsd.uchicago.edu
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_46
46
and the giant lesions are even less common. As a
result, most surgeons have little experience with
them and little opportunity to develop a rational
protocol for their treatment.
The goal of this chapter is to classify the more
common cutaneous lesions, review the patho-
physiology and natural history, summarize the
risk of malignant degeneration, and provide a
rational approach to treatment.
46.2 Congenital Nevi
Congenital nevi are those cutaneous lesions
apparent at birth or that become apparent prior to
1  year of age. The word nevus is a “generic”
term, best defined as a hamartoma that is an over-
growth of mature cells normally present in the
affected part, but with disorganization and often
with one element predominating. This broad defi-
nition applies to a variety of cutaneous lesions
that can be congenital or acquired.
The majority of congenital lesions are mela-
nocytic in nature, including common congenital
melanocytic nevi, nevi of Ota, nevi of Ito, nevus
spilus, café au lait spots, and Mongolian spots.
Other non-melanotic lesions, such as sebaceous
nevi (of Jadassohn), neural nevi, and epidermal
nevi, can be evident at birth. Several other nevi
have a propensity to appear in childhood and will
be discussed here, including intradermal nevi,
blue nevi, and Spitz nevi.
593
594
46.2.1 Congenital Melanocytic Nevi
(CMN)
It is important to have a frame of reference in dis-
cussing the treatment of CMN.  Multiple defini-
tions have been used, and without some
uniformity, it is difficult to compare different
studies. A recent expert consensus-based scheme
for categorization of the cutaneous features of
CMN takes into account the size and localization
of the nevus, the number of satellite nevi, and
additional morphological characteristics.
In this proposed classification system, small
nevi are those measuring 1.5 cm or less, medium
nevi are subcategorized into M1 (1.5–10 cm) and
M2 (10–20 cm) lesions, and large nevi are classi-
fied as L1 (20–30 cm) and L2 (30–40 cm). Giant
nevi are subcategorized as G1 (40–60  cm) and
G2 (>60 cm) lesions. CMN localization is subdi-
vided into head, trunk, and extremities. The num-
ber of satellite lesions can vary and is classed as
S0 (no satellites), S1 (<20 satellites), S2 (20–50
a b
Fig. 46.1 (a, b) This infant female was born with a giant
congenital melanocytic nevus covering large portions of
her trunk with extension to the lower extremities. This
H. Prescher et al.
satellites), and S3 (>50 satellites). Additional
morphologic characteristics that should be noted
are color heterogeneity, surface rugosity, subcu-
taneous nodules, and hypertrichosis or “hairi-
ness” (Fig. 46.1).
Congenital melanocytic nevi are composed of
nevus cells of melanocytic origin, which vary in
the amount of pigment they carry. At birth, these
lesions can be quite faint. During the first
6 months of life, some nevi can appear to “grow”
significantly as tardive pigment becomes more
visible. Some satellite nevi may become visible
for the first time over the first 2–3 years (tardive
CMN). After the first 6 months, the lesions grow
proportionally to the particular area of the body
involved. The diameter of the lesion grows by a
factor of 1.7 times in the head, 3.3  in the thigh
and leg, and 2.8  in the torso, arms, hands, and
feet. The large nevi are at least 6 cm in diameter
on the infant’s body and 9 cm on its head.
Embryologically, these lesions are ectopic
nests of nevus cells. Melanoblasts, the precursors
patient should be worked up for potential meningeal and
cerebral melanosis
46  Congenital Nevi
to melanocytes, migrate from the neural crest to
the skin, mucus membranes, eyes, mesentery,
chromaffin system, and meninges, where they
differentiate into dendritic melanocytes. When a
disturbance of this migration and differentiation
occurs, the result is an ectopic population of
nevus cells. Nevus cells are melanocytes that dif-
fer from ordinary melanocytes histologically by
being arranged in nests or clusters, having a
rounded rather than dendritic shape, and tending
to keep their pigment in their cytoplasm rather
than transferring it to surrounding keratinocytes.
Histologically, efforts have been made to
identify characteristics specific to CMN in con-
trast to nevi acquired later in life. A reliable
microscopic differentiation between the two
could help determine the true rate of melanoma
in association with these lesions. Nevus cells
when found within the eccrine ducts or glands,
follicular epithelium, and blood vessels are spe-
cific for congenital melanocytic nevus, but not all
CMN will demonstrate these findings. In large
congenital melanocytic nevi, nevus cells have
been found in underlying subcutaneous fat, fas-
cia, and musculature.
46.2.2 Small Congenital
Melanocytic Nevi
Most congenital pigmented nevi are small and
are excised easily in a single procedure. The life-
time risk for melanoma in these patients has been
quoted to occur in 4.9 out of 100 people when the
patient provides the history that the lesion is con-
genital and in 0.8–2.6 out of 100 people when
determined by histological criteria of findings
consistent with CMN in melanoma specimens.
Practically speaking, however, the risk of mela-
noma before puberty is nil, being quoted as 1 in
200,000 individuals. For this reason, many pedia-
tricians, pediatric dermatologists, and pediatric
surgeons defer the removal of these lesions to an
age when excision can be performed under local
anesthesia in the office, eliminating the risks
associated with general anesthesia. Clearly, some
lesions lie in cosmetically sensitive areas and for
the psychological benefit of the child should be
595
removed earlier, even if general anesthesia is
required. From a practical point of view, these
procedures are best done either before the child
starts toddling or just prior to school entrance.
The stage in between these two ages is fraught
with falls, scrapes, fear, and lack of patient coop-
eration. The experience is better for the patient,
parent, and surgeon alike by avoiding elective
nevus removal in the toddler. There is little ben-
efit to delaying surgery in those lesions, which,
because of their location, will likely require gen-
eral anesthesia at any age.
46.2.3 Large Congenital
Melanocytic Nevi
Two immediate concerns face the family of a
child with a large or giant nevus. The first is the
risk of the child developing melanoma, and the
second is the stigma of this very visible lesion
and how it will affect the child’s psychological
development. Early consultation with a pediatric
surgeon or pediatric dermatologist can help edu-
cate the family and decrease the stress of the situ-
ation by providing concise information about the
nature of the nevus, its natural history, and the
options for its management.
In the literature, the estimated risk of develop-
ing melanoma ranges from 2% to 31%. The dif-
ferent populations and numbers in these studies
explain the wide variance. In a retrospective
study, Quaba and Wallace examined patients with
CMN covering more than 2% of total body sur-
face and found the melanoma risk to be 8.5%
during the first 15 years of life. Sandsmark et al.
have quoted a risk of 6.7% in childhood.
Marghoob et  al. have quoted a lifetime risk of
4.5–9% for melanoma arising in large and giant
CMN, and more recently, Barbarot et al. cited a
melanoma risk of 2% in 2578 patients with large
CMN (>20  cm). Approximately 50% of the
malignancies that develop in large CMN do so in
the first 3  years of life, 60% by childhood, and
70% by puberty. Another important point is that
less than 0.5% of melanomas appear in preado-
lescent children, but 33% of those are thought to
arise from CMN.
596
There is an ongoing controversy regarding the
risk of melanoma transformation. In the Swedish
prospective trial, the risk of malignant transfor-
mation was reported as 0.2%. Other studies
reported only extracutaneous melanoma in
patients with CMN.  No studies convincingly
show that excision of large CMN effectively
reduces the rate of malignant transformation to
melanoma.
We have treated a patient who developed a
metastatic malignant melanoma 20 years after a
complete excision of a large CMN of the face.
Another factor that needs to be considered and
discussed with families is the issue of neurocuta-
neous melanosis (NCM). Recent reports have
demonstrated the association of nevus cells in the
leptomeninges in a percentage of children with
large nevi in an axial orientation or those with an
extensive number of satellite nevi (Fig.  46.1).
Although symptomatic NCM is characterized by
mental retardation, hydrocephalus, and seizures,
many children are asymptomatic. These children
can be identified by T1 shortening in MR imag-
ing. It has been reported that 23% of at-risk
patients had evidence of central nervous system
involvement (melanotic nests within the brain
and meninges) on MR imaging. Marghoob and
Dusza have seen this finding in only 3% of chil-
dren in the Nevus Outreach Registry of over 600
patients. The latter figure coincides with the
authors’ experience. Although the presence of a
lesion on MRI does not typically alter the deci-
sion to treat or not treat a child with a large or
giant nevus, the approach may be altered in cases
of symptomatic NCM.
The rationale for early treatment of large and
giant nevi has four components. These are (1) the
presence of the greatest risk for malignancy in
the first 3  years, (2) the elasticity and healing
capacity of the skin in the early years, (3) the
greater parent tolerance of surgery in this time,
and (4) the psychological benefit on the child.
Taking all this into account, and assuming that
the child is otherwise healthy, the authors begin
the treatment of the large and giant nevus by
6 months of age, in most cases, provided that they
H. Prescher et al.
have seen the child from early infancy. Although
many of the tissue-expansion procedures used in
the treatment of giant nevi can be applied to older
children and selected adults, the intolerance for
repeated procedures and the decreased elasticity
of the skin may make the excision of extensive
lesions impractical in older patients.
Patches of darker color and raised areas often
exist within large CMN. The areas can represent
neuroid nevus, which is a form of nevus with
melanocytes that appear to be like Schwann cells
histologically and with nerve organelles such as
Meissner’s and Pacinian corpuscles. The patches
also can represent areas of local proliferation but
do not necessarily behave in an aggressive man-
ner. Histological findings of low mitotic rate, lack
of necrosis, evidence of maturation in the cell
population, and lack of high grade nuclear atypia
are clues to a benign course. Sometimes, the best
description of these areas, however, is melano-
cytic tumor of uncertain potential. Unusual areas
such as these should be addressed earlier in the
course of reconstruction.
46.2.4 Other Congenital Nevi
46.2.4.1 Café Au Lait Macules
Café au lait macules are sharply demarcated
areas of light tan to brown pigmentation which
present in normal individuals or can be associ-
ated, when multiple, with syndromes such as
neurofibromatosis. Histologically, there is
increased pigment in macromelanosomes within
keratinocytes in the basal layer. These lesions are
benign. If they are in cosmetically sensitive areas,
laser ablation can be considered. Recurrence
after laser therapy is common, but successful
ablation has also been reported.
46.2.4.2 Nevus Spilus
Nevus spilus, also called speckled lentiginous
nevus, also has light tan to brown macules with
areas of speckling within it. The presence of the
“speckles” or freckles within it separates it clini-
cally from the café au lait macule. Histologically,
46  Congenital Nevi
there are both increased pigment within the kera-
tinocytes of the basal layer and an increased
number of melanocytes as well. The speckles can
be areas of freckling, congenital melanocytic
nevi, or blue nevi. Any suspicious areas within
the lesion can be excised for biopsy as a nevocel-
lular portion of the lesion may still carry a malig-
nant potential. If the entire defect is in a
cosmetically sensitive area, laser can be a suc-
cessful treatment option, or it can be removed
surgically.
46.2.4.3 Blue Nevus
Blue nevi are smooth, almost blue-black lesions,
which can be present at birth but are more likely
to appear during childhood and puberty.
Frequently they are found on the extremities or
the head. Females are affected more than males.
Two variants exist: common and cellular. The
common blue nevus is relatively small, <1  cm,
sharply demarcated, and dome-shaped. In this
benign lesion, the melanocytes are dendritic in
nature, within the dermis and possibly into the
subcutaneous tissue, but the epidermis is normal.
The cellular blue nevus tends to be larger, 1–3 cm,
has less regular borders, and is found frequently
in the lumbosacrum. Melanocytes can be spindle-­
shaped and found in aggregates admixed with
dendritic melanocytes. The lesions tend to be
wider at the surface than at the base. There are
reported cases of malignant degeneration within
cellular blue nevi. For this reason, removal of
blue nevi is recommended.
46.2.4.4 Spitz Nevi
Although not usually congenital, Spitz nevi
occur frequently in young children. They are
pink, raised, firm lesions that often are confused
with pyogenic granulomas because of the
appearance and history of rapid growth and
onset. On occasion, they are pigmented as well.
The original name for these lesions was “benign
juvenile melanoma,” and under the microscope,
the rather bizarre histology can be confusing if
the patient’s age and history are not supplied to
the pathologist. These lesions are not malignant,
597
but do grow rapidly and tend to recur aggres-
sively if not completely excised. A generous bor-
der of normal tissue (i.e., 3–4  mm) should be
excised along with the lesion to decrease the
chances of recurrence.
46.2.4.5 Mongolian Spots
Mongolian spots commonly appear as blue-gray
macular discoloration resembling a bruise over
the lumbosacral area of newborn infants, espe-
cially in darker-skinned individuals. On occa-
sion, they can appear in atypical locations such as
the upper thorax or extremities. Usually these
benign lesions regress spontaneously by the age
of 3–4  years, but can persist in unusual cases.
Histologically, widely scattered dendritic mela-
nocytes lie in the lower two thirds of the dermis.
No specific therapy is necessary; however, laser
treatments can obliterate persistent lesions.
46.2.4.6 Nevus of Ota/Nevus of Ito
The nevus of Ota and the nevus of Ito are macu-
lar, blue-gray field defects in the area of the first
and second branches of the trigeminal nerve or in
the scapular, deltoid and supraclavicular area,
respectively. The mucosae of the nose and mouth
and the sclera, retina, and conjunctiva can also be
involved in the nevus of Ota. These lesions are
field defects of dermal melanocytosis, like
Mongolian spots. Unlike Mongolian spots, these
lesions do not spontaneously regress and can
become hyperpigmented during puberty. Usually,
these lesions are present at birth, but may become
apparent around puberty, only rarely appearing
during childhood. They are more common in
females and more frequent in darker-skinned
individuals, being reported most frequently in
Indian and Asian populations. In 10% of the
cases, the nevus of Ota is bilateral, and these
cases are associated with extensive Mongolian
spots. Histologically, the dermis contains elon-
gated, dendritic melanocytes scattered among the
collagen bundles, mostly located in the upper
third of the reticular dermis; they can have raised
areas within them that are indistinguishable from
a blue nevus beneath the microscope. These
598
lesions are considered to be benign; however,
reports of malignant changes exist in a few cases,
with the tumors having the histologic appearance
of a malignant or cellular blue nevus.
Good results in the treatment of these nevi
have been obtained with the Q-switched ruby
laser, the Q-switched Alexandrite laser, and the
Q-switched Nd:YAG laser. Multiple treatments
are required with each of these modalities.
46.2.4.7 Sebaceous Nevi
The sebaceous nevus was described by Jadassohn
at the turn of the twentieth century. It presents as
a waxy, hairless, yellow-orange plague, usually
on the scalp, head, or neck (Fig.  46.2). It is a
hamartoma of sebaceous glands. The lesions tend
to become more verrucous, itchy, and excoriated
during puberty. Sebaceous nevus syndrome is the
combination of large sebaceous nevi of the scalp
and face associated with developmental delay,
seizures, and ophthalmologic and bony abnor-
malities. Removal is recommended for these
lesions because of a documented risk of malig-
nant degeneration, usually basal cell carcinoma.
For extensive lesions involving cosmetically sen-
sitive areas, some centers reported using CO2
laser ablation with good results. Complete surgi-
cal excision, however, remains the gold standard
for the treatment of these challenging nevi.
Fig. 46.2  A newborn boy with a large sebaceous nevus
covering portion of the left scalp, lateral forehead, lateral
temporal area, and the cheek. Clinically, these lesions are
significant for a chance of development of a basal cell car-
cinoma later in life
H. Prescher et al.
46.3 Treatment
of Congenital Nevi
The treatment of large and giant nevi is contro-
versial. Many feel that the risk of degeneration is
too low to warrant the unsightly scars or grafts
that may follow treatment. Others feel that, in the
presence of NCM, the greatest risk lies within the
central nervous system, so the excision of the
cutaneous lesion can only have limited benefits.
However, the appearance of these lesions clearly
produces a stigma with significant psychological
implications. Removal on this basis is often war-
ranted. The challenge for the surgeon involved in
treating these often complex lesions is to develop
treatment modalities that do not only accomplish
the excision of all or most of the nevus but also
lead to an optimal esthetic and functional
outcome.
Treatment choices include observation, derm-
abrasion or curettage, and staged excision and
reconstruction. Some giant nevi are so extensive
as to have no available “donor” tissue for recon-
struction. In other cases, the family situation or
lack of available resources may speak for a less
“aggressive” approach. The treating surgeon
should be well versed in the available treatment
options, honest about the potential risks and out-
comes of the various surgical modalities of treat-
ment, and able to present these to the family.
46.4 Dermabrasion, Curettage,
and Laser Treatment
Dermabrasion and curettage are both techniques
that have been applied in the neonatal period in
an effort to remove the more concentrated popu-
lation of nevus cells near the lesion’s surface. The
technique can be effective in reducing the overall
nevus “cell load” but cannot fully remove the
nevus, because of the well-known depth of nevus
cells in CMN. Although this treatment may result
in significant lightening of the color of the lesion,
it is quite common to see later “bleed-through” of
the deeper nevus, with gradual darkening and
reappearance of the lesion. This result may pres-
46  Congenital Nevi
ent a difficult treatment problem in visible areas
like the face, where other techniques for excision
may then be less tolerated.
The same issues arise in consideration of the
laser as a means of treating nevi. Many patients
request information about the use of the laser to
manage these lesions, hoping for removal with-
out scarring. Selective photothermolysis is appro-
priate in macular dermal melanocytosis such as
nevus of Ota or nevus of Ito, non-regressing
Mongolian spots, nevus spilus, and café au lait
macules in cosmetically sensitive locations.
These particular lesions have minimal thickness,
are not located in the epidermis, and are unlikely
to be malignant, which makes them ideally suited
for management with lasers. Treatment hinges on
the surgeon picking a laser of correct wavelength
and pulse width to allow selective destruction of
the melanocytes without damaging the overlying
epidermis and underlying adnexal structures.
Serial treatments are required. Inappropriate
selection of the wavelength or dosimetry can lead
to secondary scarring with laser treatments.
Temporary hyperpigmentation and hypopigmen-
tation can occur as well.
An increasing number of centers are treating
extensive facial sebaceous nevi with carbon diox-
ide laser, with greatly improved cosmetic appear-
ance. Unfortunately, there is no data in the
literature regarding the prevention of cancerous
transformation of these lesions with this treat-
ment modality.
Because large and giant congenital melano-
cytic nevi have nevus cells in all layers of the epi-
dermis, dermis, subcutaneous tissue, and
sometimes fascia and muscle, it is unrealistic to
think that any laser would be capable of eliminat-
ing the nevus without damaging, i.e., burning and
scarring, the tissue. Furthermore, because the
lesion is vaporized, there is no surgical specimen
for histologic confirmation of the benign or
malignant nature of the lesion. While it may
prove to be of use for reducing pigmentation in
sensitive facial areas (e.g., on the eyelids), it
would be expected to be of limited benefit and
likely to require repeated treatment over time.
Whether or not the radiant energy required for
laser treatment has a negative impact on the nevus
599
cells within the remaining lesion has yet to be
determined and may not be apparent for many
years into the future.
46.5 Methods of Excision of Small
and Intermediate Nevi
Smaller nevi can be excised with elliptical,
wedge, circular, or serial excision.
46.5.1 Elliptical Excision
Simple elliptical excision is the most commonly
used technique. Elliptical excision of inadequate
length may yield “dog ears,” which consist of
excess skin and subcutaneous fat at the end of the
closure. To prevent “dog ears,” the length of the
ellipse should be at least three times the width.
“Dog ears” do not disappear on their own.
46.5.2 Wedge Excision
Lesions located at or adjacent to free margins can
be excised by wedge excision. This applies to
lesions located on the helical rim, eyelids, and lips.
46.5.3 Circular Excision
When preservation of skin is desired or the length
of the scar must be kept to a minimum, circular
excision might be desirable. Circular defects can
be closed with full-thickness skin graft, local
flap, or a purse-string suture. A purse-string
suture causes significant bunching of the skin.
This is allowed to mature for many months and
may result in a shorter (but often a wider) scar.
46.5.4 Serial Excision
Serial excision is the excision of a lesion in more
than one stage. Serial excision is frequently
employed for the treatment of congenital nevi.
The inherent viscoelastic properties of skin are
600
used, allowing the skin to stretch over time. These
techniques enable wound closure to be accom-
plished with a shorter scar than if the original
lesion was elliptically excised in a single stage.
46.6 Overview of Current Surgical
Treatment of Large
and Giant Pigmented Nevi
As previously mentioned, the challenge in surgi-
cal treatment of large and giant nevi is to select a
treatment program that will allow complete exci-
sion and reconstruction at an early stage, mini-
mize scarring, and minimize the need for later
treatment. Surgical planning must satisfy these
requirements in order to provide an optimal func-
tional and esthetic outcome. The optimal choice
of treatment varies by body region, and the
remainder of this chapter summarizes the authors’
thoughts on these different treatment modalities.
46.6.1 Scalp
Tissue expansion is a treatment modality of choice
for excision and reconstruction of large and giant
nevi of the scalp. As surgical experience increases
a b
Fig. 46.3 (a) A 1-year-old girl with congenital large pig-
mented nevus of the left parietal-occipital scalp, with
extension into the auricle. (b) Expander in place in the
parietal-occipital scalp above the nevus. (c) Nevus exci-
H. Prescher et al.
and planning improves, larger defects can be
reconstructed with fewer procedures and better
restoration of normal hair patterns. Rectangular
expanders with soft bottoms and remote injection
ports are used, with the expanders in place for an
average of 10  weeks. Expanders are typically
injected weekly (increased to every 4–5  days in
some cases). The typical scalp expanders range
from 250 to 500 cc in size. Treatment starts with
patients as early as 8 months, with some cranial
molding expected by the time the expanders are
removed, but with no instance of long-term cra-
nial deformity noted (remodeling usually occurs
over 3–4 months) (Fig. 46.3).
Application of transposition flaps to the scalp
has dramatically improved our ability to cover
larger defects with more esthetically acceptable
restoration of a normal hair pattern and the fron-
totemporal hairline. The benefit of this modifica-
tion is most apparent with the use of the expanded
occipital transposition flap for covering the entire
parietal scalp and reconstruction of the temporal
hairline and sideburns. The nevus or the scar
must never be resected until the extent of flap
transposition is determined. If a complete exci-
sion cannot be accomplished, the remaining
nevus or scar is reconstructed after additional
expansion.
c
sion was followed by reconstruction with expanded trans-
position flap. Postoperative view demonstrates accurate
hairline and hair growth directions
46  Congenital Nevi
46.6.2 Face
Large and giant nevi of the face present some of
the greatest challenges in the treatment of these
lesions. These are the most visible nevi with which
the patient and family must deal and the ones that
are most likely to be associated with significant
psychological sequelae. They also represent the
area where unsightly scarring is most readily visi-
ble; consequently, the planning and execution of
the reconstructive plan must be very detailed. A
description of all the nuances of the treatment of
facial nevi is beyond the scope of this chapter.
What follows is the summary of the highlights.
a b
d e
Fig. 46.4 (a) Patient with an extensive sebaceous nevus
covering portion of the left scalp, lateral forehead, lateral
temporal area, and the left cheek. His reconstruction was
performed in two stages. (b) Tissue expander in place.
Note that both the hair-bearing scalp and the non-hair-­
bearing forehead are expanded. (c) Partial excision of the
nevus is followed by reconstruction with expanded flap,
601
Tissue expansion of the hemiforehead for uni-
lateral lesions or the bilateral or lateral forehead
for central lesions can very effectively treat even
extensive lesions. Because many of these nevi
involve the adjacent scalp, the combined “attack”
on both of these regions often facilitates the exci-
sion and lessens the number of stages required.
The planning of expansion and reconstruction for
nevi of the forehead must be directed at minimiz-
ing any possibility of distorting the eyebrow and
the normal distance from brow to hairline
(Fig. 46.4).
Nevi of the cheek are best reconstructed with
expanded or non-expanded postauricular flaps
c
f
advanced laterally. Postoperative view demonstrates accu-
rate hairline and hair growth directions and excellent con-
tour of the forehead without distortion of the eyebrow. (d)
In the second stage, tissue expander was placed in the
medial cheek. (e) Resurfacing of the cheek with medially
based expanded flap, transposed to reconstruct the cheek
esthetic unit. (f) Postoperative view
602
(the cheek and neck can be considered as a single
anatomic unit sharing similar hair-bearing char-
acteristics and relatively thin skin). Reconstruction
of the entire esthetic unit of the cheek may require
two or even three expansions. The use of a trans-
position flap significantly reduces the risk of
downward traction and distortion of the lower
eyelid, which are seen as common sequelae of
direct advancement of expanded flaps from below
the mandible to the cheek (Fig. 46.4).
Expanded or non-expanded full-thickness skin
grafts have been used effectively for excision and
a b
c e
Fig. 46.5 (a) Patient with a congenital melanocytic
nevus involving eyelids with extensions to the eyebrow
and the forehead, the cheek, and the nose bilaterally (areas
involved 1–8). (b, c) Tissue expanders were placed in the
contralateral forehead and the ipsilateral supraclavicular
areas. An expanded full-thickness skin graft was har-
vested from the supraclavicular area. (d) Immediate post-
operative view. The forehead and the nasal dorsum and
H. Prescher et al.
reconstruction of nevi of the periorbital and eyelid
areas and occasionally the nasal dorsum. A single,
large, expanded full-thickness graft from the supra-
clavicular area can reconstruct eyelids, canthus,
and the region between eyelid and brow, without
the multiple “seams” that follow use of multiple
smaller grafts (Fig. 46.5).
Extensive nevi of the central face (nose, lips,
chin) are some of the most challenging that we
have to deal with, and their treatment requires
both ingenuity and a solid grasp of plastic sur-
gery reconstructive techniques.
d
sidewalls were reconstructed with an expanded flap trans-
posed from the contralateral forehead. The eyelids were
resurfaced with a single-unit expanded full-thickness skin
graft split at the aperture. A small portion of the nevus was
left to mimic the eyebrow. (e) Postoperative views after
1 year. Note the thin rims of the nevus left deliberately at
the margins. These rims are excised in a second proce-
dure, combined with scar revisions
46  Congenital Nevi 603

46.6.3 Neck ers are placed adjacent to the lesion to be excised.

The posterior and the posterolateral neck are
commonly involved with large nevi. Posterior
and posterolateral neck defects can be success-
fully reconstructed with expanded flaps from the
upper back and the shoulders. The tissue expand-
The flaps are designed in such a way that they can
be wrapped around the neck, eliminating the
“webbing” created by pure upward advanced
flaps. With this design, the reconstructed neck
has a better contour and more favorable scar loca-
tion (Fig. 46.6).

a b

c d

Fig. 46.6 (a) A 2-year-old boy with a large congenital mela- 3  months of serial expansion. (c, d) Postoperative views.
nocytic nevus of the posterior neck, extending to the upper Excellent contour of the neck was achieved with expanded
back and the occipital scalp. (b) Expanders in place after flaps transposed upward and wrapped around the neck
604
46.6.4 Trunk
Some of the most significant strides have been
made in better understanding and applying tissue
expansion to the treatment of giant nevi of the
trunk. Better expanded flap design, and, when
regional expansion is not possible, using
expanded distant flaps with microvascular trans-
fer, has resulted in both functional and esthetic
outcomes where previously large grafted areas
diminished the outcome in both these aspects.
The most common location of giant nevi was
found to be over the posterior trunk, often extend-
ing anteriorly in a dermatome distribution.
Tissue expansion can be very effective on the
anterior trunk, provided that the lesion is confined
to either the lower abdomen or central abdomen
a b
d e
Fig. 46.7 (a) Patient with a bathing trunk nevus starting at
the junction of the middle and lower of the back and cover-
ing the entire buttocks, the perineum, and the left thigh cir-
cumferentially. (b, c) After expansion of the upper back,
two large medially based flaps are transposed to reconstruct
H. Prescher et al.
and that there is sufficient uninvolved skin above
or above and below the nevus to expand.
Expansion must be avoided in or around the area
of the breast bud in females, and lesions of the
breast should be left until after breast develop-
ment, regardless of the psychological implica-
tions of delaying the treatment until that age.
The use of expanded transposition flaps has
enabled excision of nevi of the upper back and
buttock/perineal region, where previously it was
thought that only skin grafting was possible.
Tissue expanders in the 500–750  cc range are
used most commonly in infants and young chil-
dren. Serial expansion with careful planning has
made possible the excision of progressively
larger nevi of the back and buttocks with excel-
lent outcomes (Fig. 46.7).
c
f
the lower trunk. (d, e) These flaps can be re-expanded to
allow for further nevus excision. (f) View after four rounds
of tissue expansion and near-total nevus excision (a small
rim of nevus was left around the anus to prevent scarring in
this area that may lead to incontinence of the sphincter)
46  Congenital Nevi
Another tool for reconstruction of giant nevi
of the upper back, shoulders, and neck has been
the expanded free transverse rectus abdominis
myocutaneous (TRAM) flap, which can be posi-
tioned in the upper back and posterior neck or
shoulder, then re-expanded, contoured, and
draped about the neck and shoulders.
46.6.5 Extremities
Large and giant nevi of the extremities present a
challenge that is still not fully met. Tissue expan-
sion has been of some help in treating smaller
lesions, where tissue is available both proximal
a b
e f
Fig. 46.8 (a). Case of a giant nevus of the arm treated
with an expanded pedicle flap from the trunk to the arm.
(b, c) The arm is positioned against the flank and abdomen
after expansion of the site. The forearm nevus is excised to
the fascia level, and the forearm is placed within the
expanded pedicle flap. The arm is placed for 3  weeks
within the expanded “tunnel,” and the pedicle is gradually
605
and distal to the lesion and the lesion is confined
to a fairly small segment of the limb. The geom-
etry of the extremity, as well as the limited flexi-
bility of the skin (particularly in the lower
extremity), makes regional expansion of limited
use.
In the past decade, the authors have begun to
find a way around these limitations, using large
expanded transposition flaps from the scapular
region to cover the upper arm and shoulder and
expanded pedicle flaps from the flank and abdo-
men for circumferential nevi from the elbow to
the wrist (Fig.  46.8). Expanded full-thickness
skin grafts have been used effectively for the dor-
sum of the hand with excellent esthetic outcome.
c d
g
tightened with through-and-through bolster sutures, grad-
ually reducing the blood flow through the pedicle. (d).
The pedicle is then divided and the flap inset. The border
of the remaining nevus is excised at a later time. (e–h).
Eight-year postoperative views of the forearm and the
abdominal donor site with excellent contour of the extrem-
ity and minimal donor site scarring
606
Although pedicle flaps are not readily avail-
able for coverage of more extensive lesions of the
arm, thigh, or leg, the authors have had some suc-
cess with expanded free flaps from the abdomen
and scapular region. These procedures have been
used only in very carefully selected cases, and the
optimum timing of these complex reconstructive
procedures is still under consideration.
46.7 Satellite Nevi
Satellite nevi may appear anywhere over the
course of the first few years of life, and their
number seems to correlate directly with the
­likelihood of NCM. They may vary in size from
small to medium lesions. To date, no case of
melanoma has been reported arising in a satel-
lite nevus. With this in mind, it is generally
agreed that the primary reason for excision of
satellite nevi is an esthetic one. The authors gen-
erally excise some of the larger lesions early,
often with serial excision, and leave the smaller
lesions until the child expresses specific con-
cerns about them. A significant benefit may also
result from excising multiple satellite nevi on
the face before the child enters his or her school
years.
46.8 Conclusions
Although the exact risk of malignant degenera-
tion may never be determined, there is still evi-
dence that large and giant congenital nevi carry
this potential. Excision and reconstruction are
warranted, provided that they can be a­ ccomplished
with an optimal esthetic and functional outcome.
The ability to present organized discussion of
current views of malignant change to parents,
patients (when old enough), and other allied
healthcare workers is critical. Experience with a
large population of children with large and giant
CMN has demonstrated that thoughtful applica-
tion of the full spectrum of reconstructive options,
H. Prescher et al.
heavily weighed toward the use of tissue expan-
sion (as well as expanded pedicle and free flaps),
can result in total or near-total excision of many
of these extensive nevi with predictably good
outcomes.
Further Reading
Bauer B, Dickie S (2021) Congenital Nevi. In: Puri P (ed)
Pediatric surgery. Springer, Cham
Bauer BS, Margulis A (2004) The expanded transposition
flap: shifting paradigms based on experience gained
from two decades of pediatric tissue expansion. Plast
Reconstr Surg 114:98–106
Bauer BS, Byun MY, Han H, Vicari FA (1997) A new
look into the treatment of congenital giant pigmented
nevi in infancy and childhood: a follow-up study and
review of 200 patients. Plast Surg Forum 20:76
Bauer BS, Few JW, Chavez CD, Galiano RD (2001) The
role of tissue expansion in the management of large
congenital pigmented nevi of the forehead in the pedi-
atric patient. Plast Reconstr Surg 107:668–675
Berg P, Lindelof B (2002) Congenital nevocytic nevi:
follow-up of a Swedish birth register sample regarding
etiologic factors, discomfort, and removal rate. Pediatr
Dermatol 19(4):293–297
Bruce SB, Julia C (2005) Treatment of large and giant
nevi. Clin Plastic Surg 32:11–18
Ceballos-Rodriguez MC, Redondo P, Tomas-Velazquez
A, Cieza-Dias D, Lopez-Guitierez JC (2021) Surgical
outcomes and psychosocial impact of giant congenital
melanocytic nevus surgery: a single-center case series
of 136 patients. J Pediatr Surg 56(11):2113–2117.
[Epub ahead of print]
Fahradyan A, Wolfswinkel EM, Tsuha M, Reinisch JF,
Magee W 3rd, Hammoudeh JA, Urata MM, Howell LK
(2019) Cosmetically challenging congenital melano-
cytic nevi. Ann Plast Surg 82(5S Suppl 4):S306–S309
Foster RD, Williams ML, Barkovich AJ et al (2001) Giant
con-genital melanocytic nevi: the significance of neu-
rocutaneous melanosis in neurologically asymptom-
atic children. Plast Reconstr Surg 107:933–941
Krengel S, Scope A, Dusza SW, Vonthein R, Marghoob
AA (2013) New recommendations for the categoriza-
tion of cutaneous features of congenital melanocytic
nevi. J Am Acad Dermatol 68(3):441–451
Marghoob AA, Dusza SW, Oliveria SO, Halpren AC
(2004) Number of satellite nevi as a correlate for
neurocutaneous melanocytosis in patients with
large congenital melanocytic nevi. Arch Derm 140:
171–175
Margulis A, Bauer BS, Fine NA (2004) Large and giant
congenital pigmented nevi of the upper extremity: an
46  Congenital Nevi
algorithm to surgical management. Ann Plast Surg
521:158–167
Merkel EA, Mohan LS, Shi K, Panah E, Zhang B, Gerami
P (2019) Paediatric melanoma: clinical update, genetic
basis, and advances in diagnosis. Lancet Child Adolesc
Health 3(9):646–654
Mustafa D, Blundell AR, Hawryluk EB (2020) Congenital
melanocytic nevi. Curr Opin Pediatr 32:491–497
Simons EA, Huang JT, Schmidt B (2017) Congenital
melanocytic nevi in young children: histopathologic
607
features and clinical outcomes. J Am Acad Dermatol
76(5):941–947
Vourc’h-Jourdain M, Martin L, Barbarot S (2013) Large
congenital melanocytic nevi: therapeutic management
and melanoma risk: a systematic review. J Am Acad
Dermatol 68(3):493-8.e1–493-8.e14
Watt AJ, Kotsis SV, Chung KC (2004) Risk of melanoma
arising in large melanocytic nevi: a systematic review.
Plast Reconstr Surg 113:1968–1974
 Lymphatic Malformations
47
Jeremy A. Goss, Mohammed Zamakhshary,
Jacob C. Langer, and Emily Christison-Lagay

47.1 Introduction macroscopic lymphatic malformation is a ballot-

table mass beneath normal skin, although superfi-
Lymphatic malformations constitute one subset of cial macroscopic lymphatic malformations may
a larger group of benign, congenital vascular cause the overlying skin to attain a blue or dusky
lesions which may affect any segment of the vas- hue. Microcystic malformations may present as
cular tree including arterial, venous, capillary, and vesicles that permeate the subcutaneous tissue
lymphatic vessels. These malformations result and muscle. Dermal lymphatic involvement may
from embryologic errors in lymphangiogenesis cause vesiculation and brawny edema. The inci-
and consist of cystic cavities lined by a single dence of lymphatic malformation is variable but is
layer of flattened epithelial cells and filled with estimated at 1 in 1750 live births. Approximately
clear or straw-colored eosinophilic and protein-­ 40–50% of lymphatic malformations are found in
rich fluid. Clinically, lymphatic malformations the neck and shoulder; however, they may also
range in presentation from small masses to large occur in the mediastinum, retroperitoneum, groin,
and sometimes debilitating, disfiguring, or inva- and other lymphatic rich regions.
sive lesions. The most common presentation of a Lymphatic malformations, in general, have an
infiltrative pattern of growth which may compli-
cate treatment. Morbidity includes bleeding,
J. A. Goss infection, lymphorrhea, obstruction of vital struc-
Division of Plastic and Reconstructive Surgery,
Department of Surgery, Yale University School of tures, disfigurement, and psychosocial distress.
Medicine, New Haven, CT, USA Patients with lymphatic disorders are best man-
e-mail: jeremy.goss@yale.edu aged in an interdisciplinary center focused on
M. Zamakhshary vascular anomalies. This chapter will provide an
Department of Pediatric Surgery, Alfaisal University, overview of the diagnosis and management of
Riyadh, Saudi Arabia these often difficult lesions.
J. C. Langer
Division of General and Thoracic Surgery, Hospital
for Sick Children and Department of Surgery, School
of Medicine Surgery, University of Toronto, 47.2 Historical Overview
Toronto, ON, Canada and Nomenclature
e-mail: jacob.langer@sickkids.ca
E. Christison-Lagay (*) Historically, there has been significant confusion
Division of Pediatric Surgery, Department of Surgery, regarding the nomenclature of vascular malfor-
Yale University, New Haven, CT, USA
mations in general and lymphatic malformations
e-mail: Emily.Christison-Lagay@yale.edu

© Springer Nature Switzerland AG 2023 609

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_47
610
in particular. “Cystic hygroma” and “lymphangi-
oma” are commonly used in the literature to
describe lymphatic malformations. These terms
should be abandoned, and the term lymphatic
malformation should be used instead. There are
several lymphatic malformation phenotypes,
each with its own natural history and treatment,
and mutations have been established for many
phenotypes over the last decade. Lesions include
microcystic, macrocystic, combined (microcys-
tic/macrocystic), primary lymphedema, Gorham-­
Stout disease (GSD), and generalized lymphatic
anomaly (GLA); lymphatic malformations can
also be associated with syndromic overgrowth
conditions such as congenital lipomatous over-
growth, vascular malformations, epidermal nevi,
and scoliosis (CLOVES) syndrome and Klippel-­
Trenaunay syndrome (KTS).
47.3 Etiopathogenesis
and Pathophysiology
Seminal observations by Sabin in the early twen-
tieth century described the development of the
lymphatic system beginning in the sixth week of
embryogenesis with the formation of five primi-
tive sacs: two in the neck, one in the retroperito-
neum, and two posterior to the sciatic veins.
Recent advances in understanding the genetic
code which guides lymphatic development impli-
cate a series of transcription factors including
Sox18, Coup-TPII, and Prox1  in initiating the
formation of lymphatic sacs from the anterior
cardinal vein. Prox1 maintains a role as “global
regulator” interacting with VEGFR3 in an auto-
regulatory feedback loop which determines lym-
phatic phenotype and facilitates separation from
the venous system. Activation of VEGFR3
requires VEGFC, which, in turn, depends upon
proteolytic processing by the matrix protein
CCBE1 and the metalloprotease ADAMTS3.
Gradients of VEGFC drive migration of lym-
phatic endothelial cells, a complex process that
also depends upon the distribution of neuropilin2
and LYVE-1 (Oliver et al. 2020).
Ongoing remodeling, proliferation, and matu-
ration of LECs with lymphatic sprouting and
J. A. Goss et al.
valve development are subject to a precise chore-
ography of growth factors and signaling mole-
cules which regulate cell polarity, sense shear
stress, and recruit muscle cells to stabilize and
strengthen lymphatics and promote lymphatic
contraction and return of lymph to the blood-
stream. Important regulators include PIK3CA,
Fat4, Rasip1, Foxc2, Orai1, Piezo1, Gata2,
Sema3a, as well as the expression of PDGFB.
Genetic discovery in vascular anomalies is an
emerging field with many clinical phenotypes
traceable to mutations in key genes of lymphan-
giogenesis. Sporadic lymphatic malformations
(microcystic and macrocystic) as well as syn-
dromic lymphatic malformations (CLOVES and
KTS) are now known to have mutations in the
PIK3CA gene (Luks et al. 2015). Kinase inactivat-
ing mutations in the human VEGFR3 gene result
in Milroy’s disease, a form of primary lymph-
edema. Mutations in Sox18 are associated with
hypotrichosis-­lymphedema-­telangiectasia, a rare
disease with eyelid edema, absent eyelashes and
eyebrows, and abnormal peripheral veins (Irrthum
et al. 2003; Tartaglia et  al. 2001). Additionally,
several germline mutations have been established
in primary lymphedema (Irrthum et al. 2000).
Mutations in FOXC2 cause the autosomal domi-
nant lymphedema-distichiasis syndrome (Meige
disease) characterized by distichiasis (a double
row of eyelashes) and bilateral lower limb lymph-
edema at puberty, thought to be a consequence of
abnormal valve development (Fang et al. 2000).
Homozygous and compound heterozygous muta-
tions in CCBE1, FAT4, ADAMTS3, and FBXL7
cause Hennekam lymphangiectasia-lymphedema
syndrome clinically manifest as severe peripheral
lymphedema, intestinal lymphangiectasias,
hydrops fetalis, growth restriction, and mental
retardation (Alders et al. 2009). Mutations in
EPHB4 and CALCRL have been associated with
lymphatic related hydrops fetalis. Additional
mutations in GJC2, GATA2, PTPN14, PTPN11/
SOS1, ITGA9, REELIN, and PIEZO1 have also
been implicated in the development of lymphatic
malformations (Table 47.1) (Roberts et al. 2007;
Tartaglia et al. 2001; Oliver et al. 2020). As our
capacity to deliver targeted therapy grows, these
genes each represent potential targets for pharma-
47  Lymphatic Malformations 611

Table 47.1  Diseases of lymphangiogenesis and associated genes

Disease and clinical phenotype Genes
• 
Klippel-Trenaunay: extremity capillary-lymphatic-venous malformation characterized by soft PIK3CA
tissue and skeletal overgrowth. Lymphatic abnormalities are typically microcystic changes of the
abdominal wall, buttock and extremities, as well as pelvic and thigh microcysts.
• 
CLOVES (Congenital Lipomatous Overgrowth, Vascular malformations, Epidermal nevi, and
Scoliosis Syndrome): an overgrowth syndrome characterized by slow flowing vascular
malformations (lymphatic, venous, capillary), lipomatous masses, hand/foot anomalies
(overgrowth, macrodactyly, first web-space gap), arteriovenous malformation, scoliosis
• 
Isolated macrocystic and microcystic lymphatic malformations
Milroy’s Disease: congenital bilateral lower limb lymphedema VEGFR3
Hypotrichosis-lymphedema-­telangiectasia syndrome: eyelid edema, absent eyelashes and eyebrows, SOX18
abnormal peripheral veins
Meige Disease: bilateral lower limb lymphedema at puberty, distichiasis FOXC2
Hennekam lymphangiectasia-­lymphedema syndrome Type 1: severe peripheral edema with intestinal CCBE1,
lymphangiectasis, hydrops fetalis, growth restriction, mental retardation, flattened nasal bridge and FAT4,
hypertelorism ADAMTS3,
FBXL7
Lymphatic related fetal hydrops EPHB4,
CALCRL
Late-onset lymphedema GJC2
Emberger syndrome: primary lymphedema with myelodysplasia progressing to acute myeloid GATA2
leukemia
Choanal atresia and lymphedema PTPN14
Primary lymphedema ITGA9,
REELIN
Generalized lymphatic dysplasia: widespread edema, systemic involvement including intestinal and PIEZO1
pulmonary lymphangiectasia, pleural effusions, chylothorax, and pericardial effusions

cotherapy. Moreover, as lymphatic malformations Table 47.2  Syndromes commonly associated with pre-
exhibit dynamic behavior, with growth often natally diagnosed LM
accompanying the onset of adolescence, under- Non-aneuploides Aneuplodies
standing causal elements of progression is inte- Noonan’s Turner
Multiple pterygium Down
gral to treatment. Pubertal hormones have been
Achondrogenesis type-1 Edward
implicated in stimulating their growth and may Short-rib-polydactyly syndrome Patau
contribute to disrupted angiogenesis, lymphangi- Fryn’s
ogenesis, or dilation of vascular spaces Robert’s
(Kulungowski et al. 2012; Hassanein et al. 2012). Fetal alcohol
Some future treatments may target prevention of
progression rather than primary treatment of
disease. lent prognosis, prenatal sonography has revealed
a high “hidden mortality” among fetuses with
this condition. In contrast to the incidence of
47.4 Prenatal Diagnosis lymphatic malformation in live born infants (1 in
1750), the incidence in spontaneous abortions is
Lymphatic malformations are often first diag- much higher, estimated at 1  in 200. Prenatally
nosed by prenatal ultrasound in the late first tri- diagnosed lymphatic malformation may be a
mester. The natural history ranges from marker of aneuploidy (>50%, chiefly Turner or
progression to hydrops, to complete resolution. Down syndrome), syndromicity (most com-
Although most lymphatic malformations pre- monly Noonan syndrome), and other congenital
senting to the pediatric surgeon have an excel- structural anomalies (Table 47.2). This subgroup
612 J. A. Goss et al.

Fig. 47.1  Prenatal ultrasound of a fetus with Turner syn- b

drome and a large posterior lymphatic malformation.
Note the diffuse subcutaneous edema (E), which is indica-
tive of hydrops fetalis. H fetal head, C cystic hygroma
(lymphatic malformation)

is usually diagnosed early in pregnancy, is pre-

dominantly nuchal, and is often associated with
diffuse lymphatic abnormalities (Fig. 47.1).
The natural history of cervical lymphatic
lesions detected prenatally varies according to
the gestational age at which nuchal thickening
appears and is further influenced by the presence
or absence of hydrops or abnormal karyotyping.
Cases of isolated lymphatic malformation diag-
nosed in the first trimester typically have a good
prognosis with spontaneous resolution in the
majority. Conversely, those with hydrops and
abnormal karyotype have a poorer outlook. This
is in sharp contrast to isolated lymphatic malfor-
mation diagnosed late in pregnancy, which are
similar to those presenting to pediatric surgeons
postnatally. This “late” group is rarely associated
with aneuploidy (<1%), affects a variety of sites,
and almost never resolves.
Because of the high risk of associated syn-
dromes in patients with abnormal karyotype or
hydrops, parental counseling should take place in
a specialized center, where complete fetal inves-
tigation, including detailed ultrasound exam, is
available.
Prenatal diagnosis of a large cervical mass
should raise suspicion for potential airway
Fig. 47.2 (a) Prenatal sonogram showing a large pretra-
cheal lymphatic malformation. LM lymphatic malforma-
tion, F face. (b) The EXIT procedure, in which intubation
is accomplished prior to delivery while still on support
obstruction that may impact planning for deliv-
ery. While the majority of these patients may
still be orotracheally intubated in the delivery
room, others are best managed using the ex
utero intrapartum (EXIT) procedure to gain
access to the airway prior to dividing the
umbilical cord. A multidisciplinary team of
high-risk obstetricians, neonatologists, and
pediatric surgeons or otolaryngologists should
be present at delivery. In severe cases, a trache-
ostomy and/or partial resection can be done, or
the child can be placed on extracorporeal mem-
brane oxygenation as part of the EXIT proce-
dure (Fig. 47.2).
47  Lymphatic Malformations 613

Fig. 47.3  Infant with large cervical lymphatic malformation

Fig. 47.4  A 5-year-old with venolymphatic malformation of the right cheek

47.5 Diagnosis Symptoms caused by lymphatic malforma-

47.5.1 Clinical Features
Most lymphatic malformations are diagnosed
postnatally with more than half clinically evident
at birth and greater than 80% diagnosed by
5 years of age. Lesions vary widely in size, and
the most common anatomic location is the head
and neck (Figs. 47.3 and 47.4). Other areas often
affected include the axilla, mediastinum, groin,
and retroperitoneum. The predilection for these
sites is likely secondary to the rich lymphatics in
these locations.
tions are related to anatomical site, size, and the
presence or absence of complications.
Complications of lymphatic malformations
include infections which may be caused by bacte-
ria contaminating cystic spaces related to the
malformation or which travel to the malforma-
tion from distant sites through lymphatic chan-
nels. Such infections must be treated expeditiously
and aggressively, since they can progress to life-­
threatening sepsis. Hemorrhage into the cyst is
also a well-described complication and may be
spontaneous or related to minor trauma. Both
infection and trauma may account for sudden
614 J. A. Goss et al.

Fig. 47.5  Chest x-ray and CT from a child with persistent chylothorax due to a mediastinal lymphatic malformation

rapid increase in the size of the malformation and

may be the first presenting symptom. Less com-
monly, large malformations may cause airway
obstruction. Such complex lesions require
prompt treatment by an expert multidisciplinary
team. Rarely, chylothorax, chylopericardium, or
chylous ascites may complicate lymphatic mal-
formations (Fig. 47.5). In the absence of any of
the above complications, most lymphatic malfor-
mations are asymptomatic, and their effect on
cosmesis is related to their site and size.

Fig. 47.6  Lymphangioma circumscriptum: small vesi-

47.5.2 Microcystic Lymphatic cles overlying a microcystic lymphatic malformation
Malformation

Microcystic lymphatic malformations are typi-
fied by cysts measuring <5 mm, often accompa-
nied by pink-red overlying skin, with cutaneous
vesicles which leak lymph fluid and/or bleed.
These lesions are not amenable to sclerotherapy
as they are too small to be cannulated by a nee-
dle; microcystic disease has low likelihood for
cure unless it is amenable to gross total resec-
tion. Occasionally, following an apparent total
resection, small vesicles appear at the closure
site, sometimes termed lymphangioma circum-
scriptum (Fig. 47.6). When localized, these can
be managed with topical sirolimus.
47.5.3 Macrocystic Lymphatic
Malformation
Macrocystic lymphatic malformations contain
an abundance of large cysts (>5 mm) which can
be cannulated and respond well to sclerotherapy
or surgical excision (Fig.  47.7). Lesions often
47  Lymphatic Malformations 615

affect the neck or axilla and are soft and com-

pressible, and overlying skin can appear normal
but may have a blue hue or pink-red discolor-
ation with cutaneous vesicles. Facial lesions
contribute to macroglossia, poor oral hygiene,
and dental caries, whereas periorbital lymphatic
malformations may cause poor vision or, less
commonly, blindness (Greene et al. 2005).

47.5.4 Combined (Microcystic
and Macrocystic) Lymphatic
Malformation

In many cases, solitary lymphatic malforma-

tions are not purely macrocystic or microcystic.
Instead, lesions may contain both micro- and
macrocysts (Fig.  47.8). A preponderance of
microcysts portends a poorer prognosis for cure
Fig. 47.7  T2-weighted MR image of posterior suboc-
cipital macrocystic lymphatic malformation
(Table 47.2).

a b

Fig. 47.8 (a) Combined macrocystic and microcystic lymphatic malformation of the retroperitoneum. Microcystic
disease can be visualized infiltrating the psoas and transversus abdominis muscles in (b)
616
Fig. 47.9  Lymphedema in a 13-year-old girl. Diffuse left
lower extremity edema extending from the left knee to the
dorsal foot
47.5.5 Primary Lymphedema
Lymphedema is the result of aplastic or hypo-
plastic lymphatic channels and nodes and causes
chronic, painless, subcutaneous swelling as
interstitial lymphatic fluid fails to return to the
systemic circulation. As fluid accumulates, it
stimulates subcutaneous fibrosis and increased
adipose deposition (Brorson et al. 1998; Brorson
et al. 2009). Often males present during infancy
with bilateral lower extremity edema, and
females develop unilateral swelling in adoles-
cence (Fig.  47.9) (Schook et al. 2011). Adult-
onset primary lymphedema and idiopathic upper
extremity lymphedema, while rare, have been
described (Goss et al. 2019a, b). Secondary
lymphedema is not a vascular anomaly but
rather the effect of irreversible injury to lym-
phatic vessels and/or lymph nodes from radia-
tion, lymphadenectomy, parasitic infection, or
obesity (Maclellan et al. 2015; Greene et al.
2012).
Patients with primary lymphedema typically
have normal-appearing skin although cutane-
ous problems such as bleeding from vesicles,
hyperkeratosis, and lymphorrhea can occur.
Rarely, lymphangiosarcoma develops. There is
J. A. Goss et al.
an increased risk for infection of the area as the
lymphatic’s role in immunogenicity may be
impacted.
47.5.6 Gorham-Stout Disease
Gorham-Stout disease (GSD) or “vanishing bone
disease” is a progressive, multifocal osteolytic
lymphatic disorder which commonly affects the
ribs, cranium, clavicle, and cervical spine. The
average number of bones involved is 7, and when
multiple sites are affected, they are contiguous.
Over time, bone resorption can contribute to sig-
nificant pain and pathologic fractures for patients
with GSD.  Additionally, 95% of lesions have
associated abnormalities of the overlying soft tis-
sue, 42% of patients experience pleural effusions,
and 21% develop splenic and/or hepatic lesions
(Lala et al. 2013).
47.5.7 Generalized Lymphatic
Anomaly
Generalized lymphatic anomaly (GLA) is a dif-
fuse, multisystem disorder with significant bony
involvement of the appendicular skeleton (e.g.,
ribs, thoracic spine, humerus, and femur) in a
non-contiguous manner. The average number of
affected bones is 30. Soft tissue abnormalities are
seen in 56% of patients, 50% have a macrocystic
lymphatic malformations, 63% experience
splenic or hepatic lesions, and 50% develop pleu-
ral effusions (Lala et al. 2013).
47.5.8 Lymphatic Malformation-­
Associated Overgrowth
Syndromes
In addition to malformations with a lymphatic
component only, complex vascular malforma-
tions with components of lymphatic, venous, and
47  Lymphatic Malformations
arterial anomalies also exist. Frequently, these
are associated with the overgrowth of soft tissue
and the skeleton. Many are named after the phy-
sicians who are credited with the most memora-
ble description of the condition.
47.5.9 CLOVES Syndrome
CLOVES is a non-familial overgrowth syndrome
whose phenotype consists of lipomatous truncal
masses, overlying slow-flow vascular malforma-
tions (commonly a capillary malformation), and
hand/foot anomalies (overgrowth, macrodactyly,
first web-space “sandal gap” deformity)
Fig. 47.10  CLOVES syndrome in a 6-year-old boy. Note
the limb asymmetry with left arm and hand hypertrophy,
fatty tissue deposition in the upper arm, and capillary
malformation
a b
Fig. 47.11 (a) Klippel-Trenaunay syndrome in a new-
born presenting as a complex vascular malformation with
right lower extremity hypertrophy with malformed digits
617
(Fig.  47.10). Individuals may also present with
an arteriovenous malformation (AVM) (28%),
scoliosis (33%), or neurologic impairment (50%)
(Alomari et al. 2009). The lipomatous masses
typically are painful and infiltrate the retroperito-
neum, mediastinum, paraspinal muscles, and epi-
dural space.
47.5.10 Klippel-Trenaunay
Syndrome and Parkes
Weber Syndrome
Klippel-Trenaunay syndrome (KTS) is epony-
mous for a capillary-lymphatic-venous malfor-
mation of an extremity with significant soft tissue
and skeletal overgrowth (Fig.  47.11). Clinical
manifestations vary widely in KTS; whereas
some patients experience slightly enlarged
extremities, others may develop grotesquely
overgrown limbs with malformed digits. The
lymphatic abnormalities in KTS are typically
microcystic in the abdominal wall, buttock, and
extremities versus pelvic or thigh macrocysts. A
capillary malformation is often observed over the
extremity, buttock, or thorax and is macular in the
neonatal period but, with time, becomes studded
with hemolymphatic vesicles. Because of pelvic
involvement hematuria, bladder obstruction, cys-
titis, and hematochezia can arise. The Parkes
Weber syndrome shares many similarities with
Klippel-Trenaunay syndrome, but is distin-
guished by a component of an additional
capillary-­arteriovenous malformation.
and increased first web space. (b) MRI of the same infant
showing microcystic lymphatic malformation diffusely
infiltrating the abdominal wall and retroperitoneum
618
47.6 Imaging and Histopathology
Most lymphatic malformations are diagnosed by
history and physical examination alone, and
therefore, imaging studies are not required. US
can be advantageous to study simple macrocystic
lymphatic malformations in the pediatric popula-
tion as it may identify extent of the lesion or
intralesional hemorrhage without the need for
sedation. In more complex cases, magnetic reso-
nance imaging (MRI) is the most useful imaging
modality for lymphatic malformations, aiding in
the classification of macrocystic and microcystic
lesions as well as defining anatomic relationships
including involvement of important neurovascu-
lar structures. Macrocystic lymphatic malforma-
tions demonstrate hyperintense signal intensity in
T2-weighted and turbo-STIR images as well as
rim enhancement after contrast application.
Microcystic lesions have an intermediate signal
in T1 sequences and an intermediate to high sig-
nal on T2 sequences. The new generation of
ultra-fast MRI scanners has permitted this tech-
nique to be used more frequently in small infants
and even in affected fetuses.
Conventional contrast lymphangiography is
infrequently performed but may be of help in
establishing the location of a lymphatic or chy-
lous leak in a patient with a diffuse thoracic lym-
phatic anomaly. Lymphoscintigraphy is 96%
sensitive and 100% specific for lymphedema
(Hassanein et al. 2017). When lymphoscintigra-
phy is implemented, a radio-labeled tracer pro-
tein, Technetium-99, is injected into the web
spaces of the extremities and tracked as it courses
from local lymphatics channels, through to
regional nodes, and finally into the systemic cir-
culation. Abnormal findings include delayed
transit time (>45  minutes), dermal backflow
(accumulation of tracer in superficial/cutaneous
lymphatics), asymmetric node uptake, and/or
collateral lymphatic channels (Szuba et al. 2003;
Gloviczki et al. 1989).
Certain patients with syndromic lymphatic
malformations require surveillance for associated
abnormalities. Patients with CLOVES are predis-
posed to forming Wilms tumors and require rou-
tine ultrasound monitoring every 3 months until
J. A. Goss et al.
age 7 (Peterman et al. 2017). Additionally, MRI
may aid in patients with CLOVES or KTS to
identify phlebectasia or enlarged veins, both of
which may potentiate thromboembolic events
and may warrant consideration of antiplatelet
therapy or anticoagulation (Arnold et al. 2011).
Histologic confirmation of lymphatic malfor-
mation is rarely required. When obtained,
findings include abnormally walled vascular
­
spaces filled with eosinophils and lymphocytes
within a protein-rich fluid milieu. Staining with
D240, LYVE1, and PROX1 (lymphatic-specific
markers) can further elucidate lymphatic malfor-
mations from other vascular lesions when in
doubt. Histopathology is nonspecific in lym-
phatic malformation-­associated overgrowth syn-
dromes (Gupta et al. 2011).
47.7 Differential Diagnosis
The diagnosis of lymphatic malformation is usu-
ally straightforward and can be easily differenti-
ated from lymphadenopathy, teratomas, and other
solid tumors based on the clinical examination
and imaging studies. Lipomas may also be con-
fused with superficial lymphatic malformations,
but will not have a cystic appearance on ultra-
sound examination, and will have a high fat sig-
nal on MRI.  Hemangiomas and venous
malformations may be present in the same loca-
tion, but do not transilluminate and tend to col-
lapse on compression. There is also a cohort of
patients with combined features of lymphatic and
venous malformations within the same lesion.
These patients may be difficult to classify
preoperatively.
47.8 Management
In general, treatment of a lymphatic malforma-
tion is reserved for symptomatic lesions, espe-
cially those that cause pain, bleed or weep
lymphatic fluid, obstruct or destroy tissues, cause
significant disfigurement, and contribute to major
psychological distress. Large, asymptomatic
macrocystic lesions are exceptions to this rule
47  Lymphatic Malformations
and should be treated prophylactically (begin-
ning as early as 6  months of age) as they are
prone to bleeding and infection later in life.
Until recently, there were only anecdotal reports
of medical management successfully being used as
treatment for unresectable lymphatic malforma-
tion—with no agent demonstrating consistent effi-
cacy. Perhaps the most excitement has been
generated by sirolimus, an mTOR inhibitor which
blocks PI3K/AKT pathway of VEGF activation. In
a recent systematic review of 20 published studies,
60 of 71 patients (85%) with complex lymphatic
malformations treated with sirolimus experienced
partial remission of the disease (Wiegand et al.
2018).
The two principal modern strategies for treat-
ing lymphatic anomalies are sclerotherapy and
surgical resection. Historically, irradiation, inci-
sion and drainage, and thermo- or irritant sclero-
sis have all been advocated as nonsurgical
treatments. With the exception of sclerotherapy,
none have demonstrated reproducible success.
No carefully controlled study comparing the effi-
cacy of sclerotherapy to surgery has ever been
reported. Many studies suggest that sclerotherapy
and surgical therapy have similar results and the
treatment selected should reflect upon the
strengths and experience of the treating team, as
well as the specific details of the individual lesion
in terms of size, location, and risk profile.
47.8.1 Sclerotherapy
Sclerotherapy uses a variety of agents to induce
obliteration of the lymphatic lumen by chemical
destruction of the endothelium with subsequent
sclerosis/fibrosis. Success parallels the degree of
damage inflicted upon the endothelial and deeper
muscular and connective tissue layers. In general,
macrocystic lymphatic malformations are more
amenable to sclerotherapy than microcystic
lesions because it is possible to drain the entirety
of the cyst cavity and induce endothelial apposi-
tion prior to administration of a sclerosant. A
number of sclerosing agents have been described,
including doxycycline, OK-432 (Picibanil, a bac-
terial product derived from Streptococcus pyo-
619
genes widely used and available in Europe and
Japan, but not available in North America), etha-
nol, sodium tetradecyl sulfate (STS), and bleo-
mycin. A meta-analysis of five retrospective
series of children with macrocystic or mixed
lesions of the head and neck treated with doxycy-
cline (at a preferred concentration of 10 mg/ml)
found 84.2% to be successfully treated, with 20%
of those requiring only a single session.
Doxycycline is generally well tolerated with
minimal side effects. The most commonly
reported adverse reactions were cyst hemorrhage,
cellulitis, pain, and transient edema. Recently,
several authors have supported the use of bleo-
mycin sclerotherapy (at a concentration of 1 U/
ml) in microcystic lymphatic malformations. The
mechanism of action which allows it to be more
effective than other sclerosants on microcystic
disease is not completely understood but may
involve either disruption of tight junctions or
induction of endothelial-mesenchymal transition.
The majority of patients exhibit either a complete
(approximately 30–40%) or partial (50–60%)
response. Concerns about bleomycin-induced
pulmonary fibrosis have not been substantiated.
47.8.2 Principles of Surgical
Management
Many lymphatic malformations can be resected
without undue morbidity. Although pathological
studies have shown that microscopic residual
lymphatic malformation is often left behind,
recurrence is rare when total excision of the mass
is accomplished. The following surgical princi-
ples must guide resection:
1. Adequate exposure must be obtained.
2. Meticulous dissection must be used in order
to preserve vital structures, including nerves,
vessels, trachea, and esophagus.
3. Since this is a benign disease, it is not justifi-
able to sacrifice a vital structure in order to
completely excise the lesion.
4. Whenever possible, the lymphatic supply to
the lesion should be ligated to prevent postop-
erative accumulation of lymph. In the head
620
and neck region, the lymphatic supply to a
lymphatic malformation is usually not visible,
but it is possible that the microbipolar dissec-
tion technique may “weld” these channels
shut.
5. Closed suction drains should be left at the
operative site to prevent accumulation of lym-
phatic fluid.
The successful surgical management of com-
plex (often mixed or microcystic) lymphatic mal-
formation is often staged. In each resection, the
surgeon should focus on a defined anatomic
region, attempt to limit blood loss, perform as
thorough a dissection as possible, and be pre-
pared to operate as long as necessary. Even with
such an intensive approach to resection of com-
plex malformations, subsequence “recurrence” is
as high as 40% after an incomplete excision and
17% after a macroscopically complete excision.
47.8.3 Microcystic Lymphatic
Malformation
Microcystic lymphatic malformations are chal-
lenging to manage. They do not respond well to
sclerotherapy; the diffuse nature of the lesions
often leads to subtotal resections, and therefore
recurrence is common. For these reasons, asymp-
tomatic microcystic lymphatic malformations are
best observed, and intervention should be
reserved for problematic lesions or those that
threaten sensitive sites such as the face. Generally,
first-line therapy is resection, and focal lesions do
not require preoperative imaging. An MRI should
be obtained prior to resection of large, diffuse
lesions to determine disease burden and involve-
ment of vital structures. However, in some cases,
the full extent of involvement may not be fully
appreciated clinically or radiographically.
Lesions on the face and other cosmetically sen-
sitive areas should be excised with minimal mar-
gins to limit iatrogenic deformity. Subcutaneous
tissues can be cauterized at the periphery of the
lymphatic malformation without skin excision as
this will induce fibrosis which may destroy resid-
ual lymphatic malformation and reduce recur-
J. A. Goss et al.
rence. Wounds can heal secondarily, through local
tissue rearrangement, or with skin grafting.
Patients and families are counseled that recurrence
following resection is common and additional
debulking may be required in the future.
Diffuse cutaneous vesicles can be managed
with carbon dioxide laser, superficial sclerother-
apy, or cauterization. Intraoral lymphatic vesicles
are best managed with radiofrequency ablation
which causes less post-treatment edema and
reduces risk of airway obstruction (Grimmer et
al. 2006). Individuals with diffuse, problematic
microcystic lymphatic malformations that have
failed other interventions are candidates for oral
sirolimus pharmacotherapy (Hammill et al.
2011).
47.8.4 Macrocystic and Combined
(Microcystic and Macrocystic)
Lymphatic Malformation
Sclerotherapy is often the preferred treatment for
macrocystic and combined lymphatic malforma-
tions in anatomically accessible locations. While
under general anesthesia and with the aid of
ultrasonography or fluoroscopy, a sclerosant is
injected into the lesion to induce fibrosis and
shrink the lymphatic malformation. Sclerotherapy
usually is performed every 6 weeks until a post-­
treatment ultrasound shows resolution and/or
cysts which are too small to benefit from further
sclerotherapy.
While sclerotherapy does not remove the
lesion, it does significantly shrink the lymphatic
malformation and may alleviate symptoms. If
symptoms are not abated following sclerotherapy
and excision is required, extirpation is more eas-
ily facilitated because the lesion will be smaller
owing to sclerotherapy-induced fibrosis. As
lesions shrink following sclerotherapy, patients
may experience skin redundancy which may ulti-
mately benefit from excision. Previous concerns
about sclerotherapy making subsequent surgical
excision more difficult have largely been dis-
counted in the literature.
Intra-abdominal and mesenteric macroscopic
lymphatic malformations may be best treated by
47  Lymphatic Malformations 621

resection. When surgical excision is chosen,
either a laparotomy or a laparoscopic approach
can be used. In either case, resection should be
done using meticulous technique, avoiding injury
to important neurovascular structures or organs.
Although a complete resection is sometimes pos-
sible, often some of the lesion must be left behind.
Remaining cysts should be unroofed, since com-
plications such as postoperative ascites are rarely
seen.
that of absent or hypoplastic lymphatics. The
most common morbidity from vascularized
lymph node transfer is donor-site lymphedema.
Physiological procedures do not reliably restore
lymphatic flow and fail to resolve excess subcuta-
neous adipose tissue deposition.
47.8.6 Gorham-Stout Disease/
Generalized Lymphatic
Anomaly

47.8.5 Primary Lymphedema GSD and GLA respond favorably to subcutane-

ous interferon-alfa 2b (PEGylated interferon
In general, patients with primary lymphedema do alfa-2b (Peg-Intron) 1.5mcg/kg subcutaneous,
not experience significant morbidity and are best weekly) and intravenous bisphosphonates (zole-
managed conservatively with static compression dronate (Zometa) 0.05  mg/kg IV, every other
and maintenance of a normal body mass index as month). This treatment regimen helps to prevent
obesity can significantly worsen lymphedema. progressive bone loss and improves bone remin-
There are two categories of operations to treat eralization. Oral sirolimus (0.8  mg/m2 per dose
lymphedema: excisional vs physiological. twice daily, with pharmacokinetic-guided target
Excisional procedures remove excess subcutane- serum 12 hour trough level of 7 to 13 ng/mL) has
ous fibro-adipose tissue through either suction-­ also been found to be efficacious (Adams et al.
assisted lipectomy (liposuction) or staged skin/ 2016). Symptomatic areas of bone loss may
subcutaneous excision. Physiological operations require fracture stabilization and/or bone grafts.
are aimed at improving lymphatic flow through
either lymphatic-venous anastomosis or vascu-
larized lymph node transfers. 47.8.7 Lymphatic Malformation-­
Symptomatic patients who are refractory to Associated Overgrowth
conservative therapies should undergo lymphos- Syndromes
cintigraphy and MRI prior to attempting exci-
sional procedures so as to 1) confirm the diagnosis 47.8.7.1 Cloves
of lymphedema and 2) determine the severity of Management should be based on symptoms. The
fat deposition (Greene et al. 2016). If MRI illus- lymphatic component may respond to sclerother-
trates significant subcutaneous adipose tissue apy or resection, although recurrence is common
deposition, the patient may be a candidate for in this patient population. Patients may require
lipectomy. While not curative, lipectomy signifi- orthopedic intervention to correct a leg-length
cantly reduces the size of the limb, offering discrepancy or for amputations or soft tissue deb-
improved cosmesis, and may also improve lym- ulking procedures to fit shoes and facilitate
phatic function (Greene et al. 2016; Greene et al. ambulation. Individuals are also monitored with
2017). serial ultrasounds for Wilms tumor.
Lymphatic-venous anastomosis and vascular-
ized lymph node transfer may have a role in treat-
ing early secondary lymphedema following 47.8.8 Klippel-Trenaunay Syndrome
axillary or inguinal lymphadenectomy or radia-
tion (before fibro-adipose tissue has developed). Symptomatic macrocystic lymphatic lesions
Physiological procedures may not benefit patients respond favorably to sclerotherapy, whereas
with primary disease as their pathophysiology is bleeding cutaneous vesicles are amenable to car-
622
bon dioxide laser or resection. Patients with large
embryonal veins should be treated with sclero-
therapy, coiling, or endovascular laser to prevent
life-threatening thromboembolic events. Leg-­
length discrepancies are common and may neces-
sitate shoe-lifts to prevent limping or scoliosis
until such time as epiphysiodesis of the distal
femoral growth plate can be performed. Ray,
midfoot, or Syme amputations may be required
for some patients to permit the use of footwear.
Circumferential leg overgrowth can be improved
with staged skin and subcutaneous excision.
47.9 Conclusion
Lymphatic malformations are rare, challenging
lesions that require significant expertise. For
complex lesions, there is no single treatment con-
sensus, but a treatment algorithm must be estab-
lished based upon the experience and strengths of
a multidisciplinary managing team. Such a multi-
disciplinary approach is optimal for providing
excellent care and maximizing the chances for a
favorable outcome. With the current increasing
interest in non-operative treatment, standardized
well-organized comparative studies are needed to
help define the role of new agents and therapies
as treatment for primary and recurrent lymphatic
malformations.
Disclosures  The authors have no disclosures.
References
Adams DM, Trenor CC 3rd, Hammill AM, Vinks AA,
Patel MN et al (2016) Efficacy and safety of sirolimus
in the treatment of complicated vascular anomalies.
Pediatrics 137(2):2
Alders M, Hogan BM, Gjini E et al (2009) Mutations in
CCBE1 cause generalized lymph vessel dysplasia in
humans. Nat Genet 41:1272–1274
Alomari AI (2009) Characterization of a distinct syn-
drome that associates complex truncal overgrowth,
vascular, and acral anomalies: a descriptive study of
18 cases of CLOVES syndrome. Clin Dysmorphol
18(1):1–7
Arnold R, Chaudry G (2011) Diagnostic imaging of vas-
cular anomalies. Clin Plast Surg 38:21–29
J. A. Goss et al.
Brorson H, Svensson H (1998) Liposuction combined with
controlled compression therapy reduces arm lymph-
edema more effectively than controlled compression
therapy alone. Plast Reconstr Surg 102:1058–1067
Brorson H, Ohlin K, Olsson G, Karlsson MK (2009)
Breast cancer-related chronic arm lymphedema is
associated with excess adipose and muscle tissue.
Lymphat Res Biol 7:3–10
Christison-Lagay ER, Fishman SJ (2006) Vascular anom-
alies. Surg Clin North Am 86:393–425
Fang J, Dagenais SL, Erickson RP et al (2000) Mutations
in FOXC2 (MFH-1), a forkhead family transcrip-
tion factor, are responsible for the hereditary
lymphedema-­ distichiasis syndrome. Am J Hum
Genet 67:1382–1388
Gloviczki P, Calcagno D, Schirger A, Pairolero PC, Cherry
KJ, Hallett JW, Wahner HW (1989) Noninvasive
evaluation of the swollen extremity: experiences with
190 lymphoscintigraphic examinations. J Vasc Surg
9:683–689
Goss JA, Maclellan RA, Greene AK (2019a) Adult-onset
primary lymphedema: a clinical-lymphoscintigraphic
study of 26 patients. Lymphat Res Biol 17(6):620–623
Goss JA, Maclellan RA, Greene AK (2019b) Primary
lymphedema of the upper extremities: clinical and
lymphoscintigraphic features in 23 patients. Lymphat
Res Biol 17(1):40–44
Greene AK, Maclellan RA (2016) Operative treatment of
lymphedema using suction-assisted lipectomy. Ann
Plast Surg 77:337–340
Greene AK, Burrows PE, Smith L et al (2005) Periorbital
lymphatic malformation: clinical course and manage-
ment in 42 patients. Plast Reconstr Surg 115:22–30
Greene AK, Grant FD, Slavin SA (2012) Lower-extremity
lymphedema and elevated body-mass index. N Engl J
Med 366(22):2136–2137
Greene AK, Voss SD, Maclellan RA (2017) Liposuction
for swelling in patients with lymphedema. N Engl J
Med 377:1788–1789
Grimmer JF, Mulliken JB, Burrows PE, Rahbar R (2006)
Radiofrequency ablation of microcystic lymphatic
malformation in the oral cavity. Arch Otolaryngol
Head Neck Surg 132(11):1251–1256
Gupta A, Kozakewich H (2011) Histopathology of vascu-
lar anomalies. Clin Plast Surg 38:31–44
Hammill AM, Wentzel M, Gupta A, Nelson S, Lucky
A, Elluru R, Dasgupta R, Azizkhan RG, Adams DM
(2011) Sirolimus for the treatment of complicated
vascular anomalies in children. Pediatr Blood Cancer
57(6):1018–1024
Hassanein AH, Mulliken JB, Fishman SJ, Quatrano NA,
Zurakowski D, Greene AK (2012) Lymphatic malfor-
mation: risk of progression during childhood and ado-
lescence. J Craniofac Surg 23(1):149–152
Hassanein AH, Maclellan RA, Grant FD, Greene AK
(2017) Diagnostic accuracy of lymphoscintigraphy for
lymphedema and analysis of false-negative tests. Plast
Reconstr Surg Glob Open 5(7):e1396
Irrthum A, Karkkainen MJ, Devriendt K, Alitalo K,
Vikkula M (2000) Congenital hereditary lymphedema
47  Lymphatic Malformations
caused by a mutation that inactivates VEGFR3 tyro-
sine kinase. Am J Hum Genet 67:295–301
Irrthum A, Devriendt K, Chitayat D et al (2003) Mutations
in the transcription factor gene SOX18 underlie
recessive and dominant forms of hypotrichosis-­
lymphedema telangiectasia. Am J Hum Genet 72:
1470–1478
Kulungowski AM, Hassanein AH, Nosé V, Fishman
SJ, Mulliken JB, Upton J, Zurakowski D, DiVasta
AD, Greene AK (2012) Expression of androgen,
estrogen, progesterone, and growth hormone recep-
tors in vascular malformations. Plast Reconstr Surg
129(6):919e–924e
Lala S, Mulliken JB, Alomari AI, Fishman SJ,
Kozakewich HP, Chaudry G (2013) Gorham-stout
disease and generalized lymphatic anomaly-clinical,
radiologic, and histologic differentiation. Skelet
Radiol 42(7):917–924
Luks VL, Kamitaki N, Vivero MP et al (2015) Lymphatic
and other vascular malformative/overgrowth disorders
are caused by somatic mutations in PIK3CA. J Pediatr
166(1048–1054):e1041–e1045
Maclellan RA, Couto RA, Sullivan JE, Grant FD, Slavin
SA, Greene AK (2015) Management of Primary and
Secondary Lymphedema: analysis of 225 referrals to a
Center. Ann Plast Surg 75(2):197–200
Oliver G, Kipnis J, Randolph GJ, Harvey NL (2020)
The lymphatic vasculature in the 21st century: novel
functional roles in homeostasis and disease. Cell
182(2):270–296
623
Peterman CM, Fevurly RD, Alomari AI, Trenor CC III,
Adams DM, Vadeboncoeur S, Liang MG, Greene AK,
Mulliken JB, Fishman SJ (2017) Sonographic screen-
ing for Wilms tumour in children with CLOVES
syndrome. Pediatr Blood Cancer 64(12). https://doi.
org/10.1002/pbc.26684
Roberts AE, Araki T, Swanson KD, Montgomery KT,
Schiripo TA, Joshi VA, Li L, Yassin Y, Tamburino AM,
Neel BG, Kucherlapati RS (2007) Germline gain-of-­
function mutations in SOS1 cause Noonan syndrome.
Nat Genet 39:70–74
Schook CC, Mulliken JB, Fishman SJ, Grant F,
Zurakowski D, Greene AK (2011) Primary lymph-
edema: clinical features and management in 138 pedi-
atric patients. Plast Reconstr Surg 127:2419–2431
Szuba A, Shin WS, Strauss HW, Rockson S (2003) The
third circulation: radionuclide lymphoscintigraphy in
the evaluation of lymphedema. J Nucl Med 44:43–57
Tartaglia M, Mehler EL, Goldberg R, Zampino G, Brunner
HG, Kremer H, van der Burgt I, Crosby AH, Ion A,
Jeffery S, Kalidas K, Patton MA, Kucherlapati RS,
Gelb BD (2001) Mutations in PTPN11, encoding the
protein tyrosine phosphatase SHP-2, cause Noonan
syndrome. Nat Genet 29(4):465–468
Wiegand S, Wichmann G, Dietz A (2018) Treatment of
lymphatic malformations with the mTOR inhibi-
tor sirolimus: a systematic review. Lymph Res Biol
16(8):330–339
 Sacrococcygeal Teratoma
Thambipillai Sri Paran and Prem Puri
48.1 Introduction
Teratomas are uncommon tumours comprised of
mixed dermal elements derived from the three
germ cell layers. They attract attention because of
their bizarre histology and gross appearance.
Though germ cell tumours only contribute to 3%
of the overall tumours in children under 16 years
of age, sacrococcygeal teratomas (SCTs) remain
the commonest teratoma in children, with an
overall incidence of 1  in 40,000 live births.
Differences in the prevalence of sacrococcygeal
teratoma have been reported in population-based
studies from different countries, 1:10700 births
from Finland (Pauniaho et al. 2013) and 1:27000
live births from the UK (Swamy et al. 2008). This
may reflect true differences between populations,
but may also be explained by accurate nation-
wide registration of sacrococcygeal teratomas.
Sacrococcygeal teratomas are congenital and
arise from the coccygeal region and can grow
into the pelvis or externally into the caudal region
or into both these regions. Though the majority
are benign, malignancy is reported in some
T. S. Paran (*)
Children’s Health Ireland at Crumlin, Dublin, Ireland
and Trinity College Dublin and University College
Dublin, Dublin, Ireland
P. Puri
Department of Pediatric Surgery, Beacon Hospital,
and University College Dublin, Dublin, Ireland
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_48
48
resected specimens. However, the overall prog-
nosis is excellent, dependent on ease of surgical
resection, timing of diagnosis, and malignant
potential of the tumour.
48.2 Embryology and Pathology
Sacrococcygeal teratomas have tissues derived
from the ectoderm, mesoderm, and endoderm.
The commonly accepted Embryonic
(Blastomeric) Cell Theory suggests that these
teratomas arise from pluripotent cells from
Hensen’s node, which have eluded the influence
of the primary organiser during early embryonic
development. Willis defined a teratoma as a ‘true
tumour or neoplasm composed of multiple tis-
sues of kinds foreign to the part in which it
arises’. Extragonadal teratomas tend to occur in
midline structures, such as the anterior mediasti-
num, retroperitoneum, sacrococcygeal region,
and pineal gland. They are composed of mixed
dermal elements, derived from the three germ
cell layers, and are known to contain virtually
any tissue type including skin, teeth, endocrine
structures, central nervous tissue, respiratory
mucosa, and alimentary mucosa. The tumour
may comprise of cells that are totally benign
(mature teratoma) to those that are frankly malig-
nant. A third group of cells that appear malignant
(usually described as ‘immature’) can confuse
the diagnosis. For this reason, the diagnosis of
625
626 T. S. Paran and P. Puri

malignant sacrococcygeal teratoma can only be a significant pelvic component (‘hour-glass

made if there are distant metastases at the time of
diagnosis.
48.3 Altman’s Classification
Altman et al. have classified sacrococcygeal tera-
tomas into four groups in 1974 (Fig. 48.1). Type
I tumours are almost exclusively exterior with a
minimal pelvic component; type II tumours have
p­ attern’); type III tumours have a larger propor-
tion of intra-abdominal and intrapelvic compo-
nent than the external component; and type IV
tumours are exclusively presacral with no exter-
nal component. Furthermore, Altman et  al. also
reported that type I tumours were rarely malig-
nant (0% in the American Academy of Pediatrics
Surgical Section Survey), while 6% of type II and
20% of type III and 8% of type IV tumours were
malignant. It is also well documented that most

Fig. 48.1 Altman’s Type I Type II

classification of
sacrococcygeal teratoma
from types I to IV

Type III Type IV

48  Sacrococcygeal Teratoma
tumours that are benign at birth can become
malignant after about 2  months. It has been
widely reported that the proportion of malignant
teratomas rose with increasing age (Wang et al.
2017), with malignant transformation in adults
reaching 40–50% (Xiang-Ming et al. 2019).
48.4 Diagnosis
Most SCTs are benign, resectable after birth
without significant morbidity, and rarely associ-
ated with other congenital anomalies. Thus, there
is the potential to lead a normal life without dis-
ability. Advances in ultrasound have enabled
accurate early diagnosis of SCTs, even in the first
trimester of pregnancy. Sonography can establish
the type of the tumour according to Altman’s
classification and can also determine whether
SCTs are cystic, solid, or mixed. Poor prognosis
indicators are predominantly solid components,
hydrops, and diagnosis before 30 weeks of gesta-
tion (Grigore and Iliev 2014).
48.4.1 Antenatal Diagnosis
In contrast to the overall prognosis of a newborn
with sacrococcygeal teratoma, a foetus remains
at high risk for perinatal complications and death.
Close follow-up with repeated ultrasound scans
is necessary to monitor an increase in tumour
size, extension into adjacent structures, tumour
vascularity, and evidence of cardiac failure.
Neonatal death may result from the maternal
obstetric complications, tumour rupture during
labour, preterm labour from polyhydramnios,
dystocia, or high-output cardiac failure due to the
tumour size. Foetal hydrops is usually the result
of a vascular steal syndrome leading to the shunt-
ing of blood away from the placenta and causing
high-output cardiac failure, the earliest signs of
which are polyhydramnios and placental thicken-
ing leading inexorably to preterm labour. Foetal
hydrops and placentomegaly also have grave
627
complications for maternal health. This is known
as maternal ‘mirror’ syndrome, because the
mother suffers the same symptoms as the sick
foetus. She will become ill and have signs of pre-
eclampsia: water retention, high blood pressure,
and proteinuria.
Hedrick (Hedrick et  al. 2004) reported their
experience with 30 antenatally diagnosed sacro-
coccygeal teratomas which included 4 termina-
tions, 5 foetal demises, 7 neonatal deaths, and 14
survivors. In three of the five intrauterine foetal
deaths, high-output cardiac failure and hydrops
were evident. One child died for partial tumour
rupture with pericardial and pleural effusions,
while the cause of death is unknown in the fifth
foetus. The seven neonatal deaths were at a mean
of 28.6  weeks of gestational age due to tumour
rupture during labour in three, intraoperative
bleeding in two, severe oligohydramnios and pul-
monary hypoplasia in one, and cardiac complica-
tions following open foetal surgery in one.
Fourteen survivors were all born by caesarean
section at a mean of 36 weeks of gestational age
and were operated on the first or second day of
life.
Makin et al. (Makin et al. 2006) followed up
41 foetuses in their series of antenatally diag-
nosed sacrococcygeal teratoma, with 6 elective
terminations, 12 elective foetal interventions
with 50% survival, and 17 non-interventions with
94% survival. Foetal intervention for hydrops
resulted in 86% mortality.
48.4.2 Neonatal Diagnosis
The most common neonatal presentation is large
sacral mass that is obvious at birth (Fig.  48.2).
The mass may be in the midline or paramedian.
The overlying skin is usually normal, but there
may be evidence of a haemangioma or necrosis
with ulceration. When there is intrapelvic exten-
sion of tumour, the anus and vagina are pushed
anteriorly. Intrapelvic tumours are often not diag-
nosed until later in life when they may present
628
Fig. 48.2  Newborn with large SCT
with symptoms due to pressure on the rectum or
bladder. Associated congenital anomalies are
found in 20% of cases and include vertebral,
renal, cardiac, and gastrointestinal anomalies.
Infants presenting with malignant tumours usu-
ally present with a rapidly growing buttock mass.
In such cases, distant metastases are usually pres-
ent at diagnosis. Even in these children with
malignant sacrococcygeal teratoma, recent
advances in multimodal therapy have resulted in
survival rates as high as 80%.
48.4.3 Investigations
Although imaging features cannot accurately
predict the histologic subtypes of the tumours,
thorough evaluation of the imaging features can
help distinguish malignant tumours from benign
tumours, as well as tumour size and extension.
Plain X-ray of the pelvis and spine may show
T. S. Paran and P. Puri
Fig. 48.3  MRI scan showing a large type II tumour
calcification or structures such as teeth and
bones, or there may be a bony defect as in neural
tube defect. Abdominal ultrasonography is use-
ful to determine the size and consistency of any
pelvic or abdominal component. Computed
tomography (CT) scan and magnetic resonance
imaging (MRI) both will provide more details
regarding tumour size, associated urinary out-
flow obstruction, extension, and even vascular
anatomy, but will need sedation or even general
anaesthesia in these newborns (Figs.  48.3 and
48.4). If oil is instilled into the rectum during
the MRI examination while T1-weighted images
are gathered, then the oil can act as contrast
medium. MRI should clearly distinguish
between sacrococcygeal teratoma and anterior
meningocele and may be able to detect the occa-
sional extension of the tumour through the
sacral hiatus into the spinal canal. Alpha-
fetoprotein is elevated when there is a malignant
component to the tumour and is a useful serum
marker in detecting postoperative malignant
recurrence.
48  Sacrococcygeal Teratoma
Fig. 48.4  MRI scan in a 1-month-old female with bilateral hydronephrosis and large SCT (Altman’s type IV)
48.5 Differential Diagnosis
The most important differential diagnosis of a
large sacral mass in a neonate is an anterior
meningocele. While most sacrococcygeal terato-
mas are solid, meningoceles are cystic with a
sacral defect. Dillarad et al. reported that during
rectal examination of a meningocele, the anterior
fontanelle is seen to bulge. Lemire et al. (1971)
have produced a list of 20 different lesions that
can possibly enter into the differential diagnosis
including a sacrococcygeal chordoma.
48.6 Management
The perinatal management of sacrococcygeal tera-
toma requires an interdisciplinary approach
between obstetricians, neonatologists, and paediat-
ric surgeons. The high perinatal/neonatal mortality
and morbidity rates associated with sacrococcygeal
teratoma are attributed to preterm delivery and to
complications such as malignant invasion, haemor-
629
rhage into the tumour, obstruction of umbilical
flow, high-output cardiac failure, hydrops fetalis,
and bladder outlet obstruction (Tuladhar et  al.
2000). Some of these ­complications can be detected
prenatally and treated appropriately.
48.6.1 Preoperative Management
Newborns with SCT require stabilisation prior to
early surgical resection. Blood should be cross-­
matched, and adequate venous access and arterial
access for blood pressure monitoring should be
established. If the tumour is ruptured, a pressure
bandage may be necessary to stop the blood loss,
while emergency surgery is being planned.
If the surgery is being carried out within the
first 2 days of life, then bowel preparation is not
necessary. However, when surgery is delayed
beyond this period for various reasons, proper
bowel prep must be carried out. Wide-spectrum
antibiotic prophylaxis is recommended for
48–72 hours.
630
48.6.2 Surgery
Following general anaesthesia and catheterisa-
tion of the bladder, the patient is placed in prone
position with rolls across the anterior superior
iliac spines and the shoulders (Fig. 48.5). A chev-
ron incision with its apex over the lower sacrum
is made with preservation of as much of the nor-
mal skin as possible (Fig. 48.6). The incision is
deepened down to the fascia, with ligation of
large subcutaneous veins.
Fig. 48.5  Positioning of the child for surgery
Fig. 48.6  Chevron incision above the coccyx
T. S. Paran and P. Puri
The incision is deepened down to the coccyx
or even the lower sacrum in the midline, and the
edges of the sacrum are defined. A clamp is
passed across the sacrum at this level, keeping
the tips of the forceps against the ventral surface
of the bone to ensure that the forceps pass
between the sacrum and the underlying middle
sacral vessels, which are usually substantial ves-
sels, supplying the bulk of the blood supply to the
tumour. Following this, the sacrum is divided,
and the middle sacral vessels are ligated in conti-
nuity and divided.
Following this, the tumour is dissected out-
side the capsule from thinned out levator and
gluteus muscle. The pelvic component is dis-
placed anteriorly until its upper extent is
reached. The tumour is dissected out from the
pelvis and rolled inferiorly. At this point, the
upper end of the rectum can be identified by
Vaseline gauze pack (placed preoperatively) or
by passing a Hagar dilator. Once the rectum is
clearly defined, the tumour is dissected off the
rectum, the anal canal, and the inferior skin flap.
Following meticulous haemostasis, the pelvic
floor is reconstructed by suturing the central
portion of levator sling to the perichondrium of
the sacrum and muscle and facial reconstruction
along the midline. These initial fascial sutures
should determine the sitting of the anus. A drain
may be placed within the presacral space and
brought out through the subcutaneous tissue of
the buttock. Excess skin is trimmed, and the
skin is closed in two layers when possible with
absorbable sutures or in a single layer with
nylon sutures. The Vaseline pack is left within
the anus.
Abdominal exploration is indicated when
there is a large abdominal component of tumour
or when there is significant bleeding due to either
tumour rupture or vascular injury during the peri-
neal approach. A transverse infra-umbilical inci-
sion will allow access to aorta below the origin of
inferior mesenteric artery. The middle sacral ves-
sels could be ligated via the abdominal approach,
and the abdominal component of the tumour
mobilised safely. The abdominal wound is tem-
porarily closed with running 3/0 nylon suture,
and patient repositioned for tumour resection as
48  Sacrococcygeal Teratoma 631

described above. At the end of tumour removal, mon in patients with sacrococcygeal teratoma
the abdominal wound is closed in layers with than in healthy children. Dysfunctional outcome
absorbable sutures. was more prevalent in children with large and
immature teratomas. In another study, Hambraeus
et  al. (2019) evaluated the functional outcome
48.6.3 Postoperative Management and health-related quality of life in patients with
sacrococcygeal teratoma and found that although
The infant is nursed in prone position until the a considerable proportion of bowel and urinary
wound is healed. The Vaseline pack is removed tract dysfunction was found, the reported quality
after 24 h, if it is not passed spontaneously. The of life was good in a majority of the patients with
infant can be fed the same day, as long as there is SCT. Low gestational age was found to be a pre-
no damage to the rectum during the tumour resec- dictor for bowel and urinary tract dysfunction.
tion, once fully recovered from anaesthesia. The Whether these complications are a result of
urinary catheter is removed as soon as baby’s tumour mass compressing the pelvic nerves or a
condition is stable. The wound drain can also be secondary result of surgical damage to the pelvic
removed after 48 h. nerves is unknown. Renal ultrasound is neces-
Alpha-fetoprotein levels are measured soon sary to assess the kidneys when symptoms sug-
after surgery before discharge. The infant is then gestive of neuropathic bladder are encountered.
followed up with three monthly checks of alpha-­ In fact, routine ultrasound monitoring of the
fetoprotein and PR examinations for 1  year to renal tracts annually is recommended by some
detect any recurrence. authors. Urodynamics, anorectal manometry,
and sphincter-­mapping stimulation studies must
be performed as dictated by clinical symptoms.
48.7 Long-Term Follow-Up

Though a 5-year follow-up was recommended in 48.8 Prognosis

the past, the latest reports by Derikx et al. (2007)
and Gabra et  al. (2006) highlight the need to
monitor these children into their teenage years.
Derikx et  al. (2007) reported malignant recur-
rence in 4 out of 148 children, with age at recur-
rence ranging from 10  months to 10.6  years.
When alpha-fetoprotein is raised or when clinical
suspicion is raised, the CT or MRI scan must be
used to detect and localise recurrent disease.
Long-term bowel and bladder dysfunction in
children following sacrococcygeal tumour resec-
tion is also well recognised. Bowel dysfunction
varied from constipation (17%), soiling (13%) to
involuntary bowel movements (9%) in the report
by Derikx et al. (2007). They also reported a 31%
urinary incontinence within their group of chil-
dren. Gabra et al. (2006) have also reported such
high incidence of bowel and bladder dysfunction.
Hambraeus et al. (2018) reported that the uncon-
trolled voiding, difficulty in bladder emptying,
pyelonephritis, and constipation were more com-
In the absence of distant metastases and when the
tumour is resected completely with the coccyx,
the life expectancy is normal. When distant
metastases are present, the prognosis must be
guarded. However, with present multimodal
treatment, > 90% survival could be expected in
children with distant metastases. Yadav et  al.
(2020) reported 95% survival in their cohort of 8
children with yolk sac tumour (out of a total of
41) at 54-month follow-up.
Bowel and bladder dysfunction are common
following sacrococcygeal teratoma resection.
Clean intermittent catheterisation for neuropathic
bladder and bowel management programmes for
faecal incontinence can improve the quality of
life for these children. As for the cosmetic results
after reconstruction of the buttock region follow-
ing resection of large tumours (>500  cm3), fur-
ther referral to plastic reconstructive surgeons
may be necessary.
632
48.9 Conclusions
With complete surgical excision in the absence of
distant metastases, life expectancy in children
with sacrococcygeal teratomas is that of normal.
Depending on the size of the tumour, a significant
cohort will have bladder and bowel dysfunction
following surgical resection and need long-term
follow-up. As for malignant tumours with distant
metastases, neoadjuvant chemotherapy and com-
plete surgical resection with clear margins will
result in good prognosis. When surgical resection
is not possible, local proton radiation therapy can
be utilised.
References
Derikx JP, De Backer A, van de Schoot L, Aronson DC, de
Langen ZJ, van den Hoonaard TL, Bax NM, van der
Staak F, van Heurn LW (2007) Long-term functional
sequelae of Sacro-coccygeal teratoma: a national study
in the Netherlands. J Pediatr Surg 42(6):1122–1126
Gabra HO, Jesudason EC, McDowell HP, Pizer BL,
Losty PD (2006) Sacrococcygeal teratoma – a 25-year
experience in a UK regional center. J Pediatr Surg
41(9):1513–1516
Grigore M, Iliev G (2014) Diagnosis of sacrococcygeal
teratoma using two and three dimensional ultrasonog-
raphy: two cases reported and a literature review. Med
Ultrason 16(3):274–277
Hambraeus M, Hagander L, Stenström P, Arnbjörnsson E,
Börjesson A (2018) Long-term outcome of sacrococ-
cygeal teratoma: a controlled cohort study of urinary
tract and bowel dysfunction and prediction of poor
outcome. J Pediatr 198:131–136e2
Hambraeus M, Al-Mashhadi A, Wester T, Svensson P-J,
Stenström P, Lilja HE (2019) Functional outcome and
T. S. Paran and P. Puri
health-related quality of life in patients with sacro-
coccygeal teratoma  – a Swedish multicenter study. J
Pediatr Surg 54(8):1638–1643
Hedrick HL, Flake AW, Crombleholme TM, Howell
LJ, Johnson MP, Wilson RD, Adzick NS (2004)
Sacrococcygeal teratoma: prenatal assessment, fetal
intervention and outcome. J Pediatr Surg 39(3):319–323
Lemire RJ, Graham CB, Beckwith JB (1971) Skin-­
covered sacrococcygeal masses in infants and chil-
dren. J Pediar 79(6):948–954
Makin EC, Hyett J, Ade-Ajayi N, Patel S, Nicolaides K,
Davenport M (2006) Outcome of antenatally diag-
nosed Sacrococcygeal teratomas: single – center expe-
rience (1993–2004). J Pediatr Surg 41(2):388–393
Pauniaho S-L, Heikinheimo O, Vettenranta K, Salonen J,
Stefanovic V, Ritvanen A, Rintala R, Heikinheimo M
(2013) High prevalence of sacrococcygeal teratoma in
Finland – a nationwide population-based study. Acta
Pediatr Jun 102(6):e251–e256
Swamy R, Embleton N, Hale J (2008) Sacrococcygeal
teratoma over two decades: birth prevalence, prena-
tal diagnosis and clinical outcomes. Prenat Diagn
110:1048–1051
Tuladhar R, Patole SK, Whitehall JS (2000)
Sacrococcygeal teratoma in the perinatal period.
Postgrad Med J 76(902):754–759
Wang Y, Yeming W, Wang L, Yuan X, Jiang M, Li Y
(2017) Analysis of recurrent Sacrococcygeal tera-
toma in children: clinical features, relapse risks, and
anorectal functional sequelae. Med Sci Monit 23:
17–23
Xiang-Ming X, Zhao F, Cheng X-F, Zhong W-X, Liu J-P,
Jiang W-Q, Xiao-Kai Y, Lin J-J (2019) Adult sacro-
coccygeal teratoma: a retrospective study over eight
years at a single institution. J Zhejiang Univ Sci B
20(8):670–678
Yadav DK, Acharya SK, Bagga D, Jain V, Dhua A, Goel P
(2020) Sacrococcygeal teratoma: clinical characteris-
tics, management and long-term outcomes in prospec-
tive study from a tertiary care centre. J Indian Assoc
Pediatr Surg 25(1):15–21
 Neuroblastoma
Edward Kiely
49.1 Introduction
Neuroblastoma arises in primitive cells of the
sympathetic nervous system. Consequently, it is
found at sites where collections of sympathetic
cells are encountered—adrenal medulla and sym-
pathetic ganglia.
A wide clinical spectrum of tumour activity is
seen ranging from spontaneous regression to
rapid growth with widespread dissemination. A
small number of tumours progress from apparent
malignancy to a more mature phenotype, and in
some children, the tumour may exhibit prolonged
periods of arrested growth.
As these tumours are relatively rare, national
and international collaborative groups have been
formed to refine all aspects of management and
to evaluate outcomes—Societe Internationale
d’Oncologie Pediatrique (SIOP) and Children’s
Oncology Group (COG) being the largest of
these groups. The International Neuroblastoma
Risk Group (INRG) Classification System was
devised by these groups and is now in use. The
expectation is that it will allow direct comparison
between different management protocols in dif-
ferent countries.
E. Kiely (*)
Great Ormond Street Hospital for Children NHS
Foundation Trust, London, UK
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_49
49
49.2 Historical Overview
In 1864, Rudolf Virchow recognised the neural
origin of neuroblastoma (Virchow 1864). By the
early part of the twentieth century, there was gen-
eral agreement about this, and the name neuro-
blastoma was coined by James Homer Wright
(Wright 1910). The tumour was considered rare,
and by 1934, only 134 cases had been reported in
the literature (Blacklock 1934). It was recognised
at the time as the fourth most common malig-
nancy in children. The prognosis remained very
poor—all of Blacklock’s 18 patients died within
a few months.
The first reported successful resection was by
Dr. Willard Bartlett in St. Louis, Missouri, in
1916, and this patient was alive and well 15 years
later (Lehman 1932). Successful surgery and sur-
vival however remained unusual at that time.
Treatment with radiotherapy was first reported
in 1928 but was not successful. Some overall
improvement in survival was noted in 1940–10
survivors of 40 patients—after surgery with and
without radiotherapy (Farber 1940). In 1953,
Phillips reported 17% survival after treatment
with a combination of surgery, radiotherapy and
chemotherapy (Phillips 1953). Fifteen years later,
little had changed—Fortner reporting a 5-year
survival rate of 9.7% (Fortner et  al. 1968). An
early version of combination chemotherapy was
first reported in neuroblastoma treatment in 1965
633
634
(James et al. 1965), and this form of management,
combined with supportive measures, remains the
mainstay of treatment for most affected children
to this day.
A systematic and coherent approach to neuro-
blastoma management awaited the acceptance of
an agreed method of assessing tumour stage. The
Evans staging system (Evans et al. 1971) was the
most widely utilised and forms the basis of sub-
sequent modern equivalents. The availability of
cross-sectional and radio-isotope imaging has
also made for more accurate staging.
Tumour excision has always been considered
essential. The timing and extent of surgery have
been more controversial. Complete tumour
removal has traditionally been the goal but has
been impossible to standardise or assess.
49.3 Incidence
The age-adjusted incidence rate is about 10 per
million children and varies a little across differ-
ent populations. In the USA, about 1  in 7000
children will be diagnosed with neuroblastoma at
some time in childhood. It is the commonest
extra-cranial solid tumour in children and the
commonest cancer in infancy and comprises
about 10% of all childhood cancer. Neuroblastoma
is responsible for 15% of deaths of children from
malignant disease.
One third of these tumours are found in
infancy; a further 50% occur between the ages of
1 and 4 years and the remainder in older children
and adolescents. It is very rare in adults. There is
a slight male preponderance.
49.4 Aetiopathogenesis
Neuroblastoma is an embryonal tumour arising
from neural crest cells of the sympathetic nervous
system. The exact cell of origin is unclear as are
the mechanisms which convert benign neuroblasts
into malignant cells. The cells that form the sym-
pathetic chain and ganglia and the cells, which
give rise to chromaffin tissue, arise from two sepa-
rate lineages of neural crest cells. The chromaffin
E. Kiely
tissue of the adrenal medulla and organ of
Zuckerkandl arise from multi-potent Schwann cell
precursors migrating later than the sympathetic
chain precursors (Tsubota and Kadomatsu 2018;
Kastriti et al. 2019). It is not clear if the poorer
prognosis of adrenal neuroblastomas is related to
this fact. Finally, it remains unclear what propor-
tion of neuroblastomas are present at the time of
birth or if they can develop postnatally.
There is no single genetic abnormality com-
mon to all neuroblastomas. Whole genome
sequencing shows a low somatic mutation rate,
unlike the situation in adult cancers (Johnsen
et al. 2019; Cheung and Dyer 2018). A mutation
of the anaplastic lymphoma kinase (ALK) gene
is the most commonly found—but only in 7–10%
of all neuroblastomas. In familial neuroblastoma
(<2% of the total), ALK mutations are usually
encountered. Mutation of the PHOX2B gene is
known also to predispose to development of
familial neuroblastoma. This gene promotes neu-
ronal differentiation. Other mutations of this
gene are associated with Ondine’s curse (congen-
ital hypoventilation syndrome), as well as with
Hirschsprung’s disease.
Amplification of the MYCN gene on chromo-
some 2 (>10 copies) is found in about 20% of
neuroblastomas and is associated with more
aggressive tumour behaviour and a poorer
prognosis.
Chromosomal abnormalities are associated
with variability in clinical behaviour. In general,
whole chromosome gains are associated with
low-risk disease—triploid DNA content or
hyperploidy. Diploid DNA content with intra-
chromosomal rearrangements is associated with
high-risk disease. 1p deletion, 11q deletion and
17q gain are the most widely studied of these
abnormalities, all conferring a worse outlook.
The manner in which these abnormalities influ-
ence tumour behaviour is unknown at present.
Spontaneous regression occurs in many can-
cers but is seen more often in neuroblastoma than
in other tumours. The causes of such regression
are unknown at this stage. In a landmark paper in
1963, Beckwith and Perrin described adrenal
nodules found incidentally at autopsy in 0.4% of
infants aged less than 3  months (Beckwith and
49 Neuroblastoma
Perrtin 1963). These infants had died of unrelated
causes, predominantly cardiac malformations.
The nodules were indistinguishable from neuro-
blastoma on histology and demonstrated invasion
of surrounding tissue. This is many times the
expected incidence and implies that spontaneous
regression may be very common. Equally, it is
not clear how many of these infants would have
later developed clinical neuroblastomas.
49.5 Pathophysiology
The manner in which these tumours affect the chil-
dren is dependent on the biology of the tumour—
the presence of bony metastases may give rise to
bone pain, peri-orbital bruising and general mal-
aise. Even very large abdominal tumours may pro-
duce no symptoms. Thoracic tumours are usually
asymptomatic and often are found on chest X-rays
performed for other reasons.
In general, therefore, neuroblastomas are not
associated with specific symptoms. Overall, it is
often surprising how large and extensive the
tumours have become before the parents recog-
nise there is a problem.
A minority have noticeable sweating. In some,
compression of a renal artery results in hyperten-
sion. Rarely, a cervico-thoracic tumour will pro-
duce a Horner’s syndrome.
Pelvic tumours which become large interfere
with bladder and bowel function. A small subset
of good biology tumours present with opsomyoc-
lonus—the dancing eye syndrome. Tumours
within the spinal canal will eventually produce
symptoms depending on the severity of cord
compression. On rare occasions, an infant is born
with established paraplegia as a consequence of
antenatal cord compression.
49.6 Pathology
The macroscopic appearance of these tumours
varies from firm pale grey discreet masses to fri-
able haemorrhagic ill-defined tumours. Small cys-
tic lesions may be found on antenatal scanning.
635
Fig. 49.1  Histology from the tumour surface: round cells
with rosette formation. (Courtesy of Michael Höllwarth)
Histologically, firmer tumours are generally
composed of mature ganglion cells with abun-
dant pink cytoplasm and well-defined nucleoli. In
between the ganglion cells is mature Schwannian
stroma. These ganglioneuromas are generally
associated with a benign clinical course.
At the opposite end of the histological spec-
trum are tumours comprised of sheets of imma-
ture small round blue cells. These sometimes
show rosette formation—a circular arrangement
of cells with central neurofibrillary material
(Fig.  49.1). Those tumours composed predomi-
nantly of neuroblasts exhibit more aggressive
behaviour.
When there is a substantial infiltration of imma-
ture neuroblasts in an otherwise mature tumour,
the lesion is termed a ganglioneuroblastoma.
In assessing the histological appearance, the
mitosis-karyorrhexis index (MKI) is also calcu-
lated. This is an index of the number of cells in
mitosis and karyorrhexis per 5000 cells.
Although many of these tumours may be diag-
nosed on standard haematoxylin and eosin stain-
ing, a battery of immuno-histochemical stains is
routinely used to confirm the diagnosis. This
allows differentiation from other small blue
round cell tumours.
Various pathological grading systems have
been used to predict tumour behaviour and to
direct therapy, and an International Neuroblastoma
Pathology Classification has been agreed.
636
The INRG Classification System uses histo-
logical category, MKI and grade of differentia-
tion to assist in assigning risk groups.
49.7 Sites of Disease
The adrenal medulla is the commonest site of pri-
mary neuroblastoma—about 50% of the total.
Other abdominal sites—sympathetic chains or
the pre-aortic region—are involved in about 20%.
The posterior mediastinum is the site of primary
tumour in perhaps 10% and the pelvis in about
7% of the total. Less than 5% of tumours occur in
the neck.
In the majority, metastatic spread has already
occurred at the time of diagnosis—up to 70%.
Metastatic spread is predominantly to lymph
nodes, bone and bone marrow. The disease is
localised in perhaps 25%, and about 10% of the
total have the special infant form of the disease—
MS stage disease.
49.8 Markers of Disease
Behaviour
49.8.1 Biochemical Markers
Several serum biochemical markers have been
used to assist in predicting likely outcome. The
three most frequently applied are lactate dehy-
drogenase (LDH, normal <1500 IU/ml), ferritin
(normal <142 ng/ml) and neuron-specific enolase
(NSE, normal <15 ng/ml). Raised levels of these
markers are associated with more aggressive
tumour biology. The INRG Classification System
considers that these markers are no longer clini-
cally useful.
49.9 Molecular Markers
Increasing numbers of genetic markers of disease
activity have been defined. The sheer number of
these markers is sometimes considered a hin-
drance to understanding disease behaviour and in
planning clinical care.
E. Kiely
The INRG Staging System (INRGSS) uses
MYCN status, presence/absence of 11q aberra-
tions and cell ploidy in assigning risk groups
(Monclair et al. 2009).
The MYCN proto-oncogene is located on
chromosome 2, and amplification (>10 copies) is
associated with more aggressive clinical behav-
iour and advanced disease stage. About 20% of
neuroblastoma patients show MYCN amplifica-
tion, and this increases to 50% in those present-
ing with metastatic disease.
11q deletion is found in about one third of all
neuroblastomas and is not associated with MYCN
amplification. Any 11q aberration is indepen-
dently associated with reduced survival.
DNA ploidy also affects tumour behaviour.
Polyploidy—greater than diploid content—is asso-
ciated with improved survival. Tumours with dip-
loid content are associated with a worse outcome.
49.10 Staging
Many staging systems have come and gone since
the Evans system was devised. The latest
(INRGSS) will in turn be replaced as knowledge
and treatment evolve. However, the INRGSS
(Table  49.1) has the advantage of simplicity in
applying risk grouping at the time of diagnosis
and utilises contemporary knowledge.
It replaces the previous International
Neuroblastoma Staging System (INSS) which
has been widely used for almost 30 years. As part
of the INSS is post-surgical, direct comparisons
of outcomes are not always possible between the
different treating groups.
Table 49.1  International neuroblastoma risk group stag-
ing system
Stage
L1 Localised tumour in single body compartment,
no image-defined risk factors
L2 Localised tumour with one or more image-­
defined risk factors
M Distant metastases (excluding MS disease)
MS Younger than 18 months (547 days) with
metastatic disease confined to the liver, skin,
+/− bone marrow (<10% of nucleated marrow
cells)
49 Neuroblastoma
The difference between Stages L1 and L2 is
based on the appearance on initial cross-sectional
imaging. Image-defined risk factors (IDRFs)
include tumour extension into adjacent body
compartments and/or displacement/encasement
of major vessels, airway or major nerves. Under
these circumstances, primary surgery might be
considered unduly hazardous.
The INRG Classification System is based on
tumour stage, patient age, histological features,
MYCN status, 11q aberrations and ploidy (Cohn
et al. 2009). Four distinct risk subsets have been
defined—very-­ low-, low-, intermediate- and
high-risk groups. These groups may be further
subdivided into different treatment groups, with
the emphasis on reducing treatment for those
with more favourable biological profile.
49.11 Presentation
49.11.1 Screening
In an attempt to reduce the incidence of children
presenting with advanced disease, screening pro-
grammes have been applied most notably in
Japan and Canada. Urinary catecholamines were
checked during the first year of life. This resulted
in a substantial increase (more than threefold) in
the number of infants being diagnosed with neu-
roblastoma, but numbers of older children and
the mortality from the disease remained
a b
Fig. 49.2 (a). Skin metastasis in a newborn with a Stage IV S neuroblastoma (courtesy of Michael Höllwarth) (b).
Multiple liver metastases in a newborn with Stage IV S neuroblastoma (courtesy of Michael Höllwarth)
637
unchanged (Brodeur 2018). Most of the tumours
diagnosed on screening were of good biology. By
inference, therefore, it seems likely that a large
number of neuroblastomas in infancy resolve
without any treatment. Because the outcome for
screened children was not improved overall,
screening has largely been abandoned.
49.11.2 Clinical Presentation
A small number—about 3%—of neuroblastomas
are detected antenatally. These infants usually
have low stage disease with good biology.
Another group of young infants present with
skin nodules and/or hepatomegaly (Fig.  49.2a,
b)—Stage MS disease.
The majority of patients, however, present with
symptoms from either the mass or the metastases.
An abdominal mass may be palpated during clini-
cal examination for non-specific symptoms.
Bone pain, arthralgia or peri-orbital bruising is a
consequence of bony metastases. Increased sweat-
ing is occasionally reported, and hypertension from
renal artery compression is not unusual. Generally,
the children with metastases have a period of gen-
eral malaise and ill health preceding diagnosis.
Thoracic tumours may be detected on a chest
X-ray taken for presumed chest infection. Pelvic
tumours interfere with bowel or bladder function
and are found during investigation for these
problems.
638
Uncommon presentations include paraplegia
from spinal cord compression or diarrhoea as a
consequence of vaso-active intestinal polypep-
tide secretion. Finally, a small group presents
with opsomyoclonus (dancing eye syndrome).
Physical examination may reveal a fixed, hard,
irregular abdominal mass. A pelvic mass may be
detected on rectal examination, but otherwise, in
the absence of peri-orbital bruising or skin nodules,
there may be little to find on clinical examination.
49.12 Diagnosis
The diagnosis may be suspected on the basis of the
history and the physical findings. The presence of
a mass together with raised urinary catecholamine
metabolites—vanillylmandelic acid (VMA) and
homovanillic acid (HVA)—initially confirms the
diagnosis. Raised levels of these metabolites are
found in about 85% of neuroblastomas.
Tissue diagnosis is mandatory. As assessment
of tumour biology is essential for staging, and
before commencing treatment, tissue from the
tumour or from the metastases is needed.
Sufficient tissue for all relevant investigations is
usually obtainable utilising multiple needle cores.
Plain X-rays of the chest or abdomen will fre-
quently give useful information. Over half the
tumours showed dystrophic calcification.
Ultrasound is used to assess the nature and
extent of the tumour as well as to define its size.
The echo pattern is heterogeneous and unlike
other childhood cancers. Ganglioneuromas pres-
ent a more uniform appearance than undifferenti-
ated neuroblastomas, and an experienced
radiologist will often suspect the diagnosis from
the ultrasound appearance.
Cross-sectional imaging in the form of MRI or
contrast-enhanced CT is also mandatory. These
images provide precise anatomical information
but may underestimate the full extent of the
tumour (Fig. 49.3).
Isotope scanning is essential for staging pur-
poses. 123I-metaiodobenzylguanidine (MIBG)
scanning is widely available and is considered the
optimal form of isotope imaging for these tumours.
Technetium bone scanning may be used in tumours
which are not MIBG-avid (about 10–15%).
E. Kiely
Fig. 49.3  Significant elongation of the left renal artery
due to a giant abdominal neuroblastoma involving most of
the major intestinal vessels; calcification can be seen
(courtesy of Michael Höllwarth)
Finally, bone marrow aspirates and trephines
are essential to fully define the clinical stage.
49.13 Differential Diagnosis
The presence of a very large abdominal mass in a
young child usually means a neuroblastoma or
Wilms’ tumour. Bone pain and arthralgia raise
the possibility of some form of inflammatory
arthritis. In most children, diagnosis presents lit-
tle difficulty although peri-orbital bruising may
give rise to suspicion of child abuse. In general,
the diagnosis is made quite rapidly after limited
investigation.
49.14 Management
Available treatment modalities include chemo-
therapy, surgery, radiotherapy and immunother-
apy. In most centres, care is coordinated by
paediatric oncologists, and the various steps in
treatment are in accordance with agreed national
and international protocols.
A large team is needed to look after these chil-
dren. The surgeon’s role includes biopsy of tissue
for diagnosis, provision of vascular access and
tumour resection.
The timing of surgical intervention is also pro-
tocol driven and should take place after d­ iscussion
49 Neuroblastoma 639

with oncologists and radiologists. The final deci- tine) and anthracyclines (doxorubicin). Dose
sion about operation can only be made by a sur- intensity and length of treatment are increased
geon who alone bears responsibility for the result for those with advanced or biologically unfavour-
of the procedure. able tumours.
At the present time, after full investigation and In addition, myeloablative therapy with stem
staging, neuroblastoma management occurs in cell rescue is part of many treatment protocols for
three phases. The Induction phase includes che- those with high-risk disease.
motherapy where indicated and surgery. The
Consolidation phase may include myeloablative
treatment with autologous stem cell rescue +/_ 49.16 Biological Therapy
radiotherapy. The Maintenance or Post-­
Consolidation phase, mainly in the high-risk Isotretinoin promotes maturation of immature
group, usually entails some form of biological neuroblasts. Maintenance treatment of neuro-
treatment. blastoma with isotretinoin is included in many
In general, localised tumours with no IDRFs protocols in an attempt to alter the behaviour of
(Stage L1) are amenable to primary operation, residual chemotherapy-resistant neuroblastoma
and this may be the only treatment needed. cells.
For Stage L2 tumours, primary surgery may More recently, this has been combined with
be the preferred option, depending on IDRFs, site the use of an anti-GD2 antibody. GD2 is a disi-
of tumour and the available expertise. For many, aloganglioside expressed on the surface of neuro-
pre-operative chemotherapy may be considered a blastoma cells, and a number of anti-GD2
better choice in order to shrink the mass and monoclonal antibodies have been evaluated with
allow for a safer and more complete excision. significant improvement in survival. This is usu-
Most neuroblastoma treatment protocols ally combined with administration of granulocyte-­
advise upfront chemotherapy for Stage M—dis- macrophage colony-stimulating factor (GMCSF)
tant metastases. As a result of such treatment, the to optimise the white cell response.
tumours are usually much smaller but, more
importantly, are firmer and less vascular, allow-
ing for the possibility of complete macroscopic 49.17 Surgery
resection.
Many infants with Stage MS will need no Neuroblastomas frequently encase or are in close
treatment of any kind. In the subgroup with proximity to major blood vessels and nerves. The
relentless hepatomegaly, chemotherapy and aim of surgery is to remove the tumour whilst
radiotherapy may be too slow to interfere with preserving vital structures. These would include
liver growth, and use of a silo is not associated aorta and inferior vena cava, together with their
with very high survival rates. In this particular major branches in the abdomen and the brachial
and uncommon scenario, hepatic artery emboli- plexus in the neck. Non-vital structures would
sation or ligation may be life-saving. include the spleen, some intestine and one kid-
ney. As many as possible of non-vital structures
should be preserved.
49.15 Chemotherapy En bloc tumour resection is frequently not an
option because of vascular encasement. The sur-
Combination chemotherapy has been responsible geon initially dissects and displays the involved
for the improved survival over the past 50 years. vessels as they traverse the tumour. The vessels
Oncology groups worldwide use a combination are then fully mobilised from tumour, and finally
of alkylating agents (cyclophosphamide, the tumour is removed in segments. For paediat-
platinum-­based agents), topoisomerase inhibitors ric surgeons, the surgery is unusual. The opera-
(topotecan, etoposide), vinca alkaloids (vincris- tion is a vascular procedure and is unlike other
640 E. Kiely

paediatric surgical operations. The usual dissec- present time. The use of anti-GD2 antibody has
tion techniques may be inadequate or unduly been a major advance and is likely the forerunner
hazardous. of future biological therapies.
However, these tumours rarely invade the For infants with Stage MS disease, reported
tunica media of major blood vessels, and there- survival varies from 50 to 80%. Progressive hep-
fore dissection may take place in the subadventi- atomegaly producing respiratory compromise
tial plane. This is most easily accomplished using has been associated with multi-organ failure and
knife dissection (Kiely 2005). death. This subgroup needs more active manage-
Complete resection is not always possible. In ment, but the remainder do well with little or no
general, complete or near-complete tumour treatment.
resection is achieved in >90% of children. The Similarly, for those with antenatal detection of
incidence of nephrectomy to achieve more com- small and/or cystic tumours, expectant manage-
plete excision is about 10%, but as some chemo- ment is usually all that is required.
therapy is nephrotoxic, every effort is made to An age of 547 days (18 months) or less at pre-
preserve renal tissue. sentation is associated with markedly improved
survival, independent of disease stage. The older
the age of presentation, the worse the outcome.
49.18 Radiotherapy Tumour histological category, MYCN status,
DNA ploidy and presence/absence of 11q aberra-
External beam radiotherapy is often used after tion are used to define risk groups in the
surgery to improve local control. Total body irra- INRGSS. Treatment protocols differ between the
diation is utilised as part of conditioning for bone different oncology groups but can, in the future,
marrow transplantation. 131I- MIBG therapy is be directly compared.
used in MIBG-avid tumours, both for treatment Surgical resection alone may be curative in
of minimal residual disease or increasingly as those with Stage L1 and L2 disease without the
part of induction treatment. use of chemotherapy. However, in Stage L2, chil-
dren chemotherapy may be advised because of
tumour biology.
49.19 Outcome The literature has been contradictory in regard
to the benefit of complete resection in those with
In the past 50 years, the overall outlook for chil- Stage M or high-risk disease (Ryan et al. 2020;
dren with neuroblastoma has been transformed— Newman et al. 2019). A recent SIOPEN study of
from a survival rate of less than 15% to over 75%. >1300 patients with high-risk disease showed
This has been the result of multimodal treatment that survival was improved after complete surgi-
using protocols designed by collaborative groups. cal resection. This would appear to finally settle
The outlook for those with very-low- and low-­ the question of whether or not to attempt a diffi-
risk disease is now >90% survival. For those with cult operation in these children.
intermediate-risk disease, over 80% will survive.
For these three patient groups, the recent empha-
sis has been on reducing treatment in an attempt 49.20 Conclusion
to have fewer long-term complications of treat-
ment whilst maintaining excellent survival rates. The outlook for children with this tumour has
There has not been a similar improvement in improved. The use of next-generation sequencing
survival for those with high-risk disease where has allowed genome-wide searches for genetic
<50% survive long term. Intensification of treat- mutations, but these are relatively infrequent in
ment has been the thrust of management up to the neuroblastoma. Consequently, there are fewer
49 Neuroblastoma
druggable targets in distinction to many adult
cancers. The aim is to find targets whereby small
molecules might exert a profound effect on
tumour biology.
It is unclear at present if the targets will be in
the genome or the epigenome, but both areas are
the focus of intense research activity. In the
meantime, it does not seem likely that significant
improvements in outcome will occur by refine-
ments in the use of current chemotherapeutic
agents. Neither is it likely that major changes in
treatment are likely in the short term.
References
Beckwith JB, Perrin EV (1963) In situ neuroblastomas:
a contribution to the natural history of neural crest
tumours. Am J Pathol 43(6):1089–1104
Blacklock JWS (1934) Neurogenic tumours of the sympa-
thetic system in children. J Pathol Bacteriol 39:27–48
Brodeur GM (2018) Spontaneous regression of neuro-
blastoma. Cell Tissue Res 372(2):277–286
Cheung N-KV, Dyer MA (2018) Neuroblastoma: devel-
opmental biology, cancer genomics, and immunother-
apy. Nat Rev Cancer 13(6):397–411
Cohn SL, Pearson ADJ, London WB et  al (2009) The
international neuroblastoma risk group (INRG) clas-
sification system: an INRG task force report. J Clin
Oncol 27(2):289–297
Evans AE, D’Angio GJ, Randolph J (1971) A proposed
staging system for children with neuroblastoma.
Children’s cancer study Group A. Cancer 27:374–378
Farber S (1940) Neuroblastoma. Am J Dis Child
60:749–750
641
Fortner J, Nicastri A, Murphy ML (1968) Neuroblastoma:
natural history and results of treating 133 cases. Ann
Surg 167:132–142
James DH, Hustu O, Wrenn EL et al (1965) Combination
chemotherapy in childhood neuroblastoma. JAMA
194:123–126
Johnsen JI, Dyberg C, Wickström M (2019)
Neuroblastoma-a neural crest derived embryonal
malignancy. Front Mol Neurosci 29(12):9
Kastriti ME, Kameneva P, Kamenev D et  al (2019)
Schwann cell precursors generate the majority of
chromaffin cells in Zuckerkandl organ and some sym-
pathetic neurons in paraganglia. Front Mol Neurosci
12:6–34
Kiely E (2005) A technique for excision of abdominal
and pelvic neuroblastomas. Ann Roy Coll Surg Engl
89:342–348
Lehman EP (1932) Adrenal neuroblastoma in infancy  –
15 year survival. Ann Surg 95:473
Monclair T, Brodeur GM, Ambros PF et  al (2009) The
international neuroblastoma risk group (INRG) stag-
ing system: an INRG task force report. J Clin Oncol
27(2):298–303
Newman EA, Abdessalam S, Aldrink JH et  al
(2019) Update on neuroblastoma. J Pediatr Surg
54(3):383–389
Phillips R (1953) Neuroblastoma. Ann R Coll Surg Engl
12:29–47
Ryan AL, Akinkuotu A, Pierro A et al (2020) The role of
surgery in high-risk neuroblastoma. J Pediatr Hematol
Oncol 42(1):1–7
Tsubota S, Kadomatsu K (2018). Origin and initia-
tion Mechanisms of neuroblastoma. Cell Tissue Res
372:211–221
Virchow R (1864) Die Krankenhaften Geschwulste, vol
11. Hirschwald, Berlin, p 149
Wright JH (1910) Neurocytoma or neuroblastoma, a
kind of tumour not generally recognised. J Exp Med
12:556–561
 Soft Tissue Sarcomas
Sandeep Agarwala and Robert Carachi
50.1 Introduction
The soft tissues include connective tissues, lymphat-
ics, vessels, smooth and striated muscles, fat, fascia,
synovium, endothelium, and reticuloendothelium.
Tumors arising from any of these are soft tissue sar-
comas (STS), and these tumors behave in a very dif-
ferent biological manner from those tumors arising
from blastemal elements. STS are uncommon in
children, accounting for about 6% of all childhood
malignancies. The most common of these are those
arising from the immature mesenchymal cells that
are committed to skeletal muscle lineage and are
called rhabdomyosarcomas (RMS). The remaining
group consists of a heterogeneous collection of sub-
types, referred to as non-rhabdomyosarcoma soft
tissue sarcomas (NRSTS). Currently, local control is
achieved in 80–90% of RMS patients, and approxi-
mately 70% of the patients survive for 5  years or
more and are probably cured.
S. Agarwala (*)
Department of Pediatric Surgery, All India Institute of
Medical Sciences, New Delhi, India
R. Carachi
Department of Surgical Pediatrics, University of
Glasgow, Queen Elizabeth University Hospital,
Glasgow, UK
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_50
50
50.2 Incidence
Soft tissue sarcomas are commoner in males
compared to females. Among the extracranial
solid tumors of childhood, RMS is the third most
common neoplasm after neuroblastoma and
Wilms’ tumor, comprising 15% of all solid
tumors. There is a bimodal incidence, with almost
two-thirds of cases of RMS being diagnosed in
children <6  years of age, with another mid-­
adolescence peak. However, all ages are suscep-
tible to develop these tumors, and some may
present in the neonatal period (Fig. 50.1), which
makes them very difficult to treat. RMS is ubiqui-
tous, occurring almost everywhere, but most
commonly in the head and neck and the genito-
urinary (GU) areas and then the extremities and
trunk (Table 50.1).
643
644 S. Agarwala and R. Carachi

50.3 Pathology

50.3.1 Histology
and Immunohistology

RMS are grossly firm, nodular, and of variable

size and consistency. They are well circum-
scribed, but not encapsulated, and often tend to
infiltrate extensively into adjacent tissues. The
sarcoma botryoides subtype has a characteristic
grape-like appearance, with its grape-like clus-
ters of tumors arising from a mucosa-lined area.
Histologically, RMS falls into the broad category
of small blue round cell tumors, which also
include neuroblastoma, Ewing’s sarcoma, small
cell osteogenic sarcoma, and lymphomas. It is
possible for the histopathologist to differentiate
between the different tumor types by a combina-
tion of microscopic appearance and immunohis-
tochemical staining and the molecular genetic
characteristics of these different types of tumors.
The standard classification of RMS is an interna-
tional classification system for childhood RMS
(Table  50.2). Light microscopy diagnosis of
Fig. 50.1  RMS of the left leg in a neonate with enlarged RMS is based on the identification of cross stria-
inguinal lymph nodes tions, characteristic of skeletal muscle, or
­characteristic rhabdomyoblasts. Cross striations
Table 50.1  The sites of RMS and their incidences and
are seen in 50–60% of cases. Histologically,
distribution to favorable and unfavorable sites embryonal rhabdomyosarcoma (ERMS) is com-
Region posed of rhabdomyoblasts and small round cells.
(incidence) Favorable sites Unfavorable sites Rhabdomyoblast, the more mature of the embry-
Head and Non-orbital Parameningeal onal components, is characterized by bright
neck (35%) non-­ (44%—Ear, eosinophilic cytoplasm (Fig. 50.2). ERMS is the
parameningeal mastoid, nasal
cavity, paranasal
commonest type of tumor (70% of all RMS).
(30%—Oral
cavity, larynx, sinuses, Eighty percent of urinary tract tumors are made
parotid region, infratemporal fossa, up of the embryonal type, whereas 60% of the
cheek, scalp, and and pterygopalatine head and neck tumors are embryonal, and only
soft tissues of the fossa)
neck)
half of the extremity ones fall into this category.
Orbital (25%) Sarcoma botryoides and spindle cell variant are
Genitourinary Vulva, vagina, Bladder, prostate, two subtypes of embryonal RMS, and the out-
(22%) uterus, and kidney, and ureters come in both these subtypes is excellent (5-year
paratesticular survival of 95%). The sarcoma botryoid can
Extremities All
occur in the vagina, uterus, bladder, and biliary
(14%)
Trunk (27%) All tract, while the spindle cell variant is most often
Others Biliary Retroperitoneum, seen in the paratesticular region and rarely in the
perineum, and head and neck. Alveolar RMS (ARMS) consti-
perianal tutes 30% of RMS and histologically consists of
50  Soft Tissue Sarcomas 645

rhabdomyoblasts mixed with larger round cells (same as pleomorphic in adults), which is
with prominent eosinophilic cytoplasm. The extremely rare in children (<1%), shows ana-
tumor grows in cords and produces cleft-like plastic cells present in large aggregates or as dif-
spaces, namely, alveoli. The anaplastic RMS fuse sheets. It occurs in the extremities and the
trunk. Skeletal muscle or muscle-specific pro-
Table 50.2  International classification system for child- teins, like antidesmin, muscle-specific actin, and
hood rhabdomyosarcomas MyoD, can be identified by immunohistochemi-
I Superior prognosis cal staining. Monoclonal antibodies, like those
 (a) Botryoid against desmin, muscle-specific actin, sarco-
 (b) Spindle cell meric actin, and myoglobin, have also been used
II Intermediate prognosis
to confirm the myogenic lineage with very good
 (a) Embryonal
III Poor prognosis specificity and sensitivity. Monoclonal antibod-
 (a) Alveolar ies against MyoD can be used in frozen section
 (b) Undifferentiated sarcomas analysis.

Fig. 50.2  Core needle biopsy specimen showing the histology and immunohistochemistry of ERMS
646
50.3.2 Molecular Cytogenetics
The ERMS and ARMS have been associated with
distinct clinical characteristics and genetic altera-
tions. Eighty percent of ERMS have an allelic loss
of gene at 11p15.5 (loss of heterozygosity). This
is the locus for insulin-like growth factor-II
(IGF-II) gene, and LOH results in the
­overexpression of the gene, and this stimulates
growth of rhabdomyoblasts. Eighty percent of
ARMS have translocation between FKHR gene
on chromosome 13 and either PAX3 gene on
chromosome 2, t(2:3)(q35;q14), or PAX7 gene on
chromosome 1, (t1:13)(p36;q14), resulting in
PAX3/FOXO1 or PAX7/FOXO1 fusion gene,
respectively. This generates PAX3-FKHR and
PAX7-FKHR fusion proteins, respectively. The
diagnosis of ARMS is now mostly dependent on
the identification of these fusion genes. Twenty
percent of ARMS are fusion negative, and these
show clinical features and outcomes closely
resembling ERMS.  Future COG studies have
risk-stratified ARMS based on fusion status
instead of histology. Therefore, despite similari-
ties, the striking cytogenetic and molecular differ-
ences between ARMS and ERMS indicate distinct
molecular etiologies in these tumor subtypes.
Prognostic and therapeutic implications of
biologic, immunologic, and cytologic techniques
make it essential for the surgeon to obtain ade-
quate quality and quantity of tissue and to ensure
proper transfer of the specimens to the laboratory
while they are fresh, so as to enable the patholo-
gist to do all the requisite work-up.
50.4 Diagnosis
50.4.1 Imaging
The commonest imaging used is a plain X-ray,
which may demonstrate a soft tissue mass with
calcification if a hemorrhage has occurred into
the tumor. In a tumor of the head and neck, the
bony structures may be eroded, especially in
orbital meningeal/parameningeal tumors. In a
tumor arising from the neck, the mediastinum
S. Agarwala and R. Carachi
Fig. 50.3  Ultrasound of the bladder with RMS of the
base of the bladder
may be shifted, and the trachea may be pushed
away from its central position. In truncal lesions,
there may be bony erosions, metastases in the
lung, or pleural effusions.
Ultrasound is another modality that can be very
helpful initially to define a solid from a cystic
lesion. It is most helpful in genitourinary RMS to
identify the origin of the tumor and detect any
obstructive lesion causing an obstructive uropathy,
i.e., a large bladder, and hydronephrosis. In RMS
of the urinary bladder and prostate, ultrasound can
form the baseline (Fig.  50.3) for subsequent
repeated tumor response assessments and for eval-
uation of the hydronephrosis. This avoids the
repeated use of ionizing radiation. Interventional
radiology is used for image-guided (US or CT)
fine needle aspiration or core needle biopsies in
sarcomas.
Cross-sectional imaging, like contrast-­
enhanced CT scan and/or magnetic resonance
imaging (MRI) scans, of the affected region is
essential in the proper evaluation of the primary
site and to detect extensions of the tumor and its
relationship with the surrounding structures
(Fig. 50.4). While a CT scan is the standard imag-
ing used, a MRI is the modality of choice for the
head and neck, the extremities, and the pelvic
regions, where they can clearly define the differ-
ent tissue planes and also provide multi-planar
images.
50  Soft Tissue Sarcomas
Fig. 50.4  CT scan of an extensive bladder neck RMS
with outlet obstruction
Even when MRI is used as the primary imag-
ing modality, a CT scan of the chest is always
required to assess the lung metastases. CT can
detect pulmonary nodules that are occult in con-
ventional chest radiographs. CT is the most sen-
sitive modality for the purpose of detecting
pulmonary metastasis and is more sensitive than
FDG-PET for this purpose. CT can also identify
small metastatic lesions at the lung periphery and
the pleura. A CT, when viewed in cine mode, can
differentiate small nodules from vessels sec-
tioned transversely.
Pre-treatment scans help the radiation oncolo-
gist to properly define the volumes at risk for
tumor invasion and plan treatment fields. These
imaging studies are also done at regular intervals
during the follow-up to assess for tumor
recurrence.
647
Nuclear scans, using technetium 99 m methy-
lene diphosphonate, are useful for detecting osse-
ous metastases. Except for the size criteria that is
used for RPLN metastasis in a paratesticular RMS,
CT and MRI are unable to distinguish whether
enlarged LN are because of neoplasm or reactive
hyperplasia. For this, MRI lymphangiography,
using ultrasmall superparamagnetic iron oxide
nanoparticles (USPIO), can differentiate nodes
involved with the neoplasm from reactive normal
nodes. Sentinel node mapping for extremity RMS,
using Tc-99 labelled radiocolloids, is injected in
and around the tumor and gets transported via
draining lymphatic channels to nodes representing
the first site of lymphatic metastasis. These nodes
can be marked, either pre-­ operatively or intra-
operatively, by hand held gamma probes and assist
in LN sampling for histological confirmation. Fast
whole body MRI is superior to conventional radi-
ography, CT, and Tc-99 bone scintigraphy and
FDG-PET in defining the presence and extent of
neoplastic infiltration of the bone marrow.
The role of positron emission tomography
(PET) scans in sarcomas is still being defined.
FDG-PET/CT or FDG-PET/MRI combines met-
abolic and morphologic imaging for localizing
viable tumor metastases. PET-CT are used to dif-
ferentiate whether a residual mass after chemo-
therapy or surgery is just fibrosis or a tumor
containing viable malignant elements (Fig. 50.5).
FDG avidity of malignant nodes in sarcomas of
childhood is only mild to moderate, and FDG
activity of lymphoid tissues can also be seen in
non-malignant conditions, like granulomatous
disease and reactive hyperplasia, making FDG-­
PET less sensitive. The role of PET-CT in detect-
ing regional LN metastases in cases of RMS of
the extremities and paratesticular regions is still
being evaluated, and if found sensitive, this may
reduce the need for LN sampling for histological
confirmation. Whole body PET scans are also
sensitive for the detection of bone marrow
involvement. Therapy with granulocyte-­
stimulating factors (G-CSF) may cause false
positives in both FDG-PET and MRI.
648
Fig. 50.5  FDG PET-CT uptake in residual RPLN demonstrating residual viable tumor
50.4.2 Sampling for Biopsy
The biopsy for STS should preferably be obtained
as an open incisional biopsy, but today, with excel-
lent interventional imaging techniques, these can
be done using co-axial core biopsy needles that
can obtain multiple samples through one puncture
that is guided by either ultrasound or CT scan.
During an open biopsy, the role of the surgeon is to
ensure that under a general anesthetic, an adequate
open incision is made at a carefully chosen site,
which will be included in the eventual formal
resection performed later. A careful incision is
made through the capsule of the tumor if there is
one. This is in order to obtain a proper representa-
tion of the tumor and an adequate sample. The
tumor must be sent fresh to the pathology depart-
ment due to the variety of biological and histo-
chemical tests that need to be carried out on this
type of tissue. It is important that the pathologist is
aware that this type of tumor surgery is being per-
formed and can provide advice as to whether ade-
quate amount of tissue has been obtained to be
able to provide a diagnosis. All attempts at removal
of the tumor en bloc at initial presentation should
be resisted. The tumors are often gross at presenta-
tion, and there is a real risk of compromising adja-
cent tissues, which may have been invaded by this
tumor. Trunk and extremity biopsies should be
performed along the long axis of the tumor so that
subsequent excisions are not compromised.
S. Agarwala and R. Carachi
50.5 Differential Diagnosis
50.5.1 Clinical Evaluation,
Symptoms, and Signs
The clinical evaluation of patients includes a care-
ful history and a physical examination because, by
their very nature, RMS may occur anywhere in the
body. A family history may be significant since
soft tissue sarcomas may occur in cancer families
as a component of the Li-Fraumeni syndrome, as
well as the Beckwith-­ Wiedemann syndrome
(Carachi R et al. 2008). The genetics of these two
syndromes are associated with soft tissue sarco-
mas. The Beckwith-Wiedemann syndrome has a
complex genetic makeup. It is often associated
with several 11p15 chromosomal changes. It is an
overgrowth syndrome often associated with hypo-
glycemia. Other associated tumors are Wilms’
tumor and liver tumors. The Li-Fraumeni syn-
drome is an autosomal dominant disorder, where
soft tissue sarcoma occurs at an early age in a can-
cer family. Other associated tumors are brain
lesions, breast cancer, and leukemia. RMS has also
been associated with neurofibrosarcomas.
Over 80% of patients have localized disease at
the time of presentation. The symptoms at pre-
sentation often depend on the primary site. In
general, there are three main areas that can be
affected. These are the head and neck, the trunk
and limbs, and the genitourinary system.
50  Soft Tissue Sarcomas
Fig. 50.6  RMS of the neck presenting as a lump
50.5.2 RMS of Head and Neck Region
These tumors often present early, with a swelling
or bleeding from an orifice, which might be the
ear or the nose or even proptosis of the eye. They
may present as a lump in the head and neck
region and mimic lymph nodal enlargement due
to commoner infective pathology (Fig.  50.6).
Such tumors may mimic parotitis or even cause
neurological dysfunction secondary to extension
of parameningeal lesions affecting the central
nervous system. These tumors may be confused
with commoner lesions initially, but gradually
the progression of tumor growth is relentless, and
it soon becomes obvious that this is not an infec-
tive condition, when they do not respond to stan-
dard regimes of antibiotics (Fig.  50.7). These
tumors are aggressive and gradually erode bony
margins and can infiltrate from one space in the
head and neck to another. Regional lymph nodes
are mostly not involved.
649
Fig. 50.7  RMS parotid region that was initially confused
as parotitis
Orbital RMS may present with proptosis, che-
mosis, or a conjunctival mass. This may lead to
blindness and ophthalmoplegia later on. Tumors
that arise from parameningeal sites are associated
with erosion of the cranial bones, which can pres-
ent with meningeal symptoms and cranial nerve
manifestations. A tumor arising in the nasopharynx
may cause airway obstruction and bleeding. Those
arising in the paranasal sinus present with pain and
nasal discharge. A tumor in the middle ear or mas-
toid region may present as a polypoidal tumor with
obvious facial nerve palsy. The commonest histol-
ogy in this group of tumors is embryonal. Thus, a
clinical presentation may at first present to the gen-
eral practitioner; to the ear, nose, and throat sur-
geon; or even to the ophthalmologist.
50.5.3 RMS of the Trunk
and Extremities
RMS of the extremities are more common in the
lower limbs and occur in the older age group; how-
ever, they can be seen in neonates and infants too
(Fig. 50.8a). Regional draining nodal involvement
is seen in 30%–50% cases, but all may not be pal-
pable. There is often a delay in presentation, and
occasionally it may present as a groin or axillary
node. Truncal masses of the chest wall may occa-
sionally have a large intra-thoracic component
causing respiratory symptoms or neurological
symptoms if cord compression results from a para-
650
a b
Fig. 50.8 (a) RMS involving the calf mimicking an abscess. (b) RMS involving the calf with extensive inguinal lymph
nodal metastases
spinal tumor with intra-spinal extension causing
cord compression. RMS of the trunk and the limbs
usually present with a localized swelling, which
can be confused with a hematoma following a
minor injury and at times as an abscess. The lesion
may be associated with bruising and may have an
inflammatory component to it, which can be very
confusing for the diagnostician. The tumor grows
relentlessly and does not behave like a simple
hematoma and spreads early to the regional glands
(Fig. 50.8b). Another potential confusion is a sud-
den change in a congenital hemangioma/lymphan-
gioma. If it suddenly changes in characteristics
and becomes hard, it can be confused with the
development of a RMS. Occasionally, clinical pre-
sentation may be with metastatic disease to other
parts of the body and especially to the lungs from
a small primary in the periphery.
50.5.4 RMS of the Genitourinary
System
The sites that are generally affected include the
bladder, prostate, and paratesticular region in the
male and the bladder, vagina, uterus, and vulva in
the female. The histopathology of this tumor is
very often of embryonal variety and carries a
good prognosis. The poor sites include the RMS
that arise in the base of the bladder and the pros-
tate. RMS of the bladder/prostate can manifest
with urinary frequency, urgency, retention, and
gross hematuria. Paratesticular tumors are u­ sually
S. Agarwala and R. Carachi
hard, painless scrotal swellings that can present
with signs and symptoms secondary to a larger
intraperitoneal mass and have a characteristic
spindle cell on histology. Paratesticular masses
can be misdiagnosed as a hydrocele if proper
physical examination is not done. RMS of the
vagina can present as bloody vaginal discharge or
the presence of sarcoma botryoides, which are
fleshy polyps that protrude from the vaginal ori-
fice and may be found in the nappy of a child
(Fig.  50.9a, b). RMS of the vulva usually pres-
ents as a small vulval nodule, and that of the
uterus presents as hypogastric mass. RMS of the
genitourinary system usually present earlier
because of the confined space within the pelvis
causing an obstructive uropathy. During the
examination of the child, a full examination of
the pelvic region and a digital rectal examination
are mandatory in order to pick this up and avoid
confusion with what maybe mistakenly diag-
nosed as a constipation. On rectal examination,
these masses are felt anterior to the rectum unlike
the sacrococcygeal tumors that are retro-rectal. In
the genitourinary tract, some of these masses
may be very large and present as a lower abdomi-
nal mass. This can be a true tumor or can be due
to an obstructed enlarged bladder, which may
enhance the size of the mass. Paratesticular
tumors are usually hard, painless scrotal swell-
ings that can present with signs and symptoms
secondary to a larger intraperitoneal mass and
have a characteristic spindle cell on histology.
50  Soft Tissue Sarcomas
a b
Fig. 50.9 (a) Sarcoma botryoides protruding from the vagina. (b) Botryoid fleshy masses in nappy of the same patient
50.6 Management
50.6.1 Staging
It is important to properly stage every RMS
patient as the treatment and prognosis depend on
the stage. Staging prior to treatment requires a
thorough clinical examination and laboratory and
imaging examinations. A biopsy is required to
establish the histologic diagnosis. Pre-treatment
size is determined by external measurement or
MRI or CT, depending upon the anatomic loca-
tion. For less accessible primary sites, CT is
employed as a means of lymph node assessment
as well. Metastatic sites require some form of
imaging (but not histologic confirmation, except
for bone marrow examination) confirmation.
Various staging systems have been described,
but the most widely used is the clinical grouping
(CG) used in the IRS studies. This is a surgico-­
pathological system and is dependent on the sur-
gical resection (Table 50.3). The IRS study group
has also adopted a modification of the TNM stag-
ing system (Table 50.4). The TNM system uses
651
Table 50.3  Clinical grouping system used by the IRSG
Group I: Localized disease, completely resected
 A  Confined to the organ or muscle of origin
 B Infiltration outside organ or muscle of origin;
regional nodes not involved
Group II: Total gross resection with evidence of
regional spread
 A Grossly resected tumors with “microscopic”
residual tumor
 B Regional disease completely resected with
regional nodes involved, tumor extension into
adjacent organs, or both
Group III: Incomplete resection or biopsy with gross
residual disease remaining
 A Localized or locally extensive tumor, gross
residual disease after biopsy only
 B Localized or locally extensive tumor, gross
residual disease after “major” resection (>50%
debulking)
Group IV: Any size primary tumor, with or without
regional lymph node involvement, with distant
metastases, irrespective of surgical approach to the
primary tumor
the site, invasiveness, size, nodal status, and dis-
tant metastasis to stage the tumors into stages 1
through 4. Both CG and TNM do not use the his-
tology in their systems.
652 S. Agarwala and R. Carachi

Table 50.4  TNM pre-treatment staging classification for RMS

Stage Sites T Size N M
1 Favorable T1 or T2 a or b N0 or N1 or Nx M0
2 Unfavorable T1 or T2 a N0 or Nx M0
3 Unfavorable T1 or T2 a N1 M0
b N0 or N1 or Nx M0
4 Either T1 or T2 a or b N0 or N1 M1
Definitions:
Site:
  Favorable sites: Orbit, head and neck (excluding parameningeal), or genitourinary (excluding bladder/prostate)
  Unfavorable sites: Bladder/prostate, parameningeal, extremities, trunk, and all others
Tumor: T1 = Tumor confined to anatomic site of origin
  (a) <5 cm in diameter
  (b) >5 cm in diameter
T2 = Extension and/or fixation to surrounding tissues
  (a) <5 cm in diameter
  (b) >5 cm in diameter
Regional nodes:
  N0: Regional nodes not clinically involved
  N1: Regional nodes clinically involved by tumor
  Nx: Clinical status of regional nodes unknown (specially sites which preclude lymph node evaluation)
Metastasis:
  M0: No distant metastases
  MI: Metastases present

50.6.2 Prognostic Factors, Risk (Table 50.5). The intermediate- and high-risk cat-
Categorization, and Outcome egories have good EFS (50%–70%) and poor
of Management of RMS EFS (35%), respectively (Lautz TB et al. 2019).
The multimodal treatment of RMS is now done
The prognosis of patients with RMS is multifac- according to these risk categories so as to opti-
torial. The most important predictor of outcome mize the outcome and at the same time limit the
is the presence of metastatic disease. Other fac- complications.
tors that have a bearing on outcome are age (age The overall 3-year FFS in IRSG IV was 77%.
1–9  years better than infants and adolescents), In IRSG IV, the common sites of treatment failure
size (≤5  cm better than >5  cm), site (the orbit, were local (51%), regional (17%), and distant
vagina, vulva, uterus, and paratesticular are (32%). Approximately 20% patients with RMS
favorable sites), resectability (CG I/II better than have metastatic disease at presentation, and the
CG III), histology (ERMS better than ARMS), 3-year failure-free survival (FFS) for these is only
subtype (botryoid and spindle cell better), and the 25%. The prognosis is better for ERMS with two
development of relapse. Both CG and TNM sys- or fewer sites of metastases. There has been little
tems are used concurrently to arrive at a risk cat- improvement in survival among metastatic patients
egorization of the tumors (Table  50.5), which over the years. Approximately 30% of RMS
uses both the CG and the TNM staging, in addi- patients relapse, and among these, 50–90% will
tion to histology, and categorizes them into low, die of progressive disease. The survival is better
intermediate, and high risk depending on the after relapse among patients with CG I/II disease
prognosis (EFS). The low-risk tumors are further than CR III/IV. The 5-year survival after relapse is
subdivided into subset A and subset B. The low-­ <20%, and aggressive surgical resection of
risk subset A has excellent EFS of >85%, while relapsed disease has been shown not to make an
subset B has very good EFS of 70%–85%
50  Soft Tissue Sarcomas
Table 50.5  Risk categorization for RMS
Risk group TNM CG Site
Low subset A 1 or 2 I or II FH or UFH
1 III Orbit only
Low subset B 1 III FH
3 I or II UFH
Intermediate 2 or 3 III UFH
1–3 I–III FH or UFH
High 4 IV FH or UFH
FH favorable site; UFH unfavorable site
(See table on TNM for list of FH and UFH sites)
ERMS embryonal RMS; ARMS alveolar RMS
impact on survival; therefore, alternate chemother-
apy regimens are recommended for such patients.
50.6.3 Treatment of RMS
Treatment of children with RMS is multimodal
including surgery, radiation therapy (RT), and
systemic chemotherapy. While multidrug combi-
nation chemotherapy is used for primary cytore-
duction and eradication of gross and
micro-metastases, RT and surgery are used for
local control of the disease either alone or in com-
bination (Malempati S et al. 2012). Sometimes,
RT and surgery may also be used for the eradica-
tion of pulmonary metastases. As the chemother-
apy goes on for many weeks, the placement of a
Hickman catheter or a portacath is useful.
50.6.3.1 Principles of Chemotherapy
All patients with RMS receive chemotherapy.
Most active agents are actinomycin D (A), vin-
cristine (V), cyclophosphamide (C), and doxoru-
bicin (D). Other agents with moderate to high
activity include melphalan, methotrexate, ifos-
famide (I), cisplatin, carboplatin, etoposide (E),
topotecan (T), and irinotecan (I). The gold stan-
dard multiagent combination has been VAC pro-
tocol that has been used by IRSG for nearly three
decades and is still the choice of treatment in the
current COG protocols. Other combinations that
have been tried and compared with VAC are VAI
and VIE.  According to the IRS-V protocol
653
Histology Outcome (EFS)
ERMS Excellent (<85%)
ERMS
ERMS Very good (70–85%)
ERMS
ERMS Good (50–70%)
ARMS
ERMS or ARMS Poor (<35%)
(D9602) (Arndt CS et al.l 2009), the cases with
low-risk subset A receive VA for 45 weeks, with
additional RT for CG II/III (Raney R et al.
2011).  The 5-year FFS and OS for this subset
were 89% and 97%, respectively. The patients
with low-risk subset B receive VAC for 45 weeks
along with RT. The 5-year FFS and OS for this
subset were 85% and 93%, respectively. The
IRS-V protocol (D9803) (Arndt CS et al. 2009;
Minn Y et al. 2010) was for intermediate-­risk
RMS. The patients were randomized to receive
42  weeks of VAC or VAC alternating with
VTC. All patients received 36–50.4 Gy RT. There
was no difference between the two groups as the
4-year FFS with VAC and VTC was 73% and
68%, respectively. In the IRS-V protocol
(D9802) for high-risk RMS, upfront window
treatment with irinotecan or combination vin-
cristine + irinotecan was evaluated with on a
backbone of VAC chemotherapy with interval
compressed VDC  +  IE (Aydin B et al. 2018;
Pappo As et al. 2007).  The 18-month FFS and
OS for this subset were 66% and 80%, respec-
tively. Similar multiagent protocols have been
described by the SIOP MMT, the German CWS,
and, more recently, the European Paediatric Soft
Tissue Sarcoma Group (EpSSG) and the COG-
STS protocols. Patients with progressive or
recurrent RMS generally have poor outcome
with 5-year OS of 17%. Various combinations
using drugs like bevacizumab (monoclonal anti-
body against VEGF), temsirolimus (mTOR
inhibitor), and vinorelbine are being tried with
654 S. Agarwala and R. Carachi

VAC chemotherapy (Dyson KA et al. 2019). For ent. Interstitial or mold brachytherapy may be
such patients with progressive or recurrent RMS, considered in the radiotherapeutic management
consolidation with high-dose chemotherapy of extremity, bladder-prostate, and paramenin-
(megatherapy) with autologous bone marrow or geal RMS and is practiced in select centers where
peripheral blood stem cell rescue (“bone marrow expertise is available. In metastatic disease (to
or stem cell transplant”) has also been tried with lungs, bones), palliative RT provides a modest
no major success. benefit, and there is recent interest in the use of
hypofractionated stereotactic ablative body
50.6.3.2 Principles of Radiation radiotherapy (SABR) in such scenarios.
Therapy There are problems associated with RT, and
RT is an important component of the multimo- they depend on the site irradiated. In the head and
dality management of patients with RMS as it neck region, xerostomia, cataract, leukoencepha-
improves local disease control and outcomes. In lopathy, dental problems, and vision and hearing
general, all patients, except low-risk subset A loss may be seen. RT to the abdomen and pelvic
(CG I), receive RT. RT may be considered as an regions may cause growth retardation, bowel
adjunct to surgery in the case of microscopic resi- obstruction, colitis, proctitis, and infertility, while
due (CG II) or gross residue (CG III) following that to the extremities may lead to fractures,
biopsy, surgical resection, or neoadjuvant chemo- growth retardation, fibrosis with limitation of
therapy. In certain sites, e.g., parameningeal function, atrophy, and peripheral nerve damage.
RMS, orbital RMS, and bladder-prostate RMS, Overall, the greatest long-term problem is the
RT is preferred to surgery for local disease con- development of second malignancies. Therefore,
trol, with the aim of achieving organ preserva- the importance of avoiding RT if possible or at
tion. The RT dose ranges from 36 to 54 Grays least limiting the radiation to surrounding
(Gy) and depends on the residual disease. RT vol- tissues.
ume is determined prior to surgical resection of
the primary tumor based on the extent of tumor at 50.6.3.3 Principles of Surgery for RMS
diagnosis. In patients with node-positive disease, Resection of primary tumor: The goal of surgery
the involved lymph node region or site is included for RMS is complete removal of the tumor with a
in the radiation portal. rim of normal tissue, preserving the cosmesis and
RT techniques have gradually evolved from function as much as possible (Wexler LH et al.
external beam two-dimensional conventional RT 2011). Complete removal with no microscopic
to 3D-conformal RT (3D-CRT) to image-guided disease offers the best chance of a cure. There is
intensity-modulated RT (IG-IMRT), with a view no role for debulking surgery. The surgical
to improving the coverage and conformity of the approach depends on the primary site, size, pres-
target volume with simultaneous sparing of the ence of lymph nodes, and distant metastases. At
adjacent critical organs at risk. However, even resection, if positive margin is suspected, biopsy
advanced techniques, e.g., IG-IMRT, may be of the margin should be performed. Unresectable
associated with significant intermediate low-dose microscopic or gross residual disease should be
bath, and this may be circumvented by the use of marked with titanium clips in the tumor bed so as
proton beam therapy, specifically intensity-­ to direct re-excision and later RT if required.
modulated proton therapy (IMPT). This approach Most tumors are usually unresectable at presenta-
may be worthwhile in tumors in critical loca- tion and so will receive chemotherapy, with or
tions, e.g., parameningeal RMS, where important without RT, for achieving a reduction in size for a
structures, e.g., the eyes, optic structures, pitu- safe and complete surgery.
itary gland, brainstem, and spinal cord, may be Primary re-excision (PRE) consists of a com-
closely juxtaposed to the target volume. High plete re-excision of a prior operative site with
cost and limited availability are some of the pathologically confirmed negative margins
major limitations of proton beam therapy at pres- (Wexler LH et al. 2011). It is prior to the institu-
50  Soft Tissue Sarcomas
tion of any other adjuvant therapy. PRE is recom-
mended for those in whom only a biopsy was
performed for a resectable tumor or if a non-
oncologic surgical excision was done and if the
status of margins is unclear. In localized lesions
of the trunk and extremities, PRE can lead to an
improved survival.
Lymph node (LN) evaluation is important for
the planning of treatment and also for overall out-
come, as positive LN is an important independent
poor prognostic factor for both failure-free sur-
vival (FFS) and overall survival (OS). Regional
LN should be assessed both clinically and radio-
logically and, for sites like the extremities, patho-
logically also. Any suspicious LN requires
pathologic confirmation. Nodal metastasis is rare
in head and neck disease (3%), and so routine
regional LN biopsy is unnecessary, and only
enlarged nodes should be biopsied. In RMS
involving the extremities, the incidence of nodal
metastases is high (40–50%), and so these should
have routine pathologic evaluation of the drain-
ing nodes: axillary nodes for upper extremity and
inguinal nodes for lower extremities. Nearly 17%
of the nodes that are clinically negative can be
pathologically positive, and so even if no nodes
are detected clinically or radiologically, they
need to be biopsied. Complete LN removal has
no therapeutic benefit. In the paratesticular RMS,
the incidence of LN metastases is around
25–30%. At this site, if the retroperitoneal nodes
(iliac and paraaortic/paracaval) are greater than
2 cm on CT scans, they are considered positive
and staged accordingly and need not be biopsied.
If the nodes are <2  cm, especially in children
above 10 years of age, they should be biopsied.
RP nodes above the level of renal hilum are con-
sidered as distant metastases (stage 4 disease).
Second-look operation (SLO): Following ini-
tial multiagent neoadjuvant chemotherapy, with
or without RT, repeat imaging is performed fol-
lowed by surgical exploration and excision of the
residual primary tumor. This is called second-­
look operation (SLO) (Wexler LH et al. 2011).
SLO can reclassify a radiologic partial response
(PR) to histologic complete response (CR) in
75% cases, and this may eliminate the need for
additional local measures like RT.  Also, about
655
10–12% cases of radiologic CR may be found to
have residual viable tumor and, therefore, be
reclassified as PR, and these may require addi-
tional local therapy. SLO may even permit dose
reduction of RT for patients that were initially
CG III.  SLO is most effective in RMS of the
extremities and trunk and least useful in head and
neck regions.
Surgery for metastatic disease and relapses:
Twenty percent of RMS cases have metastatic
disease at presentation, and resection of metasta-
ses at presentation is rarely indicated. Metastatic
disease could involve the lungs (58%), regional
LN (33%), bone (32%), liver (22%), and brain
(20%). Therefore, at the time of presentation, a
staging work-up should include evaluation of all
these sites. Visceral metastases to sites, such as
the liver and brain, are never the only site of
metastases, and so it is suggested that they be
evaluated only if the lungs and/or LN are found
to be involved. Nearly 30% of RMS patients
relapse, and 50–90% of these relapsed patients
ultimately die of progressive disease (Aydin B et
al. 2018).
Surgical principles for various sites: Tumors
of the head and neck regions are rarely amenable
for upfront surgical resection, and so an inci-
sional biopsy is done. Routine regional LN
biopsy is not required. For orbital tumors, a
biopsy followed by chemoradiotherapy is mostly
all that is required. For all other sites, surgical
excision may be required after tumor reduction is
achieved with chemotherapy and radiotherapy.
Regional LN dissection is not done, except for
alveolar histology. Lymph nodal metastatic
regions, if any, are included in radiation portal.
The surgical approach for RMS of the bladder
and prostate has evolved from pelvic exenteration in
the 1960s and 1970s to the current organ-­preserving
surgeries that have been made feasible with the cur-
rent multiagent chemotherapy and RT.  Upfront
resection at these sites is reserved for a small pro-
portion of patients who have a tumor involving the
dome of the bladder, in whom the bladder and ure-
thral functions can be preserved. Total cystectomy
and anterior pelvic exenteration are now only rec-
ommended for those patients who fail to respond to
induction chemotherapy and RT. After chemother-
656 S. Agarwala and R. Carachi

apy and RT, a number of patients with bladder/pros-
tate RMS may not require extensive resections.
Even with bladder-­preserving surgeries, the long-
term bladder functions remain suboptimal in a num-
ber of survivors with urinary incontinence,
frequency, nocturia, and high pressure systems
leading to subsequent renal damage.
All paratesticular tumors need to be resected,
with the entire spermatic cord, through an ingui-
nal incision, either upfront or following neoadju-
vant chemotherapy. Any biopsy or excision
through the scrotal route should be avoided, as it
will alter the lymphatic drainage basin and will
require hemiscrotectomy. LN are considered
involved if they are enlarged radiologically or
clinically, and pathological confirmation biopsy
is not required. RPLN biopsy and ipsilateral RP
lymph node dissection are required only for chil-
dren more than 10 years of age. For vulval, vagi-
nal, and uterine tumors, organ preservation is
important, and so primary resection has a very
limited role. Surgical resection is reserved for
those who fail to achieve CR (radiographic) or
have early disease progression on induction che-
motherapy and RT. Residual tumor of the uterus
and proximal vagina may mandate hysterectomy,
but distal vaginal preservation is nearly always
feasible. Vaginal reconstruction may be required
if vaginectomy is performed.
For RMS of the extremities, upfront excision
should only be done for small tumors that can be
excised completely with negative margins and
will not lead to major compromise of the func-
tion. All others should only have an incisional
biopsy. Regional draining LN should always be
sampled during the initial biopsy, even if they are
clinically and radiologically not involved. The
concept of sentinel LN mapping and guided
biopsy has also been described. Most patients
will need excision of the tumor following neoad-
juvant chemotherapy. Amputation may be
required for those patients who fail to respond or
in whom extensive tumors are involving the bone
or neurovascular structures. Involvement of axil-
lary nodes for upper extremity tumors and iliac or
paraaortic nodes for lower extremity tumors is
considered distal metastasis (stage 4).
Complications of surgery: Radical LN dissec-
tions are not recommended as this leads to scarring
and lymphedema. There is no convincing evidence
that radical LN dissections obviate the need for RT
or even decrease its dosage. It does not improve
outcomes. Resection in the head and neck region
could result in major disfigurement unless per-
formed skillfully. Resection done in the bladder
base/urethra region can result in voiding difficul-
ties and issues with urinary continence requiring
bladder augmentation and/or need for CIC. RPLN
dissection in cases of paratesticular RMS can
result in ejaculatory dysfunction. Resection of
extremity tumors can result in significant limb
dysfunction at times requiring use of prosthesis.
50.7 Non-rhabdomyosarcoma
Soft Tissue Sarcoma (NRSTS)
NRSTS are rare tumors in children and constitute
less than half of all STS in children (Van der Graaf
WTA et al. 2017). NRSTS, like RMS, can arise in
any part of the body, but the most common sites
are the extremities, trunk, abdomen, and pelvis
(Table 50.6). The most frequent histologic types
are synovial sarcoma, neurofibrosarcoma, and
fibrosarcoma. In the extremities, it is commoner
in the lower extremity (Fig.  50.10), and in the
trunk, the chest wall is the commonest site. Most
present as painless masses, and around 15% are
Table 50.6  Some common and uncommon NRSTS seen
in children
Common Rare
Synovial sarcoma Angiosarcoma
Fibrosarcoma Hemangiopericytoma
Peripheral neuroectodermal Clear cell sarcoma
tumor
Malignant fibrous Epithelioid cell sarcoma
histiocytoma
Leiomyosarcoma Intra-abdominal
desmoplastic tumor
Neurogenic tumors Malignant
mesenchymoma
 • Malignant
schwannoma
 • Neurofibrosarcoma
 • Malignant peripheral Extra-skeletal
nerve sheath tumor chondrosarcoma
50  Soft Tissue Sarcomas 657

Fig. 50.10 (a) Synovial sarcoma of the lower extremity. (b) CT scan of the patient

metastatic at diagnosis. Spread to regional LN is
unusual except in certain histologic subtypes,
such as epithelioid sarcoma and clear cell sar-
coma, which themselves are extremely rare in
children. The commonest site for metastasis is the
lungs and brain. Bone, liver, subcutaneous tissues
are other rare sites of metastases. Bone marrow
involvement is extremely rare. While most of the
extremity NRSTS in children are synovial sarco-
mas, tumors of the trunk are predominantly
malignant fibrous histiocytomas (MFH) or neuro-
genic in origin like malignant peripheral nerve
sheath tumor, neurofibrosarcoma, and malignant
schwannoma.
658
Table 50.7  Schema for grading of NRSTS in children
Grade 1
 • Myxoid well-differentiated liposarcoma
 •  Deep-seated dermatofibrosarcoma protuberans
 • Well-differentiated or infantile (age < 5 years)
fibrosarcoma
 • Well-differentiated or infantile (age < 5 years)
hemangiopericytoma
 • Well-differentiated malignant peripheral nerve
sheath tumor
 • Angiomatoid malignant fibrous histiocytoma
Grade 2
 • Sarcomas not specifically included in grade 1 or 3;
<15% of tumor showing geographic necrosis or the
mitotic index is <5 per 10 high-power fields
Grade 3
 •  Pleomorphic or round cell liposarcoma
 •  Mesenchymal chondrosarcoma
 •  Extra-skeletal osteosarcoma
 •  Malignant triton tumor
 •  Alveolar soft part sarcoma
50.7.1 Histology and Grading
of NRSTS
Most of the NRSTS in children, except for the
MFH and fibrosarcomas, are immature and
poorly differentiated with approximately half of
the tumors having a histologic grade 3. Because
of the inconsistencies in predicted behavior, a
grading scheme for pediatric NRSTS is used
(Table 50.7), which takes into account the cyto-
histologic features that are used for adult sarco-
mas, but with caveats of the childhood lesions
(Table 50.7). This system is not used for RMS or
for primitive neuroectodermal lesions, which
are always considered high-grade tumors.
50.7.2 Prognostic Factors for NRSTS
Because of the rarity of NRSTS in children, the
data on prognostic factors is basically derived
from small single-institution studies (van der
Graaf WTA 2017). These include extent of dis-
ease (metastatic vs non-­metastatic), histologic
grade (low vs high), size of primary tumor (≤
5 cm vs > 5 cm), and extent of resection (resect-
able vs unresectable). On the basis of these
four factors, NRSTS are risk categorized into
high-, intermediate-, and low-risk disease
S. Agarwala and R. Carachi
(Waxweiler TV et al. 2015). High-risk are met-
astatic disease, where survival is 15% and most
die of progressive metastatic disease.
Intermediate-risk are the non-­metastatic, unre-
sectable tumors or have both high-grade histol-
ogy and are >5  cm in maximum diameter.
Survival in this group is around 50%, and most
die of progressive local disease. Low-­ risk
include non-metastatic resectable tumors that
are either high-grade histology and < 5 cm or
low-grade histology (any size). They have a
survival of around 90%.
50.7.3 Clinical Evaluation, Diagnosis,
and Staging of NRSTS
In all children suspected to have NRSTS, the clin-
ical evaluation should include routine hemo-
grams, renal and liver function tests, bone scans,
and bone marrow examination. MRI scan is
­considered the imaging modality of choice for the
evaluation of local and regional disease, particu-
larly in the extremities, the pelvis, and the head
and neck regions. The staging system currently
used for NRSTS is the same as the modified TNM
staging system used for RMS (Table  50.4). In
addition, histological grading is used (Table 50.7).
Although fine needle aspiration cytology (FNAC)
is a useful diagnostic tool in the initial evaluation
of NRSTS or a possible metastatic lesion, needle
core biopsy (NCB) is better for providing enough
tissue to permit accurate histologic subtyping,
cytogenetics, and molecular studies. Excisional
biopsy rarely should be used in the initial evalua-
tion of these tumors. Simple excision of the tumor
violates the tissue planes and results in the dis-
semination of the tumor cells throughout the oper-
ative field. Subsequent surgery in the region is
thereby compromised. An excisional biopsy is
undertaken only in those instances when the
tumor is small (<2.5  cm) or situated so that an
eventual wide local resection can be done without
risk or functional deformity. When an incisional
biopsy is done, it should be planned properly. For
an extremity lesion, the incision should be planned
longitudinally or parallel to the neurovascular
bundle. As LN metastasis in NRSTS is rare, rou-
50  Soft Tissue Sarcomas
tine evaluation of LN is not required. The inci-
dence of LN involvement is around 15% in
synovial sarcoma. Other high-grade lesions, like
angiosarcoma, epithelioid sarcoma, and clear cell
sarcoma, may also have LN involvement, but
these histologies are extremely rare in the pediat-
ric age group. Formal nodal dissection even when
nodes are positive, but the area is included in the
RT field.
50.7.4 Management of NRSTS
Surgery: The treatment is multidisciplinary and is
similar to adults with few differences. The first
concern is always how best to achieve local con-
trol. If feasible, wide local excision (WLE)
should be performed. The current COG NRSTS
protocol defines negative margins as at least
0.5 cm of non-malignant tissue around the surgi-
cal specimen. When the tumor abuts the perios-
teum or fascia and is removed in continuity with
the tumor, the margins are considered negative.
For unresectable tumors, neoadjuvant chemo-
therapy or RT or both can be used. Adjuvant che-
motherapy is for cases with marginal resections.
Complete surgical resection at diagnosis is
important as demonstrated by a report from St.
Jude Children’s Research Hospital. They reported
that the 5-year survival was 89%, with 12.8%
local failure rate and 11.8% distant failure rate
when complete surgical resection was done. In
contrast, only half of the patients with unresect-
able disease and 34% of those with metastatic
disease were alive at 5 years and 2 years, respec-
tively. If the initial surgical resection was not per-
formed anticipating malignancy, or if adequacy
of resection is questionable, a PRE is mandated
to ensure negative margins as this is the most
important factor for good outcome. Besides
extremities, NRSTS at all other sites are even
more difficult to excise completely. For patients
with pulmonary metastases, even if all the visible
metastases are resected, long-term cures are
observed in <10% cases.
Radiation therapy: There are no standard RT
guidelines for pediatric NRSTS, and external
beam radiation therapy (EBRT) is the standard
659
modality. Post-operative RT is considered if sur-
gical resection is less than WLE, if the surgeon
did not anticipate a sarcoma at initial resection, if
no pre-operative imaging is available, if multiple
attempts at resection were done, if the margins
are unknown, and if WLE has resulted in micro-
scopic positive margins and re-excision is not
done. Sarcomas that are not amenable to resec-
tion without mutilating surgery like amputation
may also benefit from pre-operative
RT. Brachytherapy is frequently given as a boost
to EBRT to deliver high doses of radiation to a
focal area within the operative bed and when sur-
gical margins are difficult to achieve as in extrem-
ity neurovascular bundle.
Chemotherapy: The guidelines for chemother-
apy in pediatric NRSTS are not well established.
Most NRSTS are not very chemosensitive.
Chemotherapy is indicated when the tumor is
unresectable and for large high-grade resected
tumors that are high risk for metastatic recur-
rence. Two agents that have shown some activity
against NRSTS are doxorubicin and ifosfamide,
especially in patients with synovial sarcoma.
Both COG and EpSSG have protocols consisting
of these two drugs. mTOR and PI3K inhibitors
might hold some promise for malignant periph-
eral nerve sheath tumor (MPNST), while suni-
tinib has shown some activity in alveolar soft part
sarcomas and desmoplastic round small cell
tumors (DRSCT). Sorafenib has some activity in
angiosarcomas. Imatinib has been used success-
fully in dermatofibrosarcomas.
50.8 Conclusions
Soft tissue sarcomas are an uncommon malig-
nancy of childhood, constituting about 6% of all
malignancies of children. A larger proportion of
these sarcomas in childhood are RMS, while a
smaller proportion are the NRSTS.  Both RMS
and NRSTS arise from a varied location, and the
clinical presentation is dependent on the location
of these tumors. The commonest site of disease
among the RMS is the head and neck region and
the urinary system, while it is the extremities and
trunk for NRSTS. Most sarcomas initially pres-
660
ent as a palpable mass or due to the effect of such
a mass in a confined space like proptosis in the
orbit or urinary obstruction in the RMS of the
prostate. Treatment of sarcomas is multimodal
including surgery, chemotherapy, and radiation
therapy. RMS is a chemosensitive tumor, and
therefore, most patients are treated with neoadju-
vant chemotherapy prior to local therapy in the
form surgery and/or radiation therapy. On the
other hand, most NRSTS are not very chemosen-
sitive, and complete surgical resection is mostly
required for cure. Currently, local control is
achieved in 80–90% of RMS patients, and
approximately 70% of the patients survive for
5 years or more and are probably cured. Among
the NRSTS, low-risk resectable tumors have a
survival of around 90%, while the high-risk, met-
astatic disease have a survival of 15%, and most
die of progressive metastatic disease.
References
Arndt CAS, Stoner JA, Hawkins DS, Rodeberg DA,
Hayes-Jordan AA, Paidas CN et al (2009) Vincristine,
actinomycin, and cyclophosphamide compared with
vincristine, actinomycin, and cyclophosphamide alter-
nating with vincristine, topotecan, and cyclophos-
phamide for intermediate-risk rhabdomyosarcoma:
Children’s Oncology Group study D9803. J Clin
Oncol 27:5182–5188
Austin MT, Andrassy RJ (2008) Soft tissue sarcoma. In:
Carachi R, Grosfeld JL (eds) The surgery of childhood
tumors, 3rd edn. Springer, Heidelberg, p 345
Aydin B, Akyuz C, Varan A, Yalcin B, Kurucu N, Kutluk
T (2018) ICE regimen for relapsed/refractory bone
and soft tissue sarcomas in children. Rev Recent Clin
Trials 13(2):126–131
Dyson KA, Stover BD, Grippin A, Mendez-Gomez HR,
Lagmay J, Mitchell DA, Sayour EJ (2019) Emerging
S. Agarwala and R. Carachi
trends in immunotherapy for pediatric sarcomas. J
Hematol Oncol 12(1):78
Lautz TB, Hayes-Jordan A (2019) Recent progress in
pediatric soft tissue sarcoma therapy. Semin Pediatr
Surg 28(6):150862
Malempati S, Hawkins DS (2012) Rhabdomyosarcoma:
review of the Children’s Oncology Group (COG) soft-­
tissue sarcoma committee experience and rationale for
current COG studies. Pediatr Blood Cancer 59:5–10
Minn Y, Lyden ER, Anderson JR, Million L, Arndt CA,
Brown K et  al (2010) Early treatment failure in
intermediate-­ risk rhabdomyosarcoma: results from
IRS-IV and D9803—a report from the Children’s
Oncology Group A. J Clin Oncol 28:4288–4232
Pappo AS, Lyden E, Breitfeld P, Donaldson SS, Wiener
E, Parham D et  al (2007) Two consecutive phase II
window trials of irinotecan alone or in combination
with vincristine for the treatment of metastatic rhabdo-
myosarcoma: the Children’s Oncology Group. J Clin
Oncol 25:362–369
Raney R, Walterhouse DO, Meza JL, Andrassy RJ,
Breneman JC, Crist WM et  al (2011) Results of the
intergroup rhabdomyosarcoma study group D9602
protocol, using vincristine and dactinomycin with or
without cyclophosphamide and radiation therapy, for
newly diagnosed patients with low-risk embryonal
rhabdomyosarcoma: a report from the soft tissue sar-
coma committee of the Children’s Oncology Group. J
Clin Oncol 29:1312–1318
van der Graaf WTA, Orbach D, Judson IR, Ferrari A
(2017) Soft tissue sarcomas in adolescents and young
adults: a comparison with their paediatric and adult
counterparts. Lancet Oncol 18(3):e166–e175
Waxweiler TV, Rusthoven CG, Proper MS, Cost CR, Cost
NG, Donaldson N, Garrington T, Greffe BS, Heare T,
Macy ME, Liu AK (2015) Non-rhabdomyosarcoma
soft tissue sarcomas in children: a surveillance, epi-
demiology, and end results analysis validating COG
risk stratifications. Int J Radiat Oncol Biol Phys
92(2):339–348
Wexler LH, Meyer WH, Helman LJ (2011)
Rhabdomyosarcoma. In: Pizzo PA, Poplak DG (eds)
Principles and practice of pediatric oncology, 9th edn.
Wolters Kluwers, Philadelphia
 Lymphomas
Christian Urban
51.1 Introduction
Treatment outcome for children and adolescents
with Hodgkin lymphoma (HL) and non-Hodgkin
lymphoma (NHL) has dramatically improved
during the last decades. Within clinical multi-
center trials, cure rates between 70% and over
90% are being achieved by using chemotherapy
alone or combined with radiotherapy in certain
cases (Dörffel et al. 2013; Landmann et al. 2017).
Major challenges in the future are further optimi-
zation of primary treatment to avoid relapse and
to reduce acute and long-term toxicities
(Johannsdottir et al. 2017).
51.2 Historical Overview
In 1832, Thomas Hodgkin (1798–1866) reported
autopsy findings of seven patients with lym-
phoma who had shown swollen lymph glands
and an enlarged spleen, without evidence of
tuberculosis, purulent inflammation, or cancer.
Later, the diagnosis of “Hodgkin’s disease” was
restricted to lymphomas with giant, multinucle-
ated Reed-­Sternberg cells on microscopic exami-
nation. Thomas Hodgkin also introduced modern,
C. Urban (*)
Department of Pediatric Hematology/Oncology,
University Clinic of Pediatrics and Adolescent
Medicine, Medical University of Graz, Graz, Austria
e-mail: christian.urban@medunigraz.at
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_51
51
organ-­based medicine in England, together with
the clinicians Richard Bright and Thomas
Addison. He died of dysentery on a journey to
Palestine and lies buried in Jaffa (van Gijn and
Gijselhart 2012.)
Since the discovery of a virus in 1964 by
Anthony Epstein, Bert Achong, and Yvonne Barr,
EBV has been detected in Burkitt lymphoma,
Hodgkin lymphoma, posttransplant lymphopro-
liferative disorders, and various other malignan-
cies leading to an intensive search of other tumor
causative viruses (Epstein 2019; Morales-­
Sánchez and Fuentes-Pananá 2014).
51.3 Incidence
The estimated incidence in European children
with NHL under 15  years of age in 1988–1997
was 9.4 per million, while in adolescents aged
15–19 years, it was 15.9 per million (Izarzugaza
et al. 2006). Incidence in European children with
HL in 1988–1997 was estimated at 5.8 per mil-
lion under 15 years of age and at 29.7 per million
in adolescents (Clavel et  al. 2006). In general,
lymphoma is the most common malignancy in
children aged 15–19 years.
661
662
51.4 Etiopathogenesis
Three forms of Burkitt lymphoma (BL) exist
(sporadic, endemic, immunodeficiency associ-
ated), and the endemic form was first discovered
as being driven by the Epstein-Barr virus in areas
of the world where malaria, malnutrition, and
parasites are prevalent. BL has the characteristic
t(8;14) cytogenetic translocation that leads to
constitutive activation of the MYC gene and
IgH-c/MYC rearrangement which drives BL cell
division. Characteristic cytogenetics is also
detected in sporadic BL cases (Greenough and
Dave 2014) and immunodeficiency-associated
cases (Sperl et  al. 2012). EBV is also found in
Hodgkin lymphoma (HL), although its onco-
genic role is still disputed (Murray and Young
2019).
Table 51.4 enumerates the various genes
involved in Burkitt and Burkitt-like lymphomas
as well as in the other pediatric NHL making
them attractive for targeting them as, for exam-
ple, anaplastic large cell lymphoma (ALCL) with
the ALK inhibitor crizotinib (Mossé et al. 2017).
Fig. 51.1  A 5-year-old boy, who had clinical examina-
tion because of abdominal pain and distention.
Ultrasonography and MRI showed massive thickening of
the bowel walls with some small areas of ascites between
it. Fine needle aspiration showed typical L-3 lympho-
C. Urban
51.5 Non-Hodgkin Lymphoma
(NHL)
51.5.1 Diagnosis
The various subgroups of childhood NHLs have
different clinical behavior and are also character-
ized by the occurrence in different parts of the
body (Table 51.4).
Most frequent involved sites are the mediasti-
num (mainly lymphoblastic precursor T-NHL),
abdomen (mainly Burkitt and Burkitt-like lym-
phoma) (Fig.  51.1), and cervical lymph nodes
(Fig.  51.2) (mainly lymphoblastic precursor
B-NHL). Less frequent sites are Waldeyer’s ring,
skin, or bone and rarely the CNS (Fig. 51.2). Even
though patients may appear to have local disease,
there is often submicroscopic dissemination in
certain histologic subtypes to the bone marrow
and the CNS, and some patients, particularly with
Burkitt’s disease or mediastinal T-NHL, may have
a rapid and aggressive course. Involvement of
organs other than the lymphoid and hematopoietic
system (bone, CNS, skin) is rare.
blasts with intensively basophilic cytoplasm and distinc-
tive cytoplasmic vacuoles. Immunophenotyping and
molecular studies confirmed the diagnosis of Burkitt
lymphoma
51 Lymphomas
Fig. 51.2  A 15-year-old boy, complaining about difficul-
ties in swallowing. Physical examination revealed right
cervical lymphadenopathy with massive involvement of
Waldeyer’s ring. MRI showed that the tumor mass infil-
trated through the base of the skull with involvement of
51.5.2 Diagnostic Work-Up
and Staging
To establish the diagnosis of suspected NHL in
children, the least invasive procedure should be
preferred so that the risk of general anesthesia
may be avoided. Before surgery is considered,
examination of blood and bone marrow and in the
case of pleural effusion/ascites puncture (prefer-
ably under local anesthesia) with cytologic and
immunophenotypic analysis should be done
(Mann et al. 2006).
Only if the diagnosis cannot be established
with these simple methods should a diagnostic
biopsy be performed. Complete resection should
be done only if possible and without any risk or
functional loss for the patient. These principles
are particularly important in patients with high
tumor burdens, resulting in significant respira-
tory distress, vena cava compression, pericardial
tamponade, and metabolic disturbances
(Fig.  51.3). Particularly in certain patients with
mediastinal NHL and significant airway narrow-
ing, all diagnostic and especially invasive diag-
nostic procedures should be postponed after a
663
the CNS.  Since CNS fluid and bone marrow aspiration
were non-diagnostic, biopsy of the cervical lymph node
under local anesthesia was done, demonstrating diffuse
large B-cell lymphoma
prediagnostic cytoreductive therapy with predni-
sone up to 48 h until clinical stabilization. Under
no circumstances should a critically large medi-
astinal tumor with clinical symptoms of respira-
tory distress be treated surgically. Caution should
be also given to the tumor lysis syndrome in
patients with high tumor burdens, even before the
start of cytoreductive treatment.
Further staging evaluation must be done
quickly since most children with NHL have rap-
idly growing tumors that may cause life-­
threatening situations. It includes physical
examination, complete blood count, bone mar-
row and spinal fluid aspiration, analysis of elec-
trolytes, LDH, renal and liver function tests, and
imaging with X-ray, ultrasonography, magnetic
resonance, computerized tomography, and skele-
tal scintigraphy.
The most widely used clinical staging system
was developed at the St. Jude Children’s Research
Hospital in Memphis, USA (Murphy 1980)
(Table 51.1). For patients with Burkitt NHL, the
St. Jude staging system has been adapted by the
French Society of Pediatric Oncology (SFOP) to
a system which stratifies the patients to different
664 C. Urban

Fig. 51.3  A 4-year-old boy with swelling and plethora of
the face, cervical lymphadenopathy, and respiratory dis-
tress in lying position. Chest X-ray showed a huge medias-
tinal tumor with pleural effusion on the right side. Pleural
fluid aspiration in upright position with a thin needle using
local anesthesia revealed convoluted lymphoblasts of T-cell
type, positively stained with the acid phosphatase reaction
and confirmed by immunophenotyping. After initiation of
prednisolone, there was a rapid tumor response with regres-
sion of the existing superior vena cava syndrome. Further
treatment with a T-NHL/ALL-polychemotherapy protocol
led to a complete and permanent remission

Table 51.1  St. Jude Children’s Research Hospital staging system for non-Hodgkin lymphoma (Murphy 1980)
Stage I
A single tumor (extranodal) or single anatomic area (nodal) with the exclusion of the mediastinum or abdomen
Stage II
A single tumor (extranodal) with regional node involvement
Two or more nodal areas on the same side of the diaphragm
Two single (extranodal) tumors with or without regional node involvement on the same side of the diaphragm
A primary gastrointestinal tract tumor, usually in the ileocecal area, with or without involvement of associated
mesenteric nodes only, grossly completely resected
Stage III
Two single tumors (extranodal) on opposite sides of the Diaphragm
Two or more nodal areas above and below the diaphragm
All primary intrathoracic tumors (mediastina, pleural, thymic)
All extensive primary intra-abdominal disease
All paraspinal or epidural tumors, regardless of other tumor site(s)
Stage IV
Any of the above with initial CNS and/or bone marrow involvement
51 Lymphomas 665

Table 51.2  Risk stratification of B-cell non-Hodgkin
lymphomas (SFOP) (Patte et al. 2001)
Group Extent of tumor
A Resected stage I and abdominal stage II
B Unresected stage I, non-abdominal stage II, any
stage III or IV, B-ALL-CNS negative (with
<70% blasts in BM)
C CNS involvement or B-ALL with at least 70%
blasts in the bone marrow
SFOP, French Society of Pediatric Oncology
The stratification system utilizes the clinical stage
assigned according to the St. Jude staging system
Table 51.3  Staging system according to different risk
groups in B-cell non-Hodgkin lymphomas used by the
BFM group (Reiter et al. 1999, 2000)
Risk Initial complete resection of the lymphoma
group 1:
Risk No or incomplete resection of lymphoma
group 2: manifestation and one of the following
criteria: only extra-abdominal sites or
abdominal sites and LDH less than 500 U/l
Risk No or incomplete resection of abdominal
group 3: lymphoma and LDH > 500 U/L, all patients
with bone marrow involvement or/and CNS
disease, or/and multifocal bone involvement
treatment arms tailored to their risk and cumula-
tive tumor burden (Patte et al. 2001) (Table 51.2).
A similar staging system according to the differ-
ent risk groups in Burkitt NHL patients is being
used by the BFM group (Reiter et al. 1999, 2000)
(Table 51.3).
51.5.3 Pathophysiology
EBV is a key player in the pathophysiology of
Burkitt and Burkitt-like NHL leading to a
chromosomal translocation juxtaposing the
c-MYC oncogene on chromosome 8 most fre-
quently to the immunoglobulin heavy chain
locus on chromosome 14. Table 51.4 shows the
various T-cell antigen receptor gene rearrange-
ments as drivers in the disease process of other
pediatric NHL.  In anaplastic large cell lym-
phoma (ALCL), targeting rearrangements
involving the ALK gene with the ALK inhibi-
tor crizotinib has become a new therapeutic
option (Mossé et al. 2017).

Table 51.4  Correlation of histopathology, immunophenotype, clinical features, and cytogenetic and molecular fea-
tures in childhood
Non-Hodgkin
lymphoma*
Genes
Histology Immunology Clinical features Cytogenetics involved
Burkitt and B cell (slg+) Abdominal masses, t(8;14)(q24;q32) lgH-cMYC
Burkitt-like
Gastrointestinal t(2;8)(p11;q24) lgK-cMYC
Tract tumors, involvement of t(8;22)(q24;q11) lgX-cMYC
Waldeyer’s ring
Diffuse large B cells of germinal Abdominal masses, t(8;14)(q24;q32) lgH-cMYC
B-cell (DLBCL) center or post-germinal gastrointestinal tract tumors, t(2;17)(p23;q23) CLTC-ALK
center involvement of
Waldeyer’s ring
Mediastinal large B cells of medullary Mediastinum
B-cell thymus
Anaplastic large T cell (mostly), null cell, Skin, nodes, bone t(2;5)(p23;q35) NPM-ALK
cell
(ALCL) Or NK cell (CD30+) t(1;2)(q21;p23) TPM3-ALK
t(2;3)(p23;q21) TFG-ALK
t(2,17)(p23;q23) CLTC-ALK
t(X;2)(q11–12;p23) MSN-ALK
Inv 2 (p23;q35) ATIC-ALK
(continued)
666 C. Urban

Table 51.4 (continued)
Non-Hodgkin
lymphoma*
Genes
Histology Immunology Clinical features Cytogenetics involved
Precursor T T cell (thymocyte Anterior mediastinal t(1;14)(p32;q11) TCRα′
lymphoblastic phenotype) --TAL1
(pT-LBL) Mass with upper t(11;14)(p13;q11) TCRα′ --
Torso adenopathy t(11;14)(p15;q11) RHOMB2
t(10;14)(q24;q11) TCRα′ --
t(7;19)(q35;p13) RHOMB1
t(8;14)(q24;q11) TCRα′
--HOX11
t(1;7)(p34;q34) TCRI-LYL1
TCRα′
--MYC
TCRα′
--MYC
TCRI-LCK
Precursor B lymphoblastic B-cell precursors Cutaneous masses and (pB-LBL) isolated lymph node masses
*Reprinted with permission from P.A. Pizzo & D.G.Poplack: Principles and Practice of Pediatric Oncology, fifth edition
2006, Lippincott Williams & Wilkins, Philadelphia

51.5.4 Pathology expressing surface immunoglobulin, whereas

Early pathology classifications in use have been
confusing for clinicians and also led to disagree-
ment even between expert pathologists, since
NHL is a heterogenous collection of diseases and
their description was solely on cytomorphologic
features. Immunophenotyping, cytogenetic, and
molecular studies now allow for a more precise
classification according to the lineage of the
malignant cells (WHO and REAL classification)
(Harris et al. 1994; Jaffe et al. 2001). In contrast
to the wide diversity of adulthood NHL,
­childhood NHL can mainly be divided among
four major subgroups (Table  51.4): Burkitt and
Burkitt-like lymphomas, diffuse large B-cell
lymphomas, anaplastic large cell lymphomas,
and lymphoblastic precursor T- and B-cell lym-
phomas. The malignant cells in childhood NHL
appear to arise from different lymphocyte precur-
sors at various stages of maturation or from
mature lymphocytes (Magrath 1981).
Approximately 40–50% of childhood NHL are
either from T-cell lineage or from mature B cells
fewer than 10% are of early B-cell origin lacking
surface immunoglobulin (Table 51.4).
51.5.5 Treatment and Results
Nowadays, clinical trials are available for all
childhood NHL types, and participation in these
studies is strongly recommended which guaran-
tees treatment by an experienced multidisci-
plinary team of specialists. Chemotherapy is the
mainstay of treatment. It is the same in lympho-
blastic NHL of precursor B- and T-lineage as in
precursor B- and T-ALL as the distinction
between lymphoblastic NHL and ALL is largely
arbitrary and simply based on the percentage of
blasts in the bone marrow aspirate. CNS therapy
is included in these regimes particularly for chil-
dren with lymphoblastic and Burkitt NHL,
whereas the addition of radiotherapy does not
have any benefit on survival except in rare cases
(e.g., oncologic emergency) (Reiter et  al. 1999,
2000; Neth et al. 2000).
51 Lymphomas
Patients with diffuse large B-cell lymphomas
(DLBCL), mediastinal large B-cell lymphomas,
and anaplastic large cell lymphomas (ALCL) are
mainly treated on either protocols designed for
Burkitt lymphoma or similar protocols with simi-
larly good results (Seidemann et al. 2001, 2003).
Overall, more than 75% of children with NHL
can be now cured with modern intensive polyche-
motherapy protocols and due to advances in sup-
portive care to reduce the life-threatening
complications of NHL and of therapy. Patients
with relapsed, refractory, or advanced stage non-­
Hodgkin lymphoma still continue to have a poor
prognosis (Attabaschi et al. 2005). Novel cellular
and immunotherapies for these high-risk patients,
including hematopoietic stem cell transplant,
bispecific antibodies, viral-derived T cells, chi-
meric antigen receptor (CAR) T cells, and natural
killer (NK) cell therapy, are showing promising
results (Akbayram et al. 2010; Chu et al. 2019).
Fig. 51.4  A 12-year-old boy had extensive dermatologic
work-up because of itching and scratching. Chest X-ray
revealed polycyclic bilateral enlargement of hilar lymph
nodes. CT showed left cervical and right supraclavicular
lymph node enlargement, a bulky mediastinal mass with
667
51.6 Hodgkin Lymphoma (HL)
51.6.1 Diagnosis and Differential
Diagnosis
Most patients present with painless cervical or
supraclavicular enlarged lymph nodes which feel
rubbery and firm at palpation. Differential diag-
nosis should include inflammatory conditions,
especially with indolent course like atypical
mycobacteriosis and toxoplasmosis, but also
NHL, metastatic solid tumors (e.g., soft tissue
sarcoma, neuroblastoma, nasopharyngeal carci-
noma), histiocytoses, and benign malformations.
More than 60% of the patients with HL have also
mediastinal involvement, sometimes causing
symptoms of bronchial compression or superior
vena cava syndrome (Fig.  51.4). Axillary or
inguinal lymphadenopathy is less frequent at ini-
tial presentation. Unexplained fever, weight loss
anterior chest wall infiltration and a single left lung nod-
ule, as well as splenic involvement. Touch preparation
smear of the excised supraclavicular lymph node showed
typical Reed-Sternberg and Hodgkin cells
668
of more than 10% within the last 6 months, and
night sweats are considered as “B”-symptoms.
Pruritus is another systemic symptom, some-
times leading to enormous scratching (Fig. 51.4).
51.6.2 Diagnostic Work-Up
and Staging
After diagnosis has been established by exci-
sional biopsy, careful assessment of all lymph
node regions by clinical examination and ultraso-
nography is essential. In addition, complete
blood count, analysis of electrolytes, LDH, and
renal and hepatic function tests should be done.
The erythrocyte sedimentation rate, serum cop-
per, and ferritin may be elevated as well as
C-reactive protein, which can be used in follow-
­up evaluation (Wieland et  al. 2003). Computed
tomography (CT) of the chest provides further
information about the mediastinal lymph nodes
as well as lungs, chest wall, and pericardium. To
evaluate the abdominal and pelvic lymph nodes
and for diagnosis of splenic and hepatic involve-
ment, magnetic resonance imaging (MRI) may
provide better information in children with the
advantage of no radiation side effects. Positron
emission tomography (PET) is being used as a
promising new diagnostic tool for staging HL in
adults (Jerusalem et al. 1999); however, prospec-
tive trials evaluating PET in pediatric HL are still
awaiting (Kluge et al. 2013; Totadri et al. 2018;
Kurch et al. 2019). Since HL spreads along con-
tiguous lymph nodes, staging is based on the
natural course of the disease (Table  51.5)
(Carbone et al. 1971). Substage classification A
means “asymptomatic” disease; substage B
means fever, night sweats, and weight loss of
more than 10% over the last 6 months; and sub-
stage E involves extralymphatic disease. Bone
marrow aspiration with biopsy should be reserved
for patients with clinical stage III or IV or patients
with B-symptoms and technetium-99 bone scan
for patients with suspected skeletal metastases.
In the era when HL patients were only treated
and cured with mainly radiotherapy, it was
important to detect also minimal disease, so that
C. Urban
Table 51.5 Staging classification for Hodgkin lym-
phoma (Carbone et al. 1971)
Stage Involvement of a single lymph node (I) or of a
I single extralymphatic organ or site (IE)
Stage Involvement of two or more lymph node
II regions on the same side of the diaphragm (II)
or localized involvement of an extralymphatic
organ or site and one or more lymph node
regions on the same side of the diaphragm (IIE)
Stage Involvement of lymph node regions on both
III sides of the diaphragm (III), which may be
accompanied by involvement of the spleen
(IIIS) or by localized involvement of an
extralymphatic organ or site (IIIE) or both (IIISE)
Stage Diffuse or disseminated involvement of one or
IV more extralymphatic organs or tissues with or
without associated lymph node involvement
exploratory laparotomy and splenectomy were
the recommended staging procedure. Today, the
success of diverse non-cross-resistant chemo-
therapy cycles in conjunction with the refinement
in diagnostic imaging led to the abandonment of
surgical staging. Laparotomy or better laparo-
scopic surgery is now reserved for females, who
need transposition of the ovaries outside the irra-
diation field in the case of pelvic irradiation.
Since the alkylating chemotherapy needed for
higher HL stages is harmful to gonadal function,
sperm cell cryopreservation and ovarian tissue
cryopreservation should also be considered
(Balcerek et al. 2020; Seshadri et al. 2006; Van
der Kaaij et al. 2009, 2010).
51.6.3 Pathophysiology
Hodgkin lymphoma (HL) is a tumor derived from
B cells composed by mononucleated Hodgkin
and multinucleated Reed-Sternberg cells express-
ing CD 30 antigen. The origin of these malignant
cells still remains unknown; some theories point
to either a fusion of B and non-B cells or mutation
in germinal center cells and loss of B-cell receptor
expression (Rengstl et  al. 2013). Epstein-Barr
virus (EBV) can be observed in about 20 to 40%
of cases, but in contrast to epidemic Burkitt lym-
phoma, its oncogenic role is still disputed (Murray
and Young 2019).
51 Lymphomas 669

51.6.4 Pathology results (Dörffel et al. 2003, 2013). Most current

multicenter trials include central reviewing of all
According to the WHO pathology classification, imaging studies by the study center in order to
two major subtypes of HL are differentiated give a tailored risk-adapted treatment plan, with
(Harris et  al. 1994; Jaffe et  al. 2001): nodular precisely prescribed radiation fields and dosages
lymphocyte predominant Hodgkin lymphoma in conjunction with the stage-adapted chemother-
(NLP HL) and classical Hodgkin lymphoma apy. This has allowed not only for a more homog-
(cHL) subdivided into four histologic categories: enous treatment stratification as well as
comparison between the groups but also for fur-
Lymphocyte-rich ther improvement of the already-excellent treat-
Nodular sclerosis ment results in all treatment groups possibly also
Mixed cellularity decreasing the acute and late side effects of treat-
Lymphocyte depleted ment (Dörffel et  al. 2013; Appel et  al. 2012;
Castellino et al. 2011).
In the cHL subtypes, multinucleated Reed-­ In patients with relapsed or refractory disease,
Sternberg cells and their mononucleated variant immunotherapy, e.g., anti-CD30 monoclonal
Hodgkin cells are characteristic findings. Only antibody, is a promising new treatment option for
0.1–10% of the total cell population in HL are Hodgkin lymphoma, and anti-programmed
malignant cells, whereas the majority of the death-1 (PD1) agents have also produced striking
tumor consists of inflammatory cells (histiocytes, results (Siddiqi et  al. 2014; Michot et  al. 2017;
plasma cells, eosinophils, lymphocytes, neutro- Song et al. 2019).
phils, etc.) and fibrosis. This peculiar histologic
pattern has been attributed to the secretion of dif-
ferent cytokines by the tumor. 51.7 Management

The role of surgery in the treatment of children

51.6.5 Treatment and Results with lymphomas (Hodgkin lymphoma (HL) and
non-Hodgkin lymphoma (NHL)) is limited to
Radiotherapy still is a part of current treatment diagnostic biopsies and treatment complications.
protocols; however, due to improvements in non-­ There are no indications for performing major
cross-­resistant multiagent chemotherapy, radio- tumor resections or debulking procedures, since
therapy can now be reduced in dose and field or chemotherapy is extremely effective and major
even omitted in good response and lower stages surgery delay may complicate the immediate
(Ozuah et  al. 2018; Totadri et  al. 2018). onset of chemotherapy. Only in children with
Chemotherapy mostly consists of derivatives ileocecal intussusceptions due to Burkitt lym-
from the MOPP (Mustargen, Oncovin, procarba- phoma, complete resection of the involved bowel
zine, prednisone) combination designed by segment is advised. Second-look surgery is also
Devita et  al. (1970) and ABVD (Adriamycin, generally not recommended. Particularly, chil-
bleomycin, vinblastine, dacarbacine) regimen dren with NHL may have rapidly growing tumors
developed by Bonadonna et  al. (1975) given in that can cause life-threatening complications
alternating cycles according to the stage of dis- requiring prompt intervention and treatment.
ease. In some patients with favorable clinical pre- Thus, a rapid diagnosis with the least invasive
sentation in the early stage, it is now investigated procedure should be done. In the case of sus-
if the administration of risk-adapted multiagent pected lymphoma, other options to establish the
chemotherapy alone can completely avoid diagnosis before surgery should be considered
involved-field radiation with the intention of like examination of blood and bone marrow and
sparing their late effects without impairing the in the case of pleural effusion/ascites puncture
670
with cytologic and immunophenotypic examina-
tion (Mann et  al. 2006). If the diagnosis with
these simple measures cannot be established, the
most peripheral suspected lesion should be biop-
sied, e.g., in the case of mediastinal tumor, the
nearest extrathoracic lymph node. The optimal
way is that the surgeon, the pediatric oncologist,
and the pathologist cooperate in planning the
biopsy, so that the biopsy material can be taken
over by the pathologist already in the operation
room for further appropriate processing.
51.8 Conclusions and Future
Directions
Future challenge in the management of lympho-
mas in children and adolescents will focus on fur-
ther optimization of frontline therapy within the
multicenter studies. Novel agents, e.g., the so-­
called biologicals, might be beneficial to further
improve treatment outcome and to avoid relapses, Clavel J, Steliarova-Foucher E, Berger C et  al (2006)
since prognosis of children with relapsed disease
is still very poor (Attabaschi et  al. 2005;
Akbayram et al. 2010). Another challenge will be
the reduction of acute and long-term toxicity,
since many long-term survivors from childhood
lymphomas, namely, HD, severely suffer from Dörffel W, Lüders H, Rühl E et  al (2003) Preliminary
long-term side effects including cardiac diseases,
hormone deficits, and even second malignancies
(Lackner et  al. 2007; Galper et  al. 2011; Omer
et al. 2012). Omitting or reducing doses of radio- Dörffel W, Rühl U, Lüders H et al (2013) Treatment of chil-
therapy in selected cases and less intense alkyl-
ator chemotherapy might be associated with
lower rates of long-term side effects (Totadri
et al. 2018).
References
Akbayram S, Dogan M, Akgün C et  al (2010) Use of
rituximab in three children with relapsed/refractory Greenough A, Dave SS (2014) New clues to the molecular
Burkitt lymphoma. Target Oncol 5(4):291–294
Appel BE, Chen L, Buxton A et  al (2012) Impact of
low-dose involved-field radiation therapy on pediat-
ric patients with lymphocyte-predominant Hodgkin Harris NL, Jaffe ES, Stein H et  al (1994) A revised
lymphoma treated with chemotherapy: a report from
the Children’s Oncology Group. Pediatr Blood Cancer
59(7):1284–1289
C. Urban
Attabaschi A, Dworzak M, Steiner M et al (2005) Outcome
of children with primary resistant or relapsed non-­
Hodgkin lymphoma and mature B-cell leukemia after
intensive first line treatment: a population-based anal-
ysis of the Austrian Cooperative Study Group. Pediatr
Blood Cancer 44:70–76
Balcerek M, Schilling R, Byrne J et al (2020) PanCareLIFE
determinants of utilization of cryopreservation of germ
cells in adolescent cancer patients in four European
countries. Eur J Pediatr 179(1):51–60
Bonadonna G, Zucali R, Monfardini S et  al (1975)
Combination chemotherapy of Hodgkin’s disease with
adriamycin, bleomycin, vinblastine, and imidazole
carboxamide versus MOPP. Cancer 36:252–259
Carbone PP, Kaplan HS, Musshoff K et al (1971) Report
of the committee on Hodgkin’s disease staging clas-
sification. Cancer Res 31:1860–1861
Castellino SM, Geiger AM, Mertens AC et  al (2011)
Morbidity and mortality in long-term survivors of
Hodgkin lymphoma: a report from the childhood can-
cer survivor study. Blood 117(6):1806–1816
Chu Y, Gardenswartz A, Termuhlen AM et  al (2019)
Advances in cellular and humoral immunotherapy  -
implications for the treatment of poor risk childhood,
adolescent, and young adult B-cell non-Hodgkin lym-
phoma. Br J Haematol 185(6):1055–1070
Hodgkin's disease incidence and survival in European
children and adolescents (1978-1997): report from the
automated cancer information system project. Eur J
Cancer 42(13):2037–2049
Devita VT Jr, Serpick AA, Carbone PP (1970) Combination
chemotherapy in the treatment of advanced Hodgkin’s
disease. Ann Intern Med 73:881–895
results of the multicenter trial GPOH-HD 95 for the
treatment of Hodgkin’s disease in children and ado-
lescents: analysis and outlook. Klin Pädiatr 215:
139–145
dren and adolescents with Hodgkin lymphoma with-
out radiotherapy for patients in complete remission
after chemotherapy: final results of the multinational
trial GPOH-HD95. J Clin Oncol 31(12):1562–1568
Epstein A (2019) Why and how Epstein-Barr virus was
discovered 50 years ago. Curr Top Microbiol Immunol
390:3–15
Galper SL, Yu JB, Mauch PM et  al (2011) Clinically
significant cardiac disease in patients with Hodgkin
lymphoma treated with mediastinal irradiation. Blood
117(2):412–418
pathogenesis of Burkitt lymphoma revealed through
next-generation sequencing. Curr Opin Hematol
21(4):326–332
European-American classification of lymphoid neo-
plasms: a proposal from the International Lymphoma
Study Group. Blood 84:1361–1392
51 Lymphomas
Izarzugaza MI, Steliarova-Foucher E, Martos MC et  al
(2006) Non-Hodgkin's lymphoma incidence and sur-
vival in European children and adolescents (1978-1997):
report from the automated childhood cancer informa-
tion system project. Eur J Cancer 42(13):2050–2063
Jaffe ES, Harris NL, Stein H, Vardiman JW (2001)
Pathology and genetics of tumours of haematopoietic
and lymphoid tissues. IARC Press, Lyon
Jerusalem G, Beguin Y, Fassotte MF et  al (1999)
Whole-body positron emission tomography using
18F-fluorodeoxy-glucose for posttreatment evaluation
in Hodgkin’s disease and non-Hodgkin’s lymphoma has
higher diagnostic and prognostic value than classical
computed tomography scan imaging. Blood 94:429–433
Johannsdottir HH, Fosså SD et al (2017) Adverse Health
Outcomes and Associations with Self-Reported
General Health in Childhood Lymphoma Survivors. J
Adolesc Young Adult Oncol 6(3):470–476
Kluge R, Kurch L, Montravers F et al (2013) FDG PET/
CT in children and adolescents with lymphoma.
Pediatr Radiol 43:406–417
Kurch L, Hasenclever D, Kluge R et  al (2019) Only
strongly enhanced residual FDG uptake in early
response PET (Deauville 5 or qPET ≥ 2) is prog-
nostic in pediatric Hodgkin lymphoma: results of
the GPOH-HD2002 trial. Pediatr Blood Cancer
66(3):e27539–e27539
Lackner H, Moser A, Sovinz P et  al (2007) Long-term
care of young adults surviving childhood cancer  –
where do we go? Wien Klin Wochenschr 119:361–364
Landmann E, Burkhardt B, Zimmermann M et al (2017)
Results and conclusions of the European inter-
group EURO-LB02 trial in children and adoles-
cents with lymphoblastic lymphoma. Haematologica
102(12):2086–2096
Magrath IT (1981) Lymphocyte differentiation: an essen-
tial basis for the comprehension of lymphoid neopla-
sia. J Natl Cancer Inst 67:501–514
Mann G, Attarbaschi A, Steiner M et al (2006) Early and
reliable diagnosis of non-Hodgkin lymphoma in child-
hood and adolescence: contribution of cytomorphol-
ogy and flow cytometric immunophenotyping. Pediat
Hematol Oncol 23:167–176
Michot JM, Lazarovici J, Ghez D et al (2017) Challenges
and perspectives in the immunotherapy of Hodgkin
lymphoma. Eur J Cancer 85:67–77
Morales-Sánchez A, Fuentes-Pananá EM (2014) Human
viruses and cancer. Viruses 6(10):4047–4079
Mossé YP, Voss SD, Lim MS et al (2017) Targeting ALK
with Crizotinib in Pediatric anaplastic large cell lym-
phoma and inflammatory Myofibroblastic tumor:
a Children's Oncology Group Study. J Clin Oncol
35(28):3215–3221
Murphy SB (1980) Classification, staging and end results
of treatment of childhood non-Hodgkin’s lympho-
mas: dissimilarities from lymphomas in adults. Semin
Oncol 7:332–339
Murray PG, Young LS (2019) An etiological role for the
Epstein-Barr virus in the pathogenesis of classical
Hodgkin lymphoma. Blood 134(7):591–596
671
Neth O, Seidemann K, Jansen P et  al (2000) Precursor
B-cell lymphoblastic lymphoma in childhood and
adolescence: clinical features, treatment and results in
trials NHL 86 and 90. Med Pediatr Oncol 35:20–27
Omer B, Kadan-Lottick NS, Roberts KB et  al (2012)
Patterns of subsequent malignancies after Hodgkin
lymphoma in children and adults. Br J Haematol
158(5):615–625
Ozuah NW, Marcus KJ, LaCasce AS et al (2018) Excellent
outcomes following response-based omission of radio-
therapy in children and adolescents with intermediate
or high-risk Hodgkin lymphoma. J Pediatr Hematol
Oncol 40(6):e338–e342
Patte C, Auperin A, Michon J et  al (2001) The Societe
Francaise d’Oncologie Pediatrique LMB 89 proto-
col: highly effective multiagent chemotherapy tai-
lored to the tumor burden and initial response in 561
unselected children with B-cell lymphomas and L3
leukemia. Blood 97:3370–3379
Reiter A, Schrappe M, Tiemann M et al (1999) Improved
treatment results in childhood B-cell neoplasm with
tailored intensification of therapy: a report of the
Berlin-Frankfurt-Münster group trial NHL-BFM 90.
Blood 94:3294–3306
Reiter A, Schrappe M, Ludwig W-D et al (2000) Intensive
ALL-type therapy without local radiotherapy provides
a 90% event-free survival for children with T-cell lym-
phoblastic lymphoma: a BFM group report. Blood
95:416–421
Rengstl B, Newrzela S, Heinrich T et  al (2013)
Incomplete cytokinesis and refusion of small mono-
nucleated Hodgkin cells lead to giant multinucle-
ated reed-Sternberg cells. Proc Natl Acad Sci U S A
110(51):20729–20734
Seidemann K, Tiemann M, Schrappe M et  al (2001)
Short-pulse B-non-Hodgkin lymphoma-type che-
motherapy is efficacious treatment for pediatric
anaplastic large cell lymphoma: a report of the Berlin-­
Frankfurt-­Münster group trial NHL-BFM 90. Blood
97:3699–3706
Seidemann K, Tiemann M, Lauterbach I et  al (2003)
Primary mediastinal large B- cell lymphoma with
sclerosis in pediatric and adolescent patients: treat-
ment and results from three therapeutic studies of
the Berlin-Frankfurt-Münster Group. J Clin Oncol
21:1782–1789
Seshadri T, Gook D, Lade S et al (2006) Lack of evidence
of disease contamination in ovarian tissue harvested
for cryopreservation from patients with Hodgkin lym-
phoma and analysis of factors predictive of oocyte
yield. Br J Cancer 94(7):1007–1010
Siddiqi T, Thomas SH, Chen R (2014) Role of brentux-
imab vedotin in the treatment of relapsed or refrac-
tory Hodgkin lymphoma. Pharmgenomics Pers Med
7:79–85
Song Y, Gao Q, Zhang H et  al (2019) Treatment of
relapsed or refractory classical Hodgkin lymphoma
with the anti-PD-1, tislelizumab: results of a phase 2,
single-arm, multicenter study. Leukemia 34(2):533–
542. https://doi.org/10.1038/s41375-­019-­0545-­2
672
Sperl D, Benesch M, Urban C et al (2012) Fatal EBV infec-
tion and variable clinical manifestations in an XLP-1
pedigree  - rapid diagnosis of primary immunodefi-
ciencies may save lives. Klin Padiatr 224(6):386–389
Totadri S, Radhakrishnan V, Ganesan TS et al (2018) Can
radiotherapy be omitted in children with Hodgkin lym-
phoma who achieve metabolic remission on interim
positron emission tomography? Experience of a ter-
tiary care cancer referral center. J Glob Oncol 4:1–7
van der Kaaij MA, Heutte N, van Echten-Arends J et al
(2009) Sperm quality before treatment in patients with
early stage Hodgkin's lymphoma enrolled in EORTC-­
C. Urban
GELA lymphoma group trials. Haematologica
94(12):1691–1697
Van der Kaaij MAE, Van Echten-Arends J, Simons AHM
et al (2010) Fertility preservation after chemotherapy
for Hodgkin lymphoma. Hematol Oncol 28:168–179
Van Gijn J, Gijselhart JP (2012) Thomas Hodgkin and his
disease. Ned Tijdschr Geneeskd 156(15):A4332
Wieland A, Kerbl R, Berghold A et al (2003) C-reactive
protein (CRP) as tumor marker in pediatric and ado-
lescent patients with Hodgkin disease. Med Pediatr
Oncol 41:21–25
 Wilms’ Tumor
Michael E. Höllwarth
52.1 Introduction
Wilms’ tumor (WT) is the most common kidney
tumor in children up to 15 years of age. The prev-
alence is about 1 in 10,000 in children younger
than 15  years of age (Charlton et  al. 2017a, b).
Regarding the ASR/million (age-standardized
rate), WT is highest in children aged 0–4 y (15.1)
compared to children 5–9 y (4.2) and 10–14 y
(0.7) and is higher in high-income countries
(ASR 8.6) (Cunningham et al. 2020; Gooskens et
al. 2016). Males and females are affected equally.
92–95% of the WT occur in only one kidney and
are unilateral. Bilateral tumors occur in 5–8%.
They present at an earlier age and are either syn-
chronous or metachronous, and nearly all cases
are associated with the presence of precursor
lesions, so-called nephrogenic rests (Charlton
et al. 2017a, b). Congenital anomalies associated
with WT are common. About 2% of patients have
a positive family history, with one relative who
also suffered from a WT.  The association with
other anomalies, as well as the occurrence in
families, suggests that altered genes are involved
in the tumor pathogenesis. Several genetic loci
are involved. The two most important genes are
M. E. Höllwarth (*)
University Clinic of Paediatric and Adolescent
Surgery, Medical University of Graz, Graz, Austria
e-mail: michael.hoellwarth@medunigraz.at
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_52
52
the WT 1 and WT 2 suppressor genes on chromo-
some 11, but additional loci have been described
on several other chromosomes.
52.2 Historical Overview
The first description of a kidney tumor as myoma
sarcomatodes was by Eberth in 1872. In 1898,
Birch-Hirschfeld classified a similar kidney
tumor as embryonal adenosarcoma. In 1899,
Carl Max Wilhelm Wilms, a German pathologist
and surgeon, published a monography “Die
Mischgeschwülste der Niere,” in which he
reviewed the literature and added seven more
cases of the clinical picture of the tumor that
now bears his name, or nephroblastoma (Wilms
1899). One hundred years ago, surgery was the
only cure, but the survival rate then was under
10%. Progress in anesthesia and surgical tech-
niques, together with the introduction of local
postoperative radiation in the 1930s, was fol-
lowed by an increase of the 3-year survival rate
of around 30%. New adjuvant therapeutic strate-
gies including chemotherapy in the 1950s (acti-
nomycin D) and 1960s (vincristine) were
followed by a constant increase in 5-year sur-
vival rates. Today, international Wilms tumor
study groups (SIOP in Europe and NWTS (now
COG) in North America) achieve overall sur-
vival rates of up to 90% (Green 2013; Nakayama
and Bonasso 2016).
673
674 M. E. Höllwarth

52.3 Etiopathogenesis The cross section of the nephrectomy speci-

men shows partly solid, partly cystic regions
Most WT occur sporadically in one kidney between with hemorrhagic areas (Fig.  52.1). If the WT
0 and 4  years, mostly around 3  years old. While occupies only one pole, there exists a clear
most WTs are unilateral, bilateral WT occur syn- demarcation line to the normal kidney. The ori-
chronous in 6.3% and metachronous in 0.85% of all gin of the tumor is the metanephrogenic blas-
WTs and present at a younger age (2.6 vs. 3.3 y) tema. While normal nephrons are perfectly
(Charlton et al. 2017a, b; Millar et al. 2017). Patientsorganized, the tissue in WT is disorganized, and
with syndromic ­malformations are often associated the histology mimics the development of a nor-
with presentation of WT, such as the WAGR syn- mal kidney, showing the three tissue compo-
drome (aniridia, genitourinary malformations, and nents: blastema, epithelial cells, and stroma
mental retardation), Denys-Drash syndrome (Fig.  52.2). The proportions of these cell ele-
(nephropathy with renal failure, male pseudoher- ments may be different from one tumor to the
maphroditism), Beckwith-Wiedemann syndrome other. However, each component can exhibit
(omphalocele, hemi-hypertrophy, macroglossia, focal or diffuse anaplastic cell elements, which
hyperinsulinism at birth), Perlman syndrome (dys- are indicators for a poor outcome (Fig.  52.3).
morphic face, cryptorchidism, visceromegaly,
hyperinsulinism at birth), Frasier syndrome (male
pseudohermaphroditism, nephropathy), and others.
In the complex development of the kidneys, a
large number of transcription factors, proto-­
oncogenes, and various types of growth factors
are involved (Diniz 2016). An analysis of WT
nephrectomy specimens showed a high expres-
sion of renal progenitor genes, paternally derived
genes that characterize the early metanephric
mesenchyme and genes that maintain self-­
renewal capacity (Treger et  al. 2019). Recent
research has shown that embryonal precursors
can be detected from which unilateral and multi-
focal WT develop (Coorens et al. 2019). Over the
last few decades, many different biomarkers and Fig. 52.1  Typical appearance of the tumor consisting of
solid and cystic parts as well as hemorrhagic regions
genetic alterations in WT have been published,
but it is beyond the scope of this chapter to dis-
cuss them. The interested reader may read up on
the recent literature (Deng et  al. 2016; Treger
et  al. 2019; Cresswell et  al. 2016; Hohenstein
et  al. 2015; Mahamdallie et  al. 2019; Charlton
et al. 2017a, b).

52.4 Pathology

The tumor compromises different regions, if not

most parts of the kidney, and may also invade the
urinary collecting system, the renal vein, or the Fig. 52.2  Medium-power microscopic view of a typical
surrounding tissues. WT with elements of tubules, blastema, and spindle cells
52  Wilms’ Tumor
Fig. 52.3  High-power microscopic view of an anaplastic part of a WT with polymorphic cell nuclei and atypical cell
mitosis
Fig. 52.4  Medium-power microscopic view of perilobar
nephrogenic rest consisting of mesonephric blastema cells
within normal kidney tissue. Typically, NR have no
pseudocapsule
Fig. 52.5  Diffuse bilateral microscopic nephroblastoma-
tosis. This kidney responded well to chemotherapy, but
the patient developed a unilateral WT 2 years later, which
was resected using nephron-sparing surgery
675
Blastema cells are thought to be the most malig-
nant component (Popow et al. 2016).
Nephrogenic rests (NR) consist of foci of per-
sisting blastemic cells, which are situated either
intralobar or perilobar (Fig. 52.4). They are seen
as precursor cells of WT and can be found in 40%
of WT patients (Vujanic et al. 2017). NR can be
found beyond 36 weeks of gestation, within nor-
mal parts of the kidney (Breslow et al. 2006). NR
are common in neonatal kidneys and are most
often transformed into normal kidney tissue or,
rarely, into WT cell elements. In contrast to WT,
NR are not surrounded by a pseudocapsule.
Rarely, they may be found outside the kidney in
different locations (Ma et al. 2018).
Nephroblastomatosis is a pathology when
both kidneys consist of multifocal of diffuse peri-
lobar nephrogenic rests (Fig.  52.5). It is impor-
tant to know that the management of NR or
nephroblastomatosis is different from that of a
WT (see later). However, nearly all cases with
bilateral WT are associated with NRs.
52.5 Diagnosis
Most children present with an asymptomatic, but
rapidly growing, abdominal mass. About 30% of
patients suffer from abdominal pain, fever, mal-
aise, and weight loss. Some patients may addi-
tionally show a left-sided varicocele (occlusion
of the renal vein by tumor invasion) or a symp-
676
tomatic hydrocele. Ten percent of patients suffer
from hematuria, mostly microscopic, when the
tumor invades the urinary collecting system.
The availability of ultrasound in the private
GP and pediatricians’ practices today allows
early diagnosis of WT and referral to the pediat-
ric center. The ultrasound study documents the
position, size, and volume of the tumor, the intra-
vascular extension, the metastatic involvement of
regional lymph nodes and the liver, and the
examination of the contralateral kidney.
Furthermore, regular ultrasound investigations
during preoperative chemotherapy allow weekly
measurement of tumor volume, thereby indicat-
ing whether the tumor responds immediately to
the therapy or not. The abundant availability of
ultrasound today allows regular screening in
patients with associated anomalies and increased
risk for WT.  Recommended screening intervals
are two times a year.
The abdominal CT scan, with contrast admin-
istration, shows in detail the location of the tumor
in the kidney, whether it extends into the sur-
rounding tissues and the metastatic involvement
of regional or distant lymph nodes. A chest X-ray
is obtained to evaluate for the presence of pulmo-
nary metastases. When the chest X-ray is nega-
tive, an additional CT scan allows for more
precise evaluation of small pulmonary metasta-
ses, which has a significant impact on the tumor
staging. An MRI may provide additional infor-
mation to confirm the diagnosis by imaging
methods alone, thereby excluding different kid-
ney pathologies. Additionally, the development
of diffusion-weighted images (DWI) has been
shown to predict WT histology.
Tumor extension into the renal vein and vena
cava presents a particular problem. The preopera-
tive imaging examinations show the extension of
the invasion accurately.
52.6 Differential Diagnosis
The diagnosis is usually clear and not difficult for
the pathologist if the WT consists of the typical
three or two histological elements and a pseudo-
capsule around it. However, tumors with only
M. E. Höllwarth
one of the three histological features are more
difficult to distinguish from other renal tumors
(see overview by Popow et  al. 2016). Pure
blastema-­type WTs are similar as neuroblastoma,
Ewing sarcoma, and desmoplastic round cell
tumors. Pure epithelial-type WTs may look like
metanephric adenoma, renal cell carcinoma, or
perilobar NRs (Sarlos et al. 2018). In the case of
pure stoma-type WTs, clear cell sarcoma and
mesoblastic nephroma may look similar. In a
more cystic form of WT, differential diagnosis
includes a cystic kidney disease such as cystic
nephroma. Modern immunohistochemistry and
molecular biology investigations allow the differ-
entiation between WTs and these other tumors.
52.7 Management
The mainstay of therapy is the radical surgical
excision of the tumor without tumor spill, sample
lymph nodes, and, if needed, excision of addi-
tional metastases. The international protocols
(SIOP and COG) include detailed recommenda-
tions for staging, grading, surgery, controls, and
chemotherapeutic regimen in WT patients.
52.7.1 Staging
The basis of the therapy is the clinical staging in
regard to the tumor extent and the histological
grading. The former National Wilms’ Tumor
Study Group (NWTS), now Children Oncology
Group (COG), and the International Society for
Paediatric Oncology (SIOP-Societé Internationale
pour Oncolgie Paediatrique) use a slightly differ-
ent staging system (Vujanic et al. 2018).
52.7.1.1 SIOP Staging
Stage I: Tumor is limited to the kidney or sur-
rounded with a fibrous pseudocapsule and is
completely resected. The tumor is not infiltrating
the collecting system or the vessels of the renal
sinus. Percutaneous cutting needle biopsy is
allowed. No tumor rupture.
Stage II: Tumor extends beyond the kidney or
penetrates the renal capsule or pseudocapsule,
52  Wilms’ Tumor 677

but is completely resected; the tumor infiltrates Table 52.1  Advantages and disadvantages of the proto-
cols of SIOP and COG
the vessel in the renal sinus or vena cava or lym-
phatic vessels but is completely resected. •  Downstaging with SIOP
•  Less stage III (less radiation)
Percutaneous cutting needle biopsy is allowed.
•  More stages I and II
Stage III: Incomplete resection, residual non-­
•  More relapses in stage II
hematogenous tumor, involvement of paraaortic •  Complications higher with COG 9.8% vs. 6.8%
or paracaval lymph nodes, tumor rupture with •  Tumor rupture and spillage 15.3% vs. 2.2%
diffuse spillage, peritoneal implants, local •  Stage III—more radiation therapy
­infiltration of vital structures. The tumor has been
surgically biopsied (wedge or open) prior to pre-
operative chemotherapy or surgery. tumor leakage along the needle tract leading to
Stage IV: Distant metastases. recurrences within the biopsy tract.
Stage V: Bilateral tumor; each side gets One of the criticisms in regard to the SIOP
sub-staging. protocol was the uncertainty of diagnosis without
a preoperative biopsy. The SIOP arguments are
52.7.1.2 COG Staging that imaging methods today are so excellent that
Stage I: Tumor is limited to the kidney or sur- misdiagnosis is extremely rare. In SIOP-9, only
rounded with a fibrous pseudocapsule and is 2% of patients received preoperative chemother-
completely resected. The tumor is not infiltrating apy inappropriately for non-malignant processes,
the collecting system or the vessels of the renal and 3% had other malignant diseases. Neither
sinus. The tumor was not biopsied or ruptured actinomycin D nor vincristine is associated with
before surgery. Stage II: Tumor extends beyond much short-term toxicity, and no long-term com-
the kidney or penetrates the renal capsule or plications have been identified so far. Tumor
pseudocapsule, but is completely resected. The biopsy is mandatory if weekly control with ultra-
tumor infiltrates the vessel in the renal sinus. sound do not show an immediate and significant
Stage III: Incomplete resection, residual non-­ shrinkage of the tumor volume in response to the
hematogenous tumor, involvement of paraaortic chemotherapy within 1 or 2  weeks. The disad-
or paracaval lymph nodes, tumor rupture with vantage of the CUG protocol is caused by the fact
diffuse spillage, peritoneal implants, local infil- that the surgical procedure for very large tumors
tration of vital structures. The tumor was rup- is more difficult. Thus, surgical complication
tured or biopsied before removal.Stage IV: rates are higher (9.8% vs. 6.8%) although not sig-
Distant metastases. nificant, but tumor rupture occurs significantly
Stage V: Bilateral tumor; each side gets more often (15.3% vs. 2.2%). As a consequence,
sub-staging. the number of Stage III patients with the addi-
The difference in the two ways of staging is tional need for local radiation is significantly
related to the impact of the diagnostic biopsy on higher (30.4% vs. 14.2) (Table 52.2).
the risk of recurrence and on the impact of intra- However, preoperative chemotherapy in syn-
operative tumor spillage if primary surgery chronous bilateral WT is standard of care in the
instead of chemotherapy is chosen. However, SIOP and COG strategy, for a maximum time of
preoperative chemotherapy may result in a down- 12  weeks before surgery. Extension of chemo-
staging of the tumor (Table  52.1). This effect therapy beyond 12 weeks will not further facili-
after chemotherapy raises doubts in regard to tate the surgical procedure but increases the risk
Stage II, because, in reality, it might have been a of anaplastic transformation (Charlton et  al.
Stage III.  The WT trial in the UK recommends 2017a, b). A recent meta-analysis showed that
preoperative chemotherapy also but includes per- 72.5% of patients with bilateral WT received pre-
cutaneous fine needle biopsy to obtain histologi- operative chemotherapy, while the other patients
cal confirmation. This procedure has a risk of were operated on initially (Han et al. 2018).
678 M. E. Höllwarth

Table 52.2  Management differences between the SIOP 52.7.4 Pharmacotherapy

and COG group protocol
COG SIOP Although several differences exist between the
• Tumor nephrectomy or • Chemotherapy, biopsy protocols of SIOP and COG (Tables 52.1 and
biopsy optional
• Postoperative • Secondary tumor
52.2), outcomes are quite similar with a survival
chemo(radio)therapy nephrectomy rate between 85 and 90%.
Disadvantages Pretreatment: According to the current
• Tumor rupture and SIOP protocol, unilateral cases receive 4-week
spillage pretreatment with vincristine (weekly) and
• More complications •  Diagnostic error
actinomycin D (biweekly). Metastatic patients
•  Extensive resection • Downstaging
•  Increased mortality • Downgrading receive the same therapy for 6  weeks and
include doxorubicin on weeks 1 and 5. Bilateral
tumors receive vincristine and actinomycin D
52.7.2 Grading for a maximum of 9–12  weeks. Children
younger than 6  months are submitted to pri-
In regard to the result of histology, the postopera- mary surgery if the tumor is appropriate to be
tive therapy is stratified. SIOP classifies WTs as resected.
low risk if they are completely necrotic (after pre- Postoperative treatment depends on the stag-
operative chemotherapy); intermediate risk are ing and histological grading (Table 52.3).
epithelial, stromal, or mixed subtypes (including Further strategies aim to reduce chemothera-
focal anaplasia); and high risk are blastema types peutic toxicity, if possible, in very-low-risk cases.
and diffuse anaplasia. COG classifies all non-­ These are defined by COG in patients under
anaplastic histology types as favorable and focal 2  years of age as Stage I, favorable histology,
or diffuse anaplastic types as unfavorable. The tumor weight under 550 g. With the COG strat-
COG strategy included molecular markers into the egy, these children do not get chemotherapy after
treatment strategy. The most recent SIOP protocol nephrectomy. In contrast, in high-risk patients and
(Vujanic et al. 2018) includes updated guidelines recurrences, chemotherapy is intensified, e.g.,
for the pathologists as well as criteria for the post- MYCN is known to be associated with poor out-
operative histological classification/grading and come in many cancers in childhood, including
validates new prognostic factors, such as blastema WT (Vujanic et al. 2018). However, only 13% of
tumor volume and molecular markers, and more patients with WT have a MYCN expression
details regarding the staging for Stages II and III (Treger et al. 2019). One of the new drugs used in
(Vujanic et al. 2018). this regard is irinotecan for metastatic and relapsed
WT in patients (Oostveen and Prichard-­ Jones
2019).
52.7.3 Radiotherapy
Table 52.3  Staging and histological grading
Radiotherapy of the tumor region is recom-
mended for Stage III; therefore, tumor spillage Risk Stage I Stage II Stage III
during surgery should be avoided at all costs. Low No treatment A,V A,V
Intermediate A,V A,V,D RT,A,V,D
Lung radiation can be omitted if pulmonary R R
metastases disappear under chemotherapy or A,V RT,A,V
when they have been resected successfully. High A,V,D All + RT All + RT
However, pulmonary radiation is inevitable if R randomization, A actinomycin D, V vincristine, RT
additional pulmonary metastases reappear. radiotherapy, D doxorubicin
52  Wilms’ Tumor
52.7.5 Surgical Therapy
52.7.5.1 General Aspects
The benefit of primary chemotherapy consists of a
significant shrinkage of the tumor, evidenced by
weekly ultrasound controls, and, as a result, a
reduced incidence of intraoperative rupture
(Table  52.2) (Godzinski 2015; Godzinski et  al.
2014). The mainstay of therapy is the radical sur-
gical excision of the tumor without tumor spill.
The surgical p­ rocedure consists of a large trans-
verse incision to facilitate tumor nephrectomy.
Recently, patients’ selection and technical aspects
of a laparoscopic approach have been discussed
(Schmidt et al. 2019). Whether or not the contra-
lateral kidney must be explored during surgery is a
matter of debate. We think that today’s preopera-
tive imaging methods are so good that this exten-
sion of the procedure is rarely necessary. The
hilum of the kidney is approached first and the ves-
sels are ligated in order to avoid intraoperative
spread of malignant cell elements through the
renal vein. However, this part of the procedure is
often not feasible due to the huge size of the tumor
and a close connection to the vena cava or aorta.
The adrenal gland can be left in place if it is not
abutting the tumor. The ureter is ligated and
divided as low as possible. Careful sampling of
regional and distant lymph nodes is essential for
correct staging. At least 7–10 lymph nodes need to
be sampled to maintain a false negative rate < 10%
(Saltzman et  al. 2019). Metastatic lesions in the
liver or lungs are usually excised later if they are
still present after appropriate chemotherapy.
Preoperative imaging studies (CT, MRI, echo-
cardiography) show whether or not tumor throm-
bus invaded major renal vessels and/or the vena
cava including up to the right atrium. Preoperative
chemotherapy may reduce the tumor thrombus
size and retract it, even from the heart. If the
tumor thrombus is finally below the hepatic
veins, it might be resected by cavatomy. If the
extension is higher, or even up to the right atrium,
a sternotomy with cardiopulmonary bypass is
needed (Cox et al. 2018).
679
The incidence of a WT in horseshoe kidneys
is approximately 0.5%. We recommend preop-
erative chemotherapy to reduce the tumor size
and to allow an easier resection of the involved
side.
Early surgical complications can be local
hemorrhage, wound infection, and vascular
injury. The most problematic surgical complica-
tion is intraoperative tumor rupture and spillage
since it immediately increases the stage to III
with the need of postoperative abdominal radia-
tion. Intestinal obstruction is a rare, late postop-
erative surgical complication.
52.7.5.2 Bilateral WT
The main surgical goal in bilateral WT is to pre-
serve as much as possible of the functioning
kidney parenchyma, using nephron-sparing sur-
gery (NSS). The surgical procedure should
always start at the more difficult side; if a kid-
ney-sparing surgical procedure is not possible
and tumor nephrectomy is inevitable, the sur-
geon should do its best to save as much normal
kidney tissue as possible on the other side, even
by tumor enucleation if no anaplastic histology
is present (Fig. 52.6) Technical innovation, such
as in situ topical cooling, allows an adequate
cross clamp time and avoids acute tubular necro-
sis in most cases. In rare circumstances, the
involved kidney is completely removed to the
bench after being flushed with ice-cold preser-
vation solution. When the tumor is resected,
replantation can often be performed in an ortho-
topic position (Millar et al. 2017). According to
Charlton et  al. (Charlton et  al. 2017a, b), NSS
was performed in 344/517 (66%) patients, com-
bining radical nephrectomy on the one side and
NSS on the other side (192), bilateral NSS
(127), unilateral NSS and biopsy on the other
side (11), and unilateral NSS alone (14). In a
few patients, bilateral nephrectomy cannot be
avoided, and renal transplantation can be per-
formed after 1 or 2  years without recurrent
malignancy.
680 M. E. Höllwarth

52.7.5.3 Nephron-Sparing Surgery

Fig. 52.6  Bilateral WT.  In this case, tumor enucleation
on the right side is not a violation of the protocol
(NSS)
Nephron-sparing surgery is a well-established
method in adult patients with unilateral renal cell
carcinoma. In bilateral WTs, NSS is often needed;
however, renal salvage procedures in unilateral
WT remain a controversial issue, and tumor
nephrectomy is recommended as the standard sur-
gical procedure. Nephron-sparing surgery in the
SIOP protocol becomes possible due to the sig-
nificant reduction of the tumor size after chemo-
therapy, when a primary huge tumor is occupying
only one kidney pole (Figs. 52.7a, b and 52.8).
Any reduction of kidney tissue causes struc-
tural and functional hypertrophy of the remaining
nephrons. Hypertrophy of the tubules and glom-
eruli is accompanied by an increase of g­ lomerular
filtration rate. Both are precursors to the develop-
ment of hypertension (Fong et  al. 2014). In a
follow-up study of 72- and 65-month-old chil-
dren with nephrectomy in unilateral WT, they

a b

Fig. 52.7 (a) Huge WT on the right side before chemotherapy (figure on the left). (b) Significant reduction of tumor
size and volume after 4 weeks of chemotherapy (figure on the right)
52  Wilms’ Tumor
Fig. 52.8  The tumor is situated at one pole and perfectly
separated from the normal kidney. Thus, surgeon and
oncologist decided together for a kidney-sparing surgical
procedure
had significantly higher serum creatinine and sys-
tolic and diastolic pressure compared to children
with NSS (Cozzi et al. 2005). Among 75 children
19.6 years after unilateral nephrectomy, 21.3% had
an eGFR <90  ml/min/1.73  m2, and two of them
had proteinuria; five had hypertension (>
140/90 mm Hg) (Interiano et al. 2015). A Canadian
study in adults with renal cell carcinoma (665 par-
tial and 714 RN) showed a decrease of the esti-
mated glomerular filtration rate (eGFR) at 3, 12,
and 24  months after radical nephrectomy. Severe
renal failure developed postoperatively in 12.5%
after RN vs. 4.1% after partial nephrectomy
(Mason et  al. 2016). NSS is recommended in
patients with Beckwith-Wiedemann syndrome and
isolated hemihypertrophy with bilateral WT and
may be an option in selected cases with unilateral
WT (Scalabre et al. 2016). A recent meta-analysis
recommeds the use of nephron sparing surgery for
unilateral Wilms tumor (Chen et al. 2020).
It is important to observe clear selection crite-
ria for a NSS procedure in unilateral WT:
• Tumor occupies only one pole of the kidney
(Fig. 52.8).
• At least 50% of normal kidney tissue can be
preserved.
681
• The tumor does not invade the urinary collect-
ing system or the main renal vessels.
• At least a good rim of normal tissue can be
resected with the tumor.
• The regional and distant lymph nodes must be
free of tumor.
NSS should be performed in most of the bilat-
eral WTs, if possible (Kieran et  al. 2013). The
SIOP study showed that it is performed in eligi-
ble patients with unilateral tumors of Stages I and
II in around 3% of cases. The surgical procedures
are encoded as type A, when the tumor is resected
with a rim of normal kidney tissue, and type B,
which means enucleation without a rim of nor-
mal kidney tissue (see Fig. 52.6 on the right side).
NSS is also possible in Stage IV, provided the
regional lymph nodes are free of tumor (Linni
et al. 2003).
The surgical strategy is identical in all WT
patients using a large transverse laparotomy.
The kidney is mobilized, and the tumor is iden-
tified. It should not be adherent to any other
structures, and there should be no contact with
the hilum of the kidney. We used to decide as a
team—the surgeon and oncologist together
during surgery—whether or not a NSS is per-
formed. The main renal vessels are closed at
the hilum with a vascular clamp, and the tumor
is resected, leaving the pseudocapsule intact
and with a safe rim of normal kidney of tumor
using an ultrasound knife (Figs. 52.8 and 52.9).
If it is unclear whether resection can be per-
formed safely through normal renal tissue, an
intraoperative ultrasound is used to clarify the
situs. The warm ischemic time is no longer
than 10 to 15  minutes and has no negative
effects on the kidney function. The open col-
lecting system is closed; to avoid prolonged
urinary leak and bleeding, vessels are coagu-
lated carefully. Postoperative studies of renal
function show that 75% of the total renal tissue
can be preserved by NSS (Figs.  52.10 and
52.11).
682
Fig. 52.9  The remaining healthy kidney after tumor
resection
Fig. 52.11  The isotope scintigraphy shows a perfect urinary function on the right side (red curve). Three quarters of
the total kidney function is preserved
52.7.6 Nephroblastomatosis
Macroscopic nephroblastomatosis is characterized
by significantly enlarged and lobulated kidneys
(Fig. 52.5). It usually occurs in children younger
than 2 years old. The therapeutic regimen includes
M. E. Höllwarth
Fig. 52.10  Intravenous urography shows good excretion
on the right side
bilateral biopsy and reduced chemotherapy until
the kidneys achieve the normal appearance (Ortiz
et al. 2019). However, long-­term regular controls
are mandatory, because a unilateral or bilateral
synchronous or metachronous Wilms’ tumor may
occur months or even years later.
52  Wilms’ Tumor
52.7.7 Outcome
The long-term outcome of children with WT
depends on the stage and tumor histology. Today,
overall survival (OS) can reach 95%, and event-­
free survival (EFS) is around 90%. In very-low-­
risk tumors, the OS reaches 100% (Fernandez
et al. 2017). In a meta-analysis of bilateral WT,
the OS was 73%; recurrence and renal failure
occurred in 20% and 8% (Han et al. 2018). For
NSS, OS was 100% and EFS 94.8%. In a French
meta-analysis of children with unilateral WT,
3994 (93.1%) after radical nephrectomy (RN) vs.
294 (6.8%) after NSS showed similar overall and
event-free survival rates (95.7% and 92.8% after
RN vs. 96.0% and 90.5% after NSS). The rate of
mild to moderate reduced renal function was
higher after RN (42% vs. 10%) (Tricard et  al.
2017).
Long-term renal function is an important fac-
tor regarding the outcome after bilateral and uni-
lateral WT.  Among 9237 patients enrolled in 5
NWTS protocol studies, the cumulative inci-
dence of renal failure 20 years after WT for non-­
syndromic patients was 1.7%, and for
non-syndromic patients with unilateral tumor of
favorable histology, it was 0.6%, 4.0% in syn-
chronous bilateral WT, and 19.3% in metachro-
nous bilateral WT. The cumulative incidence of
end-stage renal disease is significantly higher in
patients with congenital syndromes such as the
Denys-­Drash syndrome (74%) and WAGR syn-
drome (36%) and males with hypospadias or
cryptorchidism (7%) (Breslow et al. 2005).
52.8 Conclusion
The history of WTs in children is a story of suc-
cess, reaching an OS of more than 90% today.
Children with low-risk tumors (favorable histol-
ogy) have excellent outcomes and rarely relapse.
Sophisticated protocols of COG and SIOP
including biological markers and storage of
tumor samples in biobanks will allow the imple-
683
mentation of new techniques in the future, to
improve the therapeutic stratifications for chil-
dren with anaplastic WT and bilateral diseases as
well as for syndromic cases.
References
Breslow NE et  al (2005) End stage renal disease in
patients with Wilms tumor: results from the National
Wilms Tumor Study Group and the U.S. Renal Data
System. J Urol 174:1972–1975
Breslow NE et  al (2006) Age distribution, birth weight,
nephrogenic rests, and heterogeneity in the pathogen-
esis of WT. Pediatr Blood Cancer 47:260–267
Charlton J et al (2017a) Bilateral Wilms tumor: a review
of clinical and molecular features. Expert Rev Mol
Med 19(e8):1–13
Charlton J et al (2017b) Bilateral Wilm tumour: a review
of clinical and molecular features. Expert Rev Mol
Med 19:e8:1 of 13
Chen H et al. (2020) Effectiveness of nephron sparing
surgery and radical nephrectomy in the management
of unilateral Wlms Tumor: A Meta-analysis. Front
Oncol. https://doi.org/10.3389/fonc2020.01248
Coorens THH et  al (2019) Embryonal precursors of
Wilms tumor. Science 366:1247–1251
Cox SG et al (2018) Surgical management and outcomes
of 12 cases of Wilms tumour with intracardiac exten-
sion from a single center. Pediatr Surg Int 34:227–235
Cozzi F et al (2005) Renal function adaptation in children
with unilateral renal tumor treated with nephron spar-
ing surgery or nephrectomy. J Urol 174:1404–1408
Cresswell GD et al (2016) Intra-tumor genetic heteroge-
neity in Wilms tumor: clonal evolution and clinical
implications. EBioMedicine 9:120–129
Cunningham ME et al (2020) Global disparities in Wilms
tumor. J Science Res 247:34–51
Deng C et  al (2016) Genetic variation frequencies in
Wilms’ tumor: meta-analysis and systematic review.
Cancer Sci 107:690–699
Fernandez CV et  al (2017) Clinical outcome and bio-
logical predictors of relapse after nephrectomy for
very low-risk Wilms tumor. A report from Children’s
oncology group AREN03532. Ann Surg 265:835–840
Fong D et  al (2014) Compensatory responses to neph-
ron deficiency: adaptive or maladaptive? Nephrology
19:119–128
Godzinski J (2015) The current status of treatment of
Wilms’ tumor as per the SIOP trials. J Indian Assoc
Pediatr Surg 20:16–20
Godzinski J et al (2014) Current concepts in surgery for
Wilms tumor – the risk and function-adapted strategy.
Eur J Pediatr Surg 24:457–460
684
Gooskens SL et al (2016) The clinical relevance of age at
presentation in nephroblastoma. In: van den Heuvel-­
Eibrink MM (ed) Wilms tumor. Cordon Publications,
Brisbane
Green DM (2013) Evaluation of renal function after suc-
cessful treatment for unlateral, non-syndromic Wilms
tumor. Pediatric Blood Cancer 60:1929–1935
Han Q et al (2018) Clinical features, treatment, and out-
comes of bilateral Wilms’ tumor: a systematic review
and meta-analysis. J Pediatr Surg 53:2465–2469
Hohenstein P et  al (2015) The yin and yang of kid-
ney development and Wilms’ tumors. Genes Dev
29:467–482
Interiano RB et  al (2015) Renal function in survi-
vors of nonsyndromic Wilm tumor treated with
unilateral radical nephrectomy. Cancer 121:
2449–2456
Kieran K et al (2013) Margin status and tumor recurrence
after nephron-sparing surgery for bilateral Wilms
tumor. J Pediatr Surg 48:1481–1485
Linni K et  al (2003) Nephron sparing procedures in
11 patients with Wilms’ tumor. Pediatr Surg Int
19:457–462
Ma Y et al (2018) Ectopic nephrogenic rests in children: a
series of 13 cases in a single institution. Pediatr Blood
Cancer 65:e26985
Mahamdallie S et al (2019) Identification of new Wilms
tumor predisposition genes: an exome sequencing
study. Lancet Child Adolesc Health 3:322–331
Mason R et al (2016) The natural history of renal function
after surgical management of renal cell carcinoma:
results from the Canadian kidney cancer information
system. Urol Oncol 34:486.e1–486e.7
Millar AJW et al (2017) Management of bilateral Wilms
tumours. Pediatr Surg Int 33:737–745
Nakayama DK, Bonasso PC (2016) The history of
multimodal treatment of Wilms’ tumor. Am Surg
82:487–492
M. E. Höllwarth
Oostveen RM, Prichard-Jones K (2019)
Pharmacotherapeutic management of Wilms tumor:
an update. Pediatr Drugs 21:1–13
Ortiz MV et  al (2019) Maintenance chemotherapy to
reduce the risk of a metachronous Wilms tumor in
children with bilateral nephroblastomatosis. Pediatr
Blood Cancer 66(e27500):1–4
Popow SD et  al (2016) Wilms’ tumour  – histology and
differential diagnosis. In: van den Heuvel- EMM (ed)
Wilms tumor. Cordon Publications, Brisbane
Saltzman AF et  al (2019) How many lymph nodes are
enough? Assessing the adequacy of lymph node yield
for staging in favorable histology Wilms tumor. J
Pediatr Surg 54:2331–2335
Sarlos DP et al (2018) Embryonal origin of metanephric
adenoma and its differential diagnosis. Anticancer Res
38:6663–6667
Scalabre A et al (2016) Is nephron sparing surgery justi-
fied in Wilms tumor with Beckwith-Wiedemann syn-
drome or isolated hemihypertrophy? Pediatr Blood
Cancer 63:1571–1577
Schmidt A et  al (2019) Patient selection and technical
aspects for laparoscopic nephrectomy in Wilms tumor.
Surg Oncol 29:14–19
Treger TD et  al (2019) The genetic changes of Wilms
tumour. Nat Rev Nephrol 15:240–251
Tricard T et al (2017) Néphrectomie partielle pour tumeur
de Wilms unilatérale chez l’enfant: Est-ce pertinent?
Arch Pediatr 24:650–658
Vujanic GM et  al (2017) Nephrogenic rests in Wilm
tumors treated with preoperative chemotherapy: the
UK SIOP Wilm tumor 2001 trial experience. Pediatr
Blood Cancer 64(e26547):1–7
Vujanic GM et al (2018) The UMBRELLA SIOP-RTSG
2016 Wilms tumour pathology and molecular biology
protocol. Nat Rev Urol 15:693–701
Wilms M (1899) Die Mischgeschwülste der Niere. Verlag
von Arthur Georg, Leipzig
 Ovarian Tumors
Alicia G. Sykes, Mary E. Fallat,
and Romeo C. Ignacio Jr.
53.1 Introduction
Ovarian tumors comprise a heterogeneous group
of benign and malignant lesions. Their clinical
presentation can vary based on the age of the
patient, size of the tumor, hormonal activity, his-
tology, and local spread or metastasis. Although
ovarian malignancies are the fifth most common
cause of cancer among adult females, they are
considered rare in children and adolescents. A
significant proportion of ovarian tumors are
potentially malignant. Therefore, a thorough
understanding of the characteristics that distin-
Disclaimer: The views expressed in this paper are those of
the authors and do not reflect the official policy or position
of the Department of the Navy, the Department of
Defense, or the US government.
A. G. Sykes
Department of General Surgery, Naval Medical
Center, San Diego, CA, USA
M. E. Fallat (*)
The Hiram C. Polk, Jr., Department of Surgery/
Division of Pediatric Surgery, University of
Louisville School of Medicine, Louisville, KY, USA
e-mail: mary.fallat@louisville.edu
R. C. IgnacioJr.
Department of Surgery/Division of Pediatric Surgery,
University of California San Diego School of
Medicine, La Jolla, CA, USA
e-mail: rignacio1@rchsd.org
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_53
53
guish malignant from benign ovarian tumors is
necessary to optimize prompt evaluation and
appropriate treatment and to consider the options
for ovarian preservation.
53.2 Incidence and Epidemiology
The annual incidence of ovarian masses in chil-
dren is approximately 2.6 cases/100,000 girls per
year (Heo et  al. 2014). The majority of ovarian
masses are benign with a reported frequency of
malignant pediatric ovarian tumors ranging from
4% to 22% (Renaud et al. 2019). Ovarian malig-
nancy accounts for approximately 1–2% of child-
hood malignancies, and the incidence of malignant
ovarian tumors is 0.1 per 100,000 per year in chil-
dren younger than 9 years old and 1.1 per 100,000
per year in girls aged 10–19 years (Brookfield
et al. 2009). Children presenting with an adnexal
mass who have a palpable mass, mass diameter
≥8 cm, solid components on imaging, or a history
of precocious puberty have an increased likeli-
hood of having a malignant ovarian tumor
(Oltmann et al. 2010; Renaud et al. 2019).
The majority of ovarian lesions have no spe-
cific etiology with regard to environmental expo-
sure, genetic background, or drugs. However,
there are certain diseases, genes, or hormonal fac-
tors that might increase the risk for certain ovarian
tumors. Nulliparity, use of ­ ovulatory-­inducing
685
686 A. G. Sykes et al.

Table 53.1  Syndromes and diseases associated with an increased risk of ovarian tumors
Associated ovarian
Syndrome/disease Description pathology
Basal cell nevus Rare autosomal disorder associated with two or more basal cell Ovarian fibromas
(Gorlin) carcinomas, cysts of the jaw, characteristic facies, rib abnormalities,
syndrome and increased risk of medulloblastoma and cardiac fibromas
Chediak-Higashi Oculocutaneous albinism, pyogenic infections, and abnormalities of Sclerosing stromal
syndrome leukocyte granules resulting in defective phagocytosis tumors
Maffucci Multiple enchondromas, hemangiomas, and bony deformities Juvenile granulosa cell
syndrome tumors
Muir-Torre Variant of HNPCC,a associated with multiple cutaneous neoplasms Granulosa cell tumors
syndrome and visceral malignancies
McCune-Albright Café au lait spots, autonomic endocrine hyperactivity, polyostotic Recurrent ovarian
syndrome fibrous dysplasia follicles causing sexual
precocity
Ollier disease Multiple enchondromas of the tubular bones of the hand and feet, Juvenile granulosa cell
increased risk for chondrosarcomas tumors
Peutz-Jeghers Autosomal dominant disorder associated with melanocytic macules Predominantly sex
syndrome of the mouth or hamartomatous or adenomatous polyps of the stromal tumors
gastrointestinal tract
HNPCC—Hereditary non-polyposis colorectal cancer
a

drugs, lack of oral contraceptive use, and high-fat Table 53.2  World Health Organization histological clas-
diets are risk factors associated with later devel- sification of non-neoplastic ovarian lesionsa
opment of adult epithelial neoplasms. There are Solitary follicle cyst
also numerous syndromes associated with various Multiple follicle cysts (polycystic ovarian disease,
ovarian tumors (Table 53.1). sclerocystic ovaries)
The most significant risk factor for ovarian Large solitary luteinized follicle cyst of pregnancy and
puerperium
cancer is genetic predisposition. The tumor sup- Hyperreactio luteinalis (multiple luteinized follicle
pressor genes BRCA1 and BRCA2 are responsi- cysts)
ble for 70–90% of familial cases of ovarian Corpus luteum cyst
malignancy. The role of genetic testing in chil- Pregnancy luteoma
dren and adolescents for such mutations is con- Ectopic pregnancy
troversial since there are no medical benefits Stromal hyperplasia
Stromal hyperthecosis
including preventive measures and therapy that
Massive edema
have been studied in this population. Fibromatosis
Endometriosis
Cyst, unclassified (simple cyst)
53.3 Pathology Inflammatory lesions
Scully RE, Young RH, Clement PB: Atlas of tumor
a

Ovarian tumors can be classified as non-­ pathology, tumors of the ovary, maldeveloped Gonads,
Fallopian Tube and Broad Ligaments, 3rd ser, fascicle 23.
neoplastic or neoplastic (Tables 53.2 and 53.3). Washington DC, American Registry of Pathology, Armed
Approximately two-thirds of these lesions are Forces Institute of Pathology, 1998 (original citation)
neoplastic, but only a small percentage (20%) Arch Gynecol Obstet 2016 Apr;293(4):695–700.
have malignant changes. The most common clin- https://doi.org/10.1007/s00404-016-4035-8. Epub 2016
Feb 19 (updated citation)
ical and pathologic staging methods for ovarian
cancer are the International Federation of
Gynaecology and Obstetrics (FIGO) (Table 53.4)
and the Children’s Oncology Group (COG) stag- and the COG classification is current as of 2020
ing systems (Table  53.5). The FIGO classifica- (Prat and FIGO Committee on Gynecologic
tion has not been formally updated since 2015, Oncology 2015; Duska and Kohn 2017).
53  Ovarian Tumors 687

Table 53.3  World Health Organization histological clas- Table 53.3 (continued)

sification of tumors of the ovarya Website: https://www.pathologyoutlines.com/topic/ova-
•  Surface epithelial-stromal tumors rytumorwhoclassif.html
 –  Serous tumors: Cite: Ehdaivand S WHO classification. PathologyOutlines.
com website. https://www.pathologyoutlines.com/topic/ova-
   Benign (cystadenoma)
rytumorwhoclassif.html. Accessed November 28th, 2020
   Borderline tumors (serous borderline tumor)
   Malignant (serous adenocarcinoma)
 – Mucinous tumors, endocervical-like and Table 53.4  Staging of primary carcinoma of the ovary:
intestinal type: International Federation of Gynaecology and Obstetrics
   Benign (cystadenoma) (FIGO)
   Borderline tumors (mucinous borderline tumor) Stage Extent of disease
   Malignant (mucinous adenocarcinoma) 0 No evidence of primary tumor
 –  Endometrioid tumors: I Tumor confined to the ovaries
   Benign (cystadenoma) IA Tumor limited to one ovary, capsule intact
  Borderline tumors (endometrioid borderline tumor) No tumor on ovarian surface
   Malignant (endometrioid adenocarcinoma) No malignant cells in ascites or peritoneal
 –  Clear cell tumors: washings
   Benign IB Tumor limited to both ovaries, capsule intact
   Borderline tumors No tumor on ovarian surface
   Malignant (clear cell adenocarcinoma) No malignant cells in ascites or peritoneal washings
 – Transitional cell tumors: IC Tumor limited to one or both
   Brenner tumor Ovaries, with one of the following:
   Brenner tumor of borderline malignancy Surgical spillage, capsule ruptured before
   Malignant Brenner tumor surgery, tumor on ovarian surface, malignant
   Transitional cell carcinoma (non-Brenner type) cells in ascites or peritoneal washings
 – Epithelial-stromal: II Tumor involves one or both ovaries with pelvic
extension
   Adenosarcoma
IIA Extension to or implants on uterus and/or
  Carcinosarcoma (formerly mixed Müllerian
fallopian tubes
tumors)
IIB Extension to other pelvic organs
•  Sex cord-stromal tumors
III Tumor involves one or both ovaries with
 –  Granulosa tumors:
cytologically or histologically confirmed spread to
 – Fibromas the peritoneum outside the pelvis and/or
 – Fibrothecomas metastases to retroperitoneal lymph nodes
 – Thecomas IIIA1 Positive retroperitoneal lymph nodes only
   Sertoli cell tumors: (cytologically or histologically proven)
 –  Leydig cell tumors IIIA1 Metastasis 10 mm or less in greatest dimension
   Sex cord tumor with annular tubules (i)
   Gynandroblastoma IIIA1 Metastasis more than 10 mm in greatest
   Steroid (lipid) cell tumors (ii) dimension
•  Germ cell tumors IIIA2 Microscopic peritoneal metastasis beyond the
 –  Teratoma: pelvis with positive retroperitoneal lymph nodes
   Immature IIIB Macroscopic peritoneal metastasis beyond the
   Mature pelvis 2 cm or less in greatest dimension with
   Solid or without retroperitoneal lymph node
   Cystic (dermoid cyst) metastasis
 –  Monodermal (e.g., struma ovarii, carcinoid) IIIC Macroscopic peritoneal metastasis beyond the
pelvis more than 2 cm in greatest dimension with
 – Dysgerminoma
or without retroperitoneal lymph node metastasis
 –  Yolk sac tumor (endodermal sinus tumor)
IV Distant metastasis beyond the peritoneal cavity
 –  Mixed germ cell tumors
•  Malignant, not otherwise specified Staging Classification and Clinical Practice Guidelines for
 –  Metastatic cancer from non-ovarian primary: Gynaecologic Cancers. FIGO Committee on Gynaecologic
Oncology. Intl J Gynaecol Obstetr 2000; 70: 207–312
   Colonic, appendiceal
Updated: Duska LR, Kohn EC. The new classifications
   Gastric of ovarian, fallopian tube, and primary peritoneal cancer
   Breast and their clinical implications. Ann Oncol. 2017 Nov
Table 53.3 reference
a 1;28(suppl_8): viii8–viii12
688
Table 53.5  Clinical and pathologic staging of ovarian
germ cell tumors: Children’s Oncology Group (COG)a
Stage Extent of disease
I Limited to ovary (ovaries) peritoneal washings
negative; tumor markers are normal after
appropriate half-life decline (AFP, 5 days;
3-HCG, 16 h)
II Microscopic residual or disease in lymph nodes
< 2 cm; peritoneal washings normal; markers
positive or negative
III Gross residual disease or biopsy only; lymph
nodes > 2 cm; contiguous spread to other organs
(omentum, intestine, bladder); peritoneal
washings positive for malignant cells; markers
negative or positive
IV Distant metastasis including the liver
a
von Allmen D. Malignant lesions of the ovary in child-
hood. Sem Pedr Surg 2005; 14:100–5
https://www.childrensoncologygroup.org/index.php/
newly-­diagnosed-­with-­germ-­cell-­tumors
53.4 Diagnosis and Evaluation
53.4.1 Clinical Presentation
The clinical signs and symptoms associated with
ovarian tumors are nonspecific. The most com-
mon presenting symptom is abdominal pain, fol-
lowed by a palpable abdominal or pelvic mass.
Patients may also present with abdominal disten-
sion, nausea, vomiting, anorexia, weight loss,
constipation, urinary frequency, precocious
puberty, and virilization or may be asymptomatic
if the tumor is found incidentally on imaging
(Heo et  al. 2014; Péroux et  al. 2015). Acute
abdominal or pelvic pain can occur due to tumor
torsion, hemorrhage, or rupture. The occurrence
of chronic pain varies depending on the type of
ovarian tumor, its size, and the extent of disease.
Large tumors can cause compression of adjacent
organs causing cramping pain, constipation, or
urinary symptoms. Abdominal distention can
also be due to ascites, although rare in the pediat-
ric population. Vaginal bleeding is an uncommon
symptom. Recent studies have shown that genito-
urinary complaints may be associated with ovar-
ian cancer. Regardless, there are no specific
symptoms that indicate if the ovarian mass is
malignant or benign (Renaud et al. 2019).
The physical exam is usually revealing for an
abdominal mass in children or pelvic mass in
A. G. Sykes et al.
adolescents. Associated tenderness may occur
with torsion, hemorrhage, or rupture of ovarian
tumors, especially germ cell tumors or sex cord
stromal tumors (Heo et  al. 2014). A fluid shift
may be detected in the presence of ascites. Pelvic
examination is usually reserved for sexually
active patients. A bimanual examination can aid
in palpation of smaller adnexal masses. Bimanual
rectal examination may be helpful in prepubertal
children and infants.
Ovarian lesions that are hormonally active can
cause precocious puberty, menstrual irregularity,
or virilization. Features of isosexual or ­ambiguous
development of the genitalia and breasts in pre-
menarchal girls or menstrual irregularity in post-
menarchal girls can be associated with a sex cord
stromal tumor (Heo et  al. 2014). Alternatively,
the presence of hirsute or masculine features can
be due to a Sertoli-Leydig tumor.
53.4.2 Laboratory Tests
The evaluation of suspected ovarian masses may
include laboratory tests such as tumor markers,
hormone levels, and cytogenetic studies. All
patients should have a quantitative human chori-
onic gonadotropin (β-HCG) since some ovarian
tumors secrete β-HCG.  In postmenarchal
patients, this is also used to exclude an intrauter-
ine or ectopic pregnancy. Ultrasound will help
differentiate between these diagnoses. Fine nee-
dle aspiration (FNA) is not recommended as a
diagnostic tool in children due to low sensitivity
and cannot be used to exclude malignancy
(Hermans et al. 2016).
Certain ovarian malignancies secrete specific
tumor markers (Table  53.6). The main specific
tumor markers for ovarian neoplasms are β-HCG,
alpha-fetoprotein (AFP), and CA-125. These
tumor markers can aid in the diagnosis of certain
ovarian pathologies, although they can also be
elevated with other nongynecological tumors or
medical conditions. The clinical presentation and
radiographic findings may dictate which tumor
markers to obtain. Incidental findings of a mixed
cystic and solid ovarian tumors will warrant pre-
operative AFP, β-hCG, and CA-125 tests. AFP
and β-hCG are highly specific for malignant
53  Ovarian Tumors 689

Table 53.6  Ovarian tumor markers

CA-125 HE4 AFP hCG INH E2 TT 17-KS Gn MIS
Germ cell tumors
Dysgerminoma -/↑ -/↑ -/↑ ↑/-
Yolk sac tumor (endodermal sinus tumor) -/↑ ↑↑
Choriocarcinoma -/↑ ↑↑ ↑/- ↑
Embryonal carcinoma -/↑ ↑/- ↑/- ↑/-
Immature teratoma -/↑ ↑/- ↑/-
Mixed germ cell tumor -/↑ ↑/- ↑/- -/↑
Epithelial-stromal tumors
Serous carcinoma ↑↑ ↑↑
Mucinous carcinoma ↑ ↑↑ ↑↑
Endometrioid carcinoma ↑↑ ↑↑ ↑
Sex cord-stromal tumors
Granulosa cell tumor -/↑ ↑ ↑↑ ↑ ↓ ↑↑
Juvenile granulosa cell tumor -/↑ ↑ ↑↑ ↑ - / ↑^ ↓ ↑↑
Thecoma-fibroma -/↑ ↑ ↑ - / ↑^
Sertoli-Leydig cell tumor -/↑ - / ↑^ ↑ - / ↑^ - / ↑^ - / ↑^ ↓ ↑
Sex cord tumor with annular tubules -/↑ ↑ ↑ ↑↑
Steroid cell tumor, NOS -/↑ ↑↑ ↑↑ ↓
Other ovarian tumors
Gonadoblastoma -/↑ - / ↑^ ↑ ↑ ↓
^Indicates rare variant of the tumor
Comments: Tumor markers may assist in diagnosis, as well as monitor response to therapy and detect relapse and/or
progression. CA-125 levels may be slightly elevated in any of the ovarian tumors. HE4 is a newer tumor marker with
improved sensitivity and specificity compared to CA-125. Lactate dehydrogenase (LDH) levels are useful for staging
and risk assessment in germ cell tumors

Table key
CA-125 Cancer antigen 125
HE4 Human epididymis protein 4
AFP Alpha-fetoprotein
hCG Human chorionic gonadotropin
INH Inhibin
E2 Estradiol
TT Testosterone
17-KS Urinary 17-ketosteroid
Gn Gonadotropin
MIS Müllerian-inhibiting substance
Table originally published in Pediatric Surgery. von Allmen D., Fallat M.E. (2019) Ovarian Tumors. In: Puri P.,
Höllwarth M. (eds) Pediatric Surgery. Springer Surgery Atlas Series. Springer, Berlin, Heidelberg. The authors would
like to acknowledge the assistance of Dr. Robert Debski, Chief of Pediatric Pathology, Norton Children’s Hospital,
Louisville, KY in creation of the table. Used with permission

masses in neonates with ovarian masses and
should be obtained for all children and adoles-
cents who present with an adnexal mass (Heo
et al. 2014). Elevated AFP, β-hCG, and CA-125
levels will raise suspicion of ovarian malignancy;
however, these markers can also be elevated in
benign ovarian lesions, and the absence of an
elevated tumor marker does not rule out malig-
nancy (Heo et  al. 2014; Spinelli et  al. 2012).
Patients who present with signs of precocious
puberty will need preoperative follicle-­
stimulating hormone (FSH), luteinizing hormone
(LH), lactate dehydrogenase (LDH), and estra-
diol serum levels (Papic et al. 2014a, b).
β-HCG is a glycoprotein with a similar structure
to luteinizing hormone (LH). The two hormones
only differ by the last 30 amino acids of the beta-
subunit. β-HCG is secreted by syncytiotrophoblas-
tic cells and functions to support the corpus luteum.
β-HCG is positive in intrauterine or ectopic preg-
690
nancies and hydatidiform moles. Neoplastic causes
of a positive result include almost all cases of cho-
riocarcinoma, 10–30% of seminomas, and 5–35%
of cases of dysgerminomas.
Alpha-fetoprotein is a product of fetal yolk
sac, liver, and gastrointestinal tissue and can
remain elevated during the first 8 months of life.
Maternal serum levels of AFP serve as an
­indicator of neural tube defects (spina bifida or
anencephaly), ventral wall defects (gastroschisis
or omphalocele), congenital nephrosis, and fetal
hemorrhage. As a tumor marker, it is a valuable
test for primary malignancies of the liver, hepatic
metastases for certain cancers (i.e., stomach or
pancreas), and germ cell tumors of the testis and
ovary. It is elevated in yolk sac tumors, embryo-
nal carcinomas, malignant teratomas, and Sertoli-­
Leydig carcinomas. Levels of AFP should return
to normal after surgery and chemotherapy treat-
ment. Recurrent elevations after treatment usu-
ally indicate relapse of cancer, often preceding
clinical or radiographic evidence.
Serum CA-125 is a glycoprotein derived from
coelomic epithelium. This tumor marker is associ-
ated with numerous malignancies in the liver, pan-
creas, breast, colon, and lung. It is also associated
with nonmucinous epithelial ovarian carcinoma,
carcinoma of the fallopian tubes, and endometrial
cancers. CA-125 can be used to distinguish
between a benign and malignant ovarian mass with
a sensitivity of 61–90% and specificity of 71–93%
when using a cutoff of 30 U/mL. CA-125 is also
useful in monitoring residual or recurrent disease
after surgery (Van Calster et al. 2007). An elevated
or rising level during chemotherapy is associated
with poor response and progression of disease.
There are other nonspecific tumor markers that
have some prognostic value. Carcinoembryonic
antigen (CEA) is more commonly associated with
colon cancer, but it is also present in 50–70% of
cases of serous carcinoma of the ovary. Serum
level of lactic dehydrogenase (LDH) may corre-
late with ovarian tumor activity due to high cell
turnover. However, an elevated LDH is also asso-
ciated with other malignancies such as neuroblas-
toma and lymphoma.
Approximately 10% of ovarian masses can pro-
duce hormones resulting in abnormal menses, early
thelarche, precocious puberty, or virilization.
A. G. Sykes et al.
Elevated estrogen levels from ovarian cysts or sex
stromal tumors can cause precocious puberty.
Sertoli-Leydig tumors, yolk sac tumors, and poly-
cystic ovaries can produce high levels of androgens
leading to virilization. Elevated CA-125 levels are
more often associated with epithelial ovarian can-
cers, which are extremely rare before menarche.
Therefore, CA-125 levels should not be routinely
obtained in premenarchal girls (Heo et al. 2014).
Inhibin is a useful marker for granulosa cell
tumor diagnosis and surveillance (Heo et al. 2014).
Inhibin is a polypeptide hormone produced by the
granulosa cells of normal ovarian follicles. This
hormone is present in certain body fluids as dimers
comprising an alpha- and beta-­subunit. Elevations
in inhibins A and B have been reported in granu-
losa cell tumors. The sensitivity of inhibin A test-
ing for the diagnosis of granulosa cell tumor was
67% with a specificity of 100%, compared to 89%
and 100%, respectively, for inhibin B. Therefore,
inhibin B appears to be the predominant form of
inhibin secreted by these tumors and reflects dis-
ease status more accurately than inhibin A.
All hormone tests should be selectively obtained
based on the clinical history and physical examina-
tion of the patient. Of note, the elevation of any
individual tumor marker (AFP, β-HCG, CA-125,
LDH, estradiol, or testosterone) has a 0.76 PPV,
0.86 NPV, and 0.83 accuracy for malignancy.
Therefore, tumor markers may be most useful in
the determination of malignancy when ordered as a
standard panel preoperatively (Renaud et al. 2019).
53.4.3 Radiologic Studies
53.4.3.1 Ultrasonography
Ultrasound is the initial imaging study used to eval-
uate ovarian masses and can be done pre- or post-
natally as an abdominal or pelvic exam. Prenatal
ultrasounds commonly identify fetal abdominal
cysts and are helpful in distinguishing ovarian cysts
from intestinal, uterine, or genitourinary abnormal-
ities. During infancy and early childhood, ovarian
masses are typically intra-­abdominal due to the
narrow and small pelvis. In children and adoles-
cents, initial imaging to evaluate the ovaries should
be done with transabdominal ultrasonography (US)
without the use of an endovaginal probe.
53  Ovarian Tumors 691

Transvaginal pelvic US may be performed in sexu-
ally active adolescents and can be useful to visual-
ize larger ovarian tumors. Ultrasonography is used
to distinguish cystic versus solid ovarian masses,
and there are imaging clues that aid in determining
whether a mass is benign or malignant (Table 53.7).
Several ultrasound scoring systems, such as the
Ueland index, which incorporate ovarian volume
and intrinsic mass characteristics have been devel-
oped to improve the ability to screen for malig-
nancy using ultrasound (Renaud et  al. 2019). In
2008, the International Ovarian Tumor Analysis
(IOTA) group introduced ultrasound-based rules
for predicting benign and malignant ovarian
tumors (Timmerman et al. 2008). The IOTA clas-
sification includes five rules to suggest a benign

Table 53.7  Radiologic findings for the most common benign and malignant ovarian lesions (Shah et al. 2011; Lam and
Chavhan 2018)
Computed
Ultrasound tomography Magnetic resonance imaging
Benign lesions
Ovarian cysts Thin-walled, anechoic, or Thin-walled, Thin-walled, fluid-filled structure
hypoechoic structure, with or fluid-filled structure with variable intensity on T1-
without septations; no solid and T2-weighted imaging
elements or mural nodules Hemorrhagic cysts will have
high signal intensity on T1- and
T2-weighted images
Germ cell tumors (GCTs)
Mature cystic Complex mass with cystic and Complex mass Unilocular fat-containing cystic
teratoma solid components, may contain containing mass
Immature teratoma echogenic components with calcifications, soft Scattered punctate foci of fat
acoustic shadowing tissue, and adipose with microcystic areas of fluid
(calcifications) or without tissue signal intensity
shadowing (adipose tissue)
Embryonal Nonspecific findings that may overlap with other GCTs
carcinoma
Dysgerminoma Nonspecific solid mass, with Solid, encapsulated Lobulated solid mass with
enhancement of fibrovascular mass; may have fine fibrovascular septa
septae stippled calcifications
Yolk sac tumor Well-circumscribed, hypervascular Heterogenous mixed solid-cystic mass with large dilated
mass with variable amount of intratumoral vessels
echogenicity (dependent upon
amount of internal hemorrhage,
necrosis, and calcification)
Choriocarcinoma Heterogenous mass with cystic and solid components
Mixed germ cell Variable appearance; combination Circumscribed, Large solid mass with extensive
tumor of other germ cell tumor features mottled mass with hemorrhage and necrosis with
(gonadoblastoma) punctate calcifications combination of other germ cell
tumor features
Sex cord stromal tumors
Sertoli-Leydig cell Well-circumscribed, hypoechoic Well-defined, solid Solid mass, often with many
tumors mass mass peripheral cysts, may be
hypointense on T1- and
T2-weighted images
Granulosa-theca cell Variable appearance; can be heterogenous with variable Mixed solid-cystic mass with
tumors amounts of cystic and solid components sponge-like appearance on
T2-weighted images
Epithelial ovarian Hypoechoic, thin-walled simple Large, simple cysts or Large unilocular simple cyst or
tumors cyst or multicystic mass multicystic masses multilocular cystic mass;
enhancing papillary projections
or solid enhancing mural
components in borderline or
malignant tumors
692
tumor (B-rules) and five rules to predict a malig-
nant tumor (M-rules). The B-rules include (1) uni-
locular cyst; (2) presence of solid components
where the largest solid component is < 7  mm in
largest diameter; (3) acoustic shadows; (4) smooth
multilocular tumor less than 100  mm in largest
diameter; and (5) no detectable blood flow on
Doppler examination. The M-rules include (1)
irregular solid tumor; (2) ascites; (3) at least four
papillary structures; (4) irregular multilocular-­
solid tumor with the largest diameter of at least
100 mm; and (5) very high color content on color
Doppler examination. These rules were most
recently assessed for utility in predicting malig-
nant ovarian tumors in the pediatric population in
2019. Of all IOTA criteria, size >100mm was
found to be most predictive of ovarian malignancy
in children, with sensitivity increasing to 100%
and specificity to 88% by increasing diameter cut-
off to >130  mm (Depoers et  al. 2019). Further
research is needed to better define the utility of
IOTA classification in the pediatric population.
An ultrasound finding found of an ovarian cres-
cent sign (OCS) is indicative of a benign ovarian
mass. The OCS is a rim of healthy ovarian tissue on
the ipsilateral ovary in the presence of an ovarian
mass. The absence of the OCS is associated with
malignancy with sensitivity 96–100% and specific-
ity 92–93%, although this is less predictive in pre-
menarchal patients, with smaller ovaries, and in the
setting of ovarian torsion (Stankovic et al. 2010).
Scoring systems are a useful adjunct for preop-
erative risk stratification; however, due to their limi-
a b
Fig. 53.1  CT scan of benign versus malignant ovarian
tumor: (a) Demonstrates a benign ovarian tumor with no
evidence of direct extension into the surrounding soft tis-
sue structures. The teratoma was composed of adipose tis-
A. G. Sykes et al.
tations, they must be used with other assessment
measures. Combining scoring systems with imaging
findings appears to be the safest means of screening
for malignancy, as a combination of the Ueland
index with the absence of the OCS was found to
have 93% sensitivity, 97% specificity, PPV 86%,
and NPV 99% for malignancy (Renaud et al. 2019).
53.4.3.2 Computed Tomography
Computed tomography (CT) has the disadvantage
of radiation exposure, a factor of particular impor-
tance in children. Nevertheless, CT of the abdo-
men and pelvis is a valuable imaging study and
useful in determining the anatomic location and
origin of an ovarian mass, the extent of local
involvement, and possibility of metastases. Prior
to obtaining a CT, providers should consider alter-
native imaging modalities that do not expose the
child to radiation, such as an MRI, which can pro-
vide information on tumor characterization, local-
ization, and staging (Lam and Chavhan 2018;
Levin 2018). If CT imaging is indicated, measures
should be taken to ensure that the radiation dose
follows the “as low as reasonable achievable“
(ALARA) concept, to minimize radiation expo-
sure (Slovis 2003; Levin 2018). Suggestive find-
ings of malignancy include large mass size, solid
nature of the tumor, peritoneal implants, ascites,
or suspicious calcifications (Fig. 53.1). Computed
tomography is more accurate for disease staging
than ultrasonography. This modality is also com-
monly used in detecting recurrent or metastatic
ovarian cancer after treatment.
sue, teeth, and hair. (b) Demonstrates a large malignant
tumor with irregular borders, thick septations, adenopa-
thy, and invasion into the adjacent pelvic organs. The final
pathology was a mixed germ cell tumor
53  Ovarian Tumors
53.4.3.3 Magnetic Resonance
Imaging
Magnetic resonance imaging (MRI) is preferred
over CT in the pediatric population because it
avoids exposure to ionizing radiation (Lam and
Chavhan 2018; Levin 2018). It is a useful adjunct
in differentiating uterine tumors from ovarian
masses and can provide additional diagnostic
information to guide management. MRI can assist
in further characterizing an ovarian lesion when
ultrasound findings are indeterminate and can help
distinguish benign from malignant lesions, which
can assist with operative planning for the potential
use of fertility-sparing approaches (Lam and
Chavhan 2018). Based on signal intensity, MRI
can further identify specific ovarian pathologies
such as dermoid cysts, endometriomas, and ovar-
ian torsion. One major disadvantage of MRI is the
prolonged imaging time, which may contribute to
motion artifact and mandate the use of sedation.
53.5 Treatment of Ovarian
Tumors
The majority of ovarian masses are benign, mak-
ing gonadal preservation an option. The treatment
for malignant lesions is generally oophorectomy,
depending on the specific tumor. Even with bilat-
eral oophorectomy, the option for uterine sparing
should be considered for potential in vitro fertil-
ization. Each treatment option should be individu-
alized based on the pathology, extent of the
disease, and patient and parental desire to preserve
fertility. Reproductive counseling with a multidis-
ciplinary team should always be performed prior
to the procedure (Van Heerden & Tjalma 2019;
Vitale et  al. 2019). Most benign tumors can be
approached laparoscopically. The laparoscopic
approach is safe in children and adolescents with
adnexal masses ≤ 8cm in diameter and without
complex features (Rogers et al. 2014).
Recent literature suggests pediatric surgeons
manage adnexal masses by oophorectomy more
often (46–55%) than pediatric gynecologists
(18%) and presence of a gynecologist significantly
improved rates of ovarian preservation (Lawrence
et al. 2019; Rogers et  al. 2014; Hermans et  al.
2015; Bergeron et al. 2017). Limited evidence has
693
suggested laparoscopic approach for presumably
benign ovarian masses has better outcomes
(shorter operating time, less intraoperative bleed-
ing, and reduced length of stay) when compared to
open surgery with similar complication rates (Liu
et al. 2013; Qazi et al. 2020).
53.5.1 Benign Tumors
53.5.1.1 Ovarian Cysts
Neonatal ovarian cysts have been described in
detail in the literature. Most of these cysts are
benign and tend to spontaneously regress.
Incidental finding of a simple cyst on ultrasound
requires only close follow-up at 6–12-week inter-
vals. Intraoperative discovery of ovarian cysts in
infants less than 12 months of age can be managed
with aspiration or fenestration. Surgical manage-
ment is recommended for cystic masses that are
symptomatic, complex in character, increasing in
size over time, or suggestive of malignancy.
Ovarian cysts are not uncommon in children
and adolescents. The majority of these cysts arise
from mature follicles and are benign. These cysts
tend to be thin-walled, unilocular, and unilateral
and can be discovered incidentally, during opera-
tions for other disease processes, or during evalu-
ation for adnexal masses. The treatment for these
lesions is based on symptomatology, endocrine
activity, and size. As ovarian cysts enlarge beyond
4 cm, they have an increased risk for torsion, rup-
ture, or hemorrhage. Some cysts have endocrine
activity that can cause early isosexual
­development or premature thelarche. These latter
groups of lesions should be excised with preser-
vation of adjacent ovarian tissue. The operative
approach can be through a midline laparotomy or
Pfannenstiel incision or via laparoscopy. Other
indications for cyst excision include persistence
or size increase over 2–3 months of surveillance
or concern for malignancy.
Treatment of newborn ovarian cysts remains
controversial. While the majority of pediatric sur-
geons selectively aspirate or operate on newborn
cysts greater than 4  cm in diameter, some have
found good results with the performance of close
observation and serial ultrasonography (Polis
et al. 2017).
694
53.5.1.2 Mature Cystic Teratomas
The risk for malignancy in mature cystic terato-
mas (MCTs) is only 2%, and ovarian preserva-
tion with cyst removal via laparoscopy or
laparotomy and postoperative surveillance is the
preferred treatment. The postoperative follow-up
consists of annual pelvic ultrasound in prepuber-
tal children to evaluate any disease in the contra-
lateral ovary or early signs of recurrent lesions
and an annual pelvic exam or pelvic ultrasound in
postmenarchal adolescents.
53.5.2 Malignant Tumors
53.5.2.1 Malignant Germ Cell Tumors
Germ cell tumors (GCTs) are the most common
type of ovarian tumor in children and adolescents
(Heo et  al. 2014; Papic et  al. 2014a, b).
Approximately 60–80% of pediatric ovarian
malignancies are GCTs, and 20% of ovarian
GCTs are malignant (Rescorla 2012; Cecchetto
2014; Heo et al. 2014). These tumors vary in their
presentation and malignant potential and can also
arise in other areas of the body. GCTs are more
likely than any other malignant ovarian tumor to
cause ovarian torsion in pre- and postmenarchal
girls (Savic et al. 2008). GCTs originate from one
or more embryonic cell layers (endoderm, meso-
derm, and ectoderm). Because of their variability
in tissue origin and cellular behavior, they can be
benign or malignant. Proper staging for malig-
nant GCTs depends on intraoperative findings
including the extent of tumor involvement, peri-
toneal washings, and results of tumor markers
(Table 53.5). The surgical therapy for germ cell
tumors is similar regardless of the tumor type.
Operative management includes (1) surgical
resection (i.e., unilateral oophorectomy or sal-
pingo-oophorectomy) for optimal cure; (2) intra-
operative evaluation of potential tumor
involvement in the contralateral ovary, adjacent
structures, regional and retroperitoneal lymph
nodes, all peritoneal surfaces, diaphragm, and
liver; (3) collection of peritoneal washings for
cytology; and (4) omentectomy. Any suspicious
lesions on the contralateral ovary are biopsied.
Preservation of fertility should be always consid-
ered and is dependent on the extent of disease.
A. G. Sykes et al.
Chemotherapy regimens for germ cell tumors
are predominantly platinum based. The most
common regimen for advanced stages of dysger-
minoma or endodermal sinus tumor consists of
bleomycin, etoposide, and cisplatin. The advan-
tages of this regimen include less risk of future
infertility and improved survival for children
with endodermal sinus carcinoma. Embryonal
carcinoma can be treated with the previously
mentioned regimen or a cyclophosphamide-­
methotrexate-­dactinomycin combination.
Immature stage I or II teratomas can be treated
with surgery only, whereas higher-stage tumors
will require adjuvant chemotherapy such as vin-
cristine, actinomycin D, and cyclophosphamide
with or without cisplatin.
The use of conservative surgery with platinum-­
based chemotherapy for all stages of disease has
demonstrated excellent survival with outcomes
similar to more radical resections. Therefore,
contralateral ovarian preservation is considered
safe in the treatment of GCTs; however, tumor
biology and appropriate use of chemotherapy
have a direct effect on recurrence (Renaud et al.
2019).
53.5.2.2 Sex Cord-Stromal Tumors
The majority of the remaining 20–40% of pediat-
ric ovarian tumors are sex-stromal and epithelial
tumors (Heo et  al. 2014). Sex cord-stromal
tumors (SCSTs) originate from stromal
­components of the ovary such as granulosa-theca
cells or Sertoli-Leydig cells. These tumors are
more common in patients less than 15 years old
and account for 10–20% of all pediatric ovarian
tumors (Heo et  al. 2014). Occasionally, these
tumors can produce hormones leading to andro-
genization or isosexual precocious puberty.
Juvenile granulosa-theca cell tumors are the
most common SCSTs and account for 70% of all
ovarian SCSTs in patients younger than 20 years
and up to 10% of pediatric ovarian malignancies
(Heo et  al. 2014). These tumors are commonly
functional and associated with pseudo-­precocious
puberty in 80% of patients. Other symptoms that
have been described with these neoplasms are
galactorrhea, menstrual abnormalities, vaginal
bleeding, labial swelling, early development of
axillary or pubic hair, and increase in overall
53  Ovarian Tumors
growth. Due to high associated estrogen levels, a
consequent increase in vascularity of these
tumors can lead to significant hemoperitoneum
and peritonitis if rupture occurs. These tumors
tend to be more aggressive than their adult coun-
terpart. The degree of malignancy is related to the
proportion of granulosa cells present. Pure granu-
losa cell tumors are frequently malignant. Tumors
composed of mixed granulosa and theca cells are
less frequently malignant, and pure thecoma cell
tumors are usually benign. Most patients diag-
nosed with juvenile granulosa-theca cell tumors
present with stage I malignancies and have a
favorable prognosis. These patients are ade-
quately treated with a unilateral oophorectomy or
salpingo-oophorectomy with survival rates close
to 100%. However, those with more advanced
disease (stage III) require hysterectomy with
bilateral salpingo-oophorectomy, radiotherapy,
and multi-drug chemotherapy (i.e., carboplatin
and etoposide).
Sertoli-Leydig cell tumors, also known as
arrhenoblastomas, account for up to 30% of sex
stromal neoplasms, but less than 0.5% of all
malignant ovarian masses (Heo et  al. 2014).
This tumor is the most common androgen-pro-
ducing tumor with approximately 30% of
patients presenting with masculinizing changes
such as amenorrhea, loss of female habitus,
breast atrophy, hirsutism, thinning of hair, acne,
clitoral enlargement, and accelerated growth
(Shanbhogue et al. 2010; Heo et al. 2014). The
physiologic effects are due to the accumulation
of testosterone, which can also be used as a bio-
logical marker for these tumors. These neo-
plasms tend to have elevated levels of AFP and
CA-125. Most often these tumors are diagnosed
early-­stage and have a favorable prognosis (Heo
et al. 2014). Treatment consists of surgery and
chemotherapy. Unilateral oophorectomy or
adnexal resection is adequate therapy and pre-
serves potential childbearing options in the
future. Advanced stages, bilateral disease, and
ruptured and/or poorly differentiated tumors
require more aggressive surgical resection (sim-
ilar to granulosa cell tumors) and chemotherapy.
Gonadotropin-releasing hormone agonists and
oral contraceptives can aid in treatment during
and after chemotherapy.
695
53.5.2.3 Epithelial Tumors
Epithelial tumors derive from the germinal epi-
thelium and are considered well-differentiated
neoplasms. These ovarian tumors account for
15–20% of childhood ovarian malignancies (Heo
et al. 2014). These tumors are rare before men-
arche, which is thought to be due to hormonal
stimulation triggering their development (Tsai
et al. 2001; Heo et al. 2014). They are commonly
serous or mucinous tumors and have an indolent
course. CA-125 is a useful marker in malignant
epithelial tumors and detecting recurrent disease
after chemotherapy. Treatment of epithelial
tumors is based on the FIGO staging system.
Stage 1A disease is amenable to ovarian preser-
vation with unilateral salpingo-oophorectomy.
Stage IB tumors require bilateral salpingo-­
oophorectomy, with preservation of the uterus for
potential in  vitro fertilization. Since serous
tumors can be bilateral in 20% of cases, evalua-
tion of the contralateral ovary is essential. More
advanced disease requires surgical cytoreduction,
total abdominal hysterectomy, bilateral salpingo-­
oophorectomy, and omentectomy. Adjuvant che-
motherapy has proved to have a 60–70% response
rate, but the overall 5-year survival is only 20%.
Radiation therapy for advanced epithelial tumors
currently remains controversial.
53.5.2.4 Borderline Epithelial Tumors
Borderline ovarian tumors (BOT) are in the WHO
epithelial ovarian tumor classification in the cat-
egories of serous and mucinous, endometrioid,
clear cell, and transitional (Brenner) borderline
tumors. These are tumors with increased mitotic
activity and nuclear atypia but without stromal
invasion. They generally present in early stages
in adolescent patients, have a good prognosis, but
share biochemical and radiological characteris-
tics with both benign and malignant tumors
(Morotti et al. 2012; Renaud et al. 2019). These
tumors can be difficult to diagnose preoperatively
and should be in the differential diagnosis when
preoperative tests do not strongly favor a diagno-
sis of either a benign or malignant lesion.
Ovarian-sparing cystectomy is an option and may
be associated with a higher risk of recurrence, but
can be salvaged with additional surgery, often
without compromising the effect on survival.
696
53.6 Conclusions
Ovarian tumors in children and adolescents are
rare. Although the majority of tumors are benign,
the diagnosis and management present various
challenges that require a wide range of expertise.
The diagnosis of ovarian cancer in children and
adolescents and young adults is always challeng-
ing. Multidisciplinary teams consisting of pediat-
ric surgeons, gynecologists, and oncologists may
be the most appropriate to care for these complex
and diverse patients. Because surgery is the pri-
mary treatment for ovarian tumors, ovarian sal-
vage with fertility preservation and the use of a
minimally invasive surgical technique are impor-
tant in children and adolescents.
Conflicts of Interest
None.
Funding
None.
Acknowledgements None.
References
Bergeron LM, Bishop KC, Hoefgen HR et  al (2017)
Surgical management of benign adnexal masses in the
pediatric/adolescent population: an 11-year review. J
Pediatr Adolesc Gynecol 30(1):123–127
Brookfield KF, Cheung MC, Koniaris LG, Sola JE,
Fischer AC (2009) A population-based analysis of
1037 malignant ovarian tumors in the pediatric popu-
lation. J Surg Res 156(1):45–49
Cecchetto G (2014) Gonadal germ cell tumors in
children and adolescents. J Indian Assoc Pediatr
Surg 19:189–194. [serial online] [cited 2021 Feb
21]. Available from: https://www.jiaps.com/text.
asp?2014/19/4/189/141995
Depoers C, Martin FA, Nyangoh Timoh K et al (2019) A
preoperative scoring system for adnexal mass in chil-
dren and adolescents to preserve their future fertility. J
Pediatr Adolesc Gynecol 32(1):57–63
Duska LR, Kohn EC (2017) The new classifications
of ovarian, fallopian tube, and primary peritoneal
cancer and their clinical implications. Ann Oncol
28(suppl_8):viii8–viii12
Heo SH, Kim JW, Shin SS et  al (2014) Review of
ovarian tumors in children and adolescents:
A. G. Sykes et al.
radiologic-­pathologic correlation. Radiographics
34(7):2039–2055
Hermans AJ, Kluivers KB, Wijnen MH, Bulten J,
Massuger LF, Coppus SF (2015) Diagnosis and
treatment of adnexal masses in children and adoles-
cents. Obstet Gynecol 125(3):611–615. https://doi.
org/10.1097/AOG.0000000000000665
Hermans AJ, Kluivers KB, Janssen LM et  al (2016)
Adnexal masses in children, adolescents and women
of reproductive age in the Netherlands: a nation-
wide population-based cohort study. Gynecol Oncol
143(1):93–97
Lam CZ, Chavhan GB (2018) Magnetic resonance imag-
ing of pediatric adnexal masses and mimics. Pediatr
Radiol 48(9):1291–1306
Lawrence AE, Gonzalez DO, Fallat ME, Aldrink JH,
Hewitt GD, Hertweck SP, Onwuka A, Bence C, Burns
RC, Dillon PA, Ehrlich PF, Fraser JD, Grabowski JE,
Hirschl RB, Kabre R, Kohler JE, Lal DR, Landman
MP, Leys CM, Mak GZ, Sato TT, Scannell M, Sujka
JA, Minneci PC, Deans KJ (2019 Jul) Factors associ-
ated with management of pediatric ovarian neoplasms.
Pediatrics 144(1):e20182537
Levin TL (2018) Pediatric imaging: radiation exposure
and how we image. Pediatr Rev 39(1):50–52
Liu H, Wang X, Lu D, Liu Z, Shi G (2013) Ovarian
masses in children and adolescents in China: analy-
sis of 203 cases. J Ovarian Res 6:47. Published 2013
Jul 4
Morotti M, Manada M, Gillott D, Venturini PL, Ferrero
S (2012) The preoperative diagnosis of borderline
ovarian tumors: a review of current literature. Arch
Gynecol Obstet 285:1103–1112
Oltmann SC, Garcia N, Barber R, Huang R, Hicks B,
Fischer A (2010) Can we preoperatively risk strat-
ify ovarian masses for malignancy? J Pediatr Surg
45(1):130–134
Papic JC, Finnell SM, Slaven JE, Billmire DF, Rescorla
FJ, Leys CM (2014a) Predictors of ovarian malig-
nancy in children: overcoming clinical barriers of
ovarian preservation. J Pediatr Surg 49(1):144–148
Papic JC, Billmire DF, Rescorla FJ, Finnell SM, Leys
CM (2014b) Management of neonatal ovarian cysts
and its effect on ovarian preservation. J Pediatr Surg
49(6):990–994
Péroux E, Franchi-Abella S, Sainte-Croix D et al (2015)
Ovarian tumors in children and adolescents: a series of
41 cases. Diagn Interv Imaging 96(3):273–282
Polis RL, Fallat ME, Wiesenauer C (2017) Chapter 95:
Neonatal ovarian masses. In: Puri P (ed) Newborn
surgery, 4th edn. CRC Press, Boca Raton, FL,
pp 1029–1037
Prat J, FIGO Committee on Gynecologic Oncology
(2015) FIGO’s staging classification for cancer of the
ovary, fallopian tube, and peritoneum: abridged repub-
lication. J Gynecol Oncol 26(2):87–89
Qazi SH, Jeelani SM, Dogar SA, Das JK, Saxena AK
(2020) Approaches to the management of pediatric
53  Ovarian Tumors
ovarian masses in the 21st century: systematic review
and meta-analysis. J Pediatr Surg 55(3):357–368
Renaud EJ, Sømme S, Islam S et al (2019) Ovarian masses
in the child and adolescent: an American Pediatric
Surgical Association outcomes and evidence-based
practice committee systematic review. J Pediatr Surg
54(3):369–377
Rescorla FJ (2012) Pediatric germ cell tumors. Semin
Pediatr Surg 21:51–60
Rogers EM, Casadiego Cubides G, Lacy J, Gerstle JT,
Kives S, Allen L (2014) Preoperative risk stratifica-
tion of adnexal masses: can we predict the optimal
surgical management? J Pediatr Adolesc Gynecol
27(3):125–128
Savic D, Stankovic ZB, Djukic M, Mikovic Z, Djuricic S
(2008) Torsion of malignant ovarian tumors in child-
hood and adolescence. J Pediatr Endocrinol Metab
21(11):1073–1078
Shah RU, Lawrence C, Fickenscher KA, Shao L, Lowe
LH (2011) Imaging of pediatric pelvic neoplasms.
Radiol Clin N Am 49(4):729–48, vi
Shanbhogue AK, Shanbhogue DK, Prasad SR, Surabhi
VR, Fasih N, Menias CO (2010) Clinical syndromes
associated with ovarian neoplasms: a comprehensive
review. Radiographics 30(4):903–919. https://doi.
org/10.1148/rg.304095745
Slovis TL (2003 Oct) Children, computed tomography radi-
ation dose, and the As Low As Reasonably Achievable
(ALARA) concept. Pediatrics 112(4):971–972
Spinelli C, Pucci V, Buti I et al (2012) The role of tumor
markers in the surgical approach of ovarian masses in
697
pediatric age: a 10-year study and a literature review.
Ann Surg Oncol 19(6):1766–1773
Stankovic ZB, Bjelica A, Djukic MK, Savic D (2010)
Value of ultrasonographic detection of normal ovarian
tissue in the differential diagnosis of adnexal masses
in pediatric patients. Ultrasound Obstet Gynecol
36(1):88–92
Timmerman D, Testa AC, Bourne T et al (2008) Simple
ultrasound-based rules for the diagnosis of ovarian
cancer. Ultrasound Obstet Gynecol 31(6):681–690
Tsai JY, Saigo PE, Brown C, La Quaglia MP (2001)
Diagnosis, pathology, staging, treatment, and outcome
of epithelial ovarian neoplasia in patients age < 21
years. Cancer 91(11):2065–2070
Van Calster B, Timmerman D, Bourne T et  al (2007)
Discrimination between benign and malignant
adnexal masses by specialist ultrasound examina-
tion versus serum CA-125. J Natl Cancer Inst 99(22):
1706–1714
Van Heerden J, Tjalma WA (2019 Dec) The multidis-
ciplinary approach to ovarian tumours in children
and adolescents. Eur J Obstet Gynecol Reprod Biol
243:103–110
Vitale SG, Capriglione S, Zito G et  al (2019)
Management of endometrial, ovarian and cervical
cancer in the elderly: current approach to a challeng-
ing condition. Arch Gynecol Obstet 299(2):299–315
Von Allmen D, Fallat ME (2019) Ovarian tumors. In: Puri
P, Höllwarth M (eds) Pediatric surgery. Springer sur-
gery atlas series. Springer, Heidelberg
 Testicular Tumors
54
Amanda F. Saltzman and Jonathan Ross

54.1 Introduction 54.2 Historical Overview,

Testicular cancer has a bimodal distribution dur-
ing childhood, first a small peak in infancy and
then a much larger peak in adolescence/young
adulthood. Historically, testicular tumors were
approached with radical orchiectomy, regardless
of age. Recent insights into histology and out-
comes over the past 20 years have challenged this
dogma, revealing that the relationship of tumor
development and puberty is crucial. Prepubertal
testicular masses generally represent benign
pathology and can often be managed with testis-­
sparing surgery, a significant deviation from his-
torical norms.
Special permission was obtained from the original author
of this chapter, Dr. Jonathan Ross.
A. F. Saltzman (*)
Department of Urology, University of Kentucky,
Lexington, KY, USA
e-mail: Amanda.saltzman@uky.edu
J. Ross
Department of Urology, Rush University, Chicago,
IL, USA
e-mail: jonathan_ross@rush.edu
Incidence and
Etiopathogenesis
Testicular tumors account for 1% to 2% of pedi-
atric solid tumors. Testicular tumors can gener-
ally be divided into 2 groups based on the cells or
origin, i.e. stromal or germ cell. Stromal tumors
are generally rare but more common in prepuber-
tal patients, but the incidence of germ cell tumors
(GCTs) increases with age. Malignant GCTs are
the most common solid tumor (15%) in adoles-
cents and young adults (AYA). Oncologic out-
comes for patients with all testis tumors, including
malignant GCTs are generally excellent, such
that research priorities have shifted to reducing
late effects of treatment while maintaining excel-
lent survival rates. However, AYA males with tes-
tis cancer have worse survival than their younger
pediatric and older adult counterparts, making
this a particularly vulnerable population.
54.3 Pathophysiology &
Pathology
The most important aspect to consider when
approaching a testicular tumor is the patient’s
pubertal status, using age 11 years as a surrogate.
Critical histologic changes appear to occur at
puberty, so tumors occurring in post-pubertal

© Springer Nature Switzerland AG 2023 699

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_54
700 A. F. Saltzman and J. Ross

Table 54.1  Characteristics of testis tumors in patients intratesticular from extratesticular scrotal masses.
based on pubertal status
The US characteristics of various testis tumors
Prepubertal Post-Pubertal have been described, though imaging alone can-
Malignant 70–75% 75% malignant
not determine a histologic diagnosis. The one
potential benign
Predominant Pure YST Mixed NSGCT exception is an epidermoid cyst, which while
type of benign and uncommon, has a classic pattern of
malignancy alternating hyper and hypoechoic concentric
Metastatic 5% at 20–30% at rings referred to as “onion skinning.” Despite a
potential presentation presentation
strong suspicion for this preoperatively, these
Teratoma Usually pure, Usually part of mixed
common, tumor, treated as lesions should still be excised to confirm pathol-
benign malignancy due to ogy, but TSS is the preferred approach, regardless
behavior potential for of age.
malignant With respect to tumor markers, αFP is most
degeneration
Typical TSS Radical orchiectomy
useful for prepubertal patients while βhCG and
management LDH are also important in post-pubertal patients.
Serum αFP is elevated in 90% of children with
YST, either pure or in combination with other
adolescent males are best managed with adult histologies and it is never elevated in pure semi-
algorithms. noma. While αFP is very specific for YST, one
A brief comparison of differences based on caveat is that it is physiologically elevated at
pubertal status is included in Table  54.1. The birth eventually decreasing to normal levels by
majority of testis tumors in prepubertal patients 8–12 mos of age. Important to note is the gesta-
are benign, most likely mature teratoma. While tional age of the child; a premature infant may
this is universally benign before puberty, it have a serum αFP that is 8–10 times higher than
becomes malignant after. The most common pre- a term infant, emphasizing that gestational age as
pubertal malignant testicular tumor is a yolk sac much as chronologic age needs to be considered.
tumor (YST). Post-pubertal testis tumors are usu- As such, elevated αFP in an infant does not nec-
ally mixed non-seminomas germ cell tumor essarily reflect the presence of a malignancy and
(NSGCT) and these behave more aggressively, obtaining repeat levels after a 5-day interval can
like their adult counterparts. Generally, radical be helpful with stable or increasing values raising
orchiectomy is the preferred approach for these concern for malignancy.
patients, but there are emerging data supporting In general, a boy >12 mos of age with a tes-
the use of testis-sparing surgery (TSS). ticular mass and an elevated αFP should be pre-
sumed to have YST and should undergo radical
orchiectomy. Conversely, a prepubertal patient
54.4 Diagnosis with a testis tumor and normal αFP almost cer-
tainly has a benign etiology and can be man-
Most testicular tumors present as a painless, firm aged with TSS. Infants <12 mos of age with an
mass; though rarely, they may present as testicu- elevated αFP present a difficult scenario as they
lar asymmetry, swelling, hydrocele, trauma or may have benign tumors, though αFP levels are
pain related to an acute bleed. On physical exam- rarely >100 ng/mL in males >6 mos of age with
ination, a mass that cannot be separated from the benign pathology. Therefore, TSS may be con-
testis is assumed to be a testis tumor, until proven sidered in patients 6–12 mos of age with an
otherwise. Scrotal ultrasound (US) is very help- AFP <100 ng/ml).
ful in making this distinction. Computed tomography (CT) traditionally
Preoperative evaluation of all males with a adds little value to the initial diagnosis of testicu-
suspected testis tumor includes an ultrasound lar masses though is imperative for staging. MRI
scan and measurement of tumor markers (αFP, is generally not used for testicular cancer staging,
βhCG, and LDH). US is helpful in distinguishing unless there are contraindications for CT scan.
54  Testicular Tumors
54.5 Differential Diagnosis
Scrotal swelling can be any number of things,
including testicular cancer, epididymo-orchi-
tis, paratesticular mass (rhabdomyosarcoma,
adenomatoid tumor, etc.), inguinal hernia,
­ Either a caudal or ilioinguinal block is given. An
hydrocele, etc. The most important first key
step is to determine the etiology of the swell-
ing, namely the testis itself, or extratesticular.
Once a testicular mass has been confirmed with
scrotal US, the differential is narrowed, with
malignancy being the most common and worri-
some option.
• YST
• Teratoma
• Seminoma
• Mixed NSGCT
• Granulosa cell tumor
• Leydig cell tumor
• Epidermoid cyst
• Other benign etiology
54.6 Management: Surgical
Approach and Technique
Pathology must be obtained for a testis mass for
both diagnostic, staging and therapeutic indica-
tions. In many cases, radical orchiectomy alone is
curative. Avoidance of the scrotal approach for
any reason is imperative, as this changes the typi-
cally reliable lymphatic spread of testis tumors to
the retroperitoneum. In the event a malignancy is
discovered during a scrotal surgery, the orchiec-
tomy can be completed scrotally without altering
stage provided that the entire testis and mass are
removed and there is no local spillage or spread.
Generally, a counter inguinal incision should be
made at the same operation to ensure an adequate
length of cord is removed.
If, based on the αFP level, a tumor is felt to be
malignant, then an inguinal orchiectomy should
be performed. Tumors that are likely to be benign
based on αFP level should be managed initially
with TSS (details below). Even tumors that
appear on US to replace the entire testis may be
701
enucleated as these tumors usually compress, not
destroy or replace surrounding tissue, leaving
significant residual testicular tissue important for
future hormonal and reproductive function.
The patient is placed in the supine position.
inguinal skin incision is made, extending later-
ally from just above the pubic tubercle following
Langer’s lines, toward the ipsilateral anterior
superior iliac spine. The subcutaneous tissue is
divided with cautery. The superficial epigastric
vein is often encountered in the lateral corner of
the incision and should be controlled with
cautery.
The superficial fat may then be separated
bluntly with retractors, exposing Scarpa’s fascia.
Scarpa’s fascia is grasped with forceps and
opened. The tissue between Scarpa’s fascia and
the external oblique aponeurosis is developed by
gentle spreading, and Scarpa’s fascia is then
divided throughout the length of the incision
exposing the external oblique fascia.
The spermatic cord is identified, freed cir-
cumferentially, and isolated. The testis is deliv-
ered up into the operative field with gentle
pressure. If the tumor is too large to fit through
the incision, it is critical to enlarge the incision
to avoid tumor spillage during delivery. The
medial aspect may be extended caudally down
the ipsilateral scrotum creating a “hockey stick”
incision (Fig. 54.1). The gubernaculum is care-
fully divided with cautery, avoiding injury to the
Fig. 54.1  Caudal extension of the incision (dotted line)
down the ipsilateral scrotum to accommodate a large mass
702
underlying scrotum (Fig. 54.2). It is critical that
the surgeon evert the scrotal skin to identify this
plane and avoid scrotal violation. Gentle traction
on the testicle away from the scrotal skin also
helps. Once the testis and tumor are delivered
through the incision, the field may be draped per
surgeon preference.
Fig. 54.2  Delivering the testis and dividing the guber-
naculum. Note the traction cranially of the testis and cau-
dally of the everted scrotal skin to avoid scrotal violation
(photo from A. Saltzman)
Fig. 54.3  Opening the external oblique aponeurosis sharply (figure used in prior edition, photo from A. Saltzman)
A. F. Saltzman and J. Ross
54.7 Radical Orchiectomy
If the preoperative assessment suggests a malig-
nancy, high control of the spermatic cord is para-
mount. The external oblique aponeurosis is
opened, either near the lateral corner of the inci-
sion so that it comes through above the internal
ring (Fig. 54.3), where it is unlikely to injure the
ilioinguinal nerve, or from the external ring. Care
should be taken to identify and preserve the ilio-
inguinal nerve.
The cremasteric fibers are dissected away
from the testicular vasculature and vas deferens
distally by entering the cremasteric sheath. The
cremasteric fibers give the inguinal floor strength
and thus dissection within the cremasteric sheath
eliminates the need for routine inguinal canal
floor repair. The spermatic cord is further traced
proximally to the internal ring, ideally seeing
peritoneal sac/fat and medial deviation of the vas
from the testicular vessels (Fig.  54.4). This
ensures complete cord resection and facilitates
future management at RPLND if necessary. If
used, early tourniquet control of the spermatic
cord should not compromise proximal ligation of
the cord and testis, and thus should be adjusted.
The vessels and vas are isolated separately to
facilitate vascular control and each is doubly
54  Testicular Tumors
Fig. 54.4  High ligation of the spermatic cord (photo
from A. Saltzman)
clamped. The specimen, with tunica vaginalis left
intact, is amputated at the internal ring and passed
off the field for pathological analysis.
Each of the stumps is controlled for hemosta-
sis. This can be accomplished by a variety of
ways, but a permanent material on each of the
stumps, in addition to a metal clip(s) on the vas-
cular stump is generally recommended. This
allows for easier identification of the spermatic
cord stump to ensure complete removal at
RPLND, should this be necessary. The clips help
with radiographic identification of the distal cord
stump as well. The cord stumps should be tucked
into the retroperitoneum as high as possible.
The incision is then closed in layers with
absorbable suture.
54.8 TSS
If the preoperative assessment suggests a benign
diagnosis, an attempt at TSS should be made.
Except in very specific scenarios detailed below,
tumor markers must be normal to consider
TSS.  Large tumors which appear to occupy the
entire testicle on US can be enucleated as they do
not replace or destroy the normal testicular tis-
703
sue, but merely compress it. Additionally, TSS
should not be performed without real-time intra-­
operative frozen section analysis. This has been
shown to be highly correlated with final pathol-
ogy and guides intra-operative management.
The patient is approached the same way, using
an inguinal incision, up to delivery of the testis.
The cord has already been circumferentially iso-
lated with a loose tourniquet and the testis is up
through the incision, tunica vaginalis intact, with
the field draped in case of the need to convert to
radical orchiectomy.
The tunica vaginalis is opened. If the mass is
not easily palpable or identified, intra-operative
US may be used for localization. Once the mass
is identified, the tourniquet is tightened for
hemostasis. An incision is made through the
tunica albuginea to, but not through, the tumor,
or along the periphery of the tumor if it is abut-
ting the tunica. The tumor is then excised com-
pletely, usually using both blunt and sharp
dissection (Fig.  54.5). There is usually a clear
plane between the tumor and testicular tissue.
Incisional biopsies are not appropriate as the
violation of the tumor capsule results in an up-
staging of the tumor, should it be malignant. The
tumor should then be sent for frozen section
analysis.
Once the tumor has been excised, the tourni-
quet may be released to restore blood flow. While
awaiting the frozen section report, additional tes-
ticular tissue is biopsied to evaluate for germ cell
neoplasia in situ (GCNIS), as this requires dis-
cussion about future treatment, if present. The
testicle is closed with fine absorbable suture.
If the frozen section confirms a benign histol-
ogy, then the testicle is returned to the scrotum
with or without fixation. Fixation of the replaced
testicle is not necessary as there have not been
any reported cases of testicular torsion after par-
tial orchiectomy for tumors. If the frozen section
is indeterminate or suspicious for malignancy,
then the radical orchiectomy is completed in a
standard fashion described above, without any
changes to staging.
704
a b
Fig. 54.5  Opening the hydrocele sac (a), excising the tumour (b), and closing the testis (c) in removing a benign
tumour (figure used in prior edition)
If a teratoma is diagnosed and the child is near
pubertal age or pubertal status is unclear, the sur-
rounding parenchyma should be examined to
help make this determination. If the tubules are
immature, then the tumor may be treated as
benign, but if the tubules show evidence of matu-
ration, the tumor should be treated as potentially
malignant according to post-pubertal patients.
There is insufficient evidence to recommend
cold ischemia with the above approach. While
there may be changes to the remaining paren-
chyma during this incomplete warm ischemia,
this is brief without impacting oncologic out-
comes. In fact, animal models have demonstrated
worse long-term fertility and hormonal function
in rats treated with cold ischemia over warm,
although these times exceeded 30 minutes.
A. F. Saltzman and J. Ross
c
54.9 Retroperitoneal Lymph
Node Dissection (RPLND)
RPLND plays a very limited role in prepubertal
testis tumors but is an important treatment strat-
egy for some post-pubertal patients with malig-
nant GCTs. Unlike adults, only a minority of
prepubertal patients with metastases have metas-
tases limited to the retroperitoneum, and meta-
static disease is very responsive to chemotherapy.
Furthermore, the complication rate following
RPLND is significantly higher in children than in
adults. The only relative indications for retroperi-
toneal surgery in a prepubertal patient are to
biopsy an equivocal node or to excise a persistent
retroperitoneal mass following chemotherapy—a
rare occurrence.
54  Testicular Tumors
54.10 Complications
Surgical complications are rare following ingui-
nal orchiectomy or TSS. The occasional compli-
cations that can occur following inguinal
orchiectomy are scrotal infection or hematoma
and retroperitoneal hematoma. These complica-
tions generally can be avoided with meticulous
surgical technique and thorough hemostasis of
the raw surface of the inner scrotum.
Local recurrence following simple excision
has not been reported for benign tumors or fol-
lowing inguinal orchiectomy for malignancy.
Local recurrence in the scrotum has occurred fol-
lowing trans-scrotal biopsy of malignant tumors;
however, it appears that tumor spillage at the time
of surgery plays more of a role than scrotal
approach (though best to generally avoid both).
54.11 Special Situations
54.11.1 Bilateral Masses
Bilateral masses are rare and every effort should
be made to preserve normal testicular tissue for
both hormonal and reproductive function.
Workup and staging are the same as for a solitary
mass (US and tumor markers). TSS is the pre-
ferred approach, even in the face of elevated
tumor markers. In this scenario, it is critical to
sample surrounding testicular tissue for GCNIS
as its presence may impact subsequent therapy
(observation vs. prophylactic radiation vs. com-
pletion orchiectomy).
54.11.2 Solitary Testis
This is commonly encountered for metachronous
development of GCT in a male who has already
undergone radical orchiectomy for a personal
history of GCT. Workup and staging are the same
as all testis masses. Similar to bilateral masses,
every effort should be made to preserve normal
testicular tissue so TSS is preferred, with sur-
rounding tissue sampling to determine possible
705
adjuvant therapy (observation vs. prophylactic
radiation vs. completion orchiectomy).
54.11.3 Small Testicular Masses
There are situations where TSS may be consid-
ered in post-pubertal patients beyond bilateral
masses or a solitary testis as discussed above. In
adult men with testis masses <2 cm, 50–80% will
be benign, but this same size cutoff has not
proven to be valid in patients <18y old.
54.11.4 Microlithiasis
Patients with GCTs may have surrounding tes-
ticular microlithiasis, but it is more commonly
found incidentally. While stringent guidelines
have not been published, isolated microlithiasis
without a solid mass or risk factors for GCT does
not confer an increased risk of GCT development
and does not require follow-up.
54.11.5 GCNIS Remaining after TSS
GCNIS portends an increased risk of GCT recur-
rence in both testes. It is recommended to discuss
surveillance vs. adjuvant therapy in the form of
further surgical treatment (completion radical
orchiectomy), chemotherapy (single cycle of
BEP), or radiation therapy (18–20  Gy). This
potential for future malignancy should be bal-
anced with the desire for future fertility and
Leydig cell (testosterone) function.
54.11.6 GCNIS/Gonadoblastoma/
GCT in Patient with a
Difference of Sex
Development
GCNIS (testicular) and gonadoblastoma (ovar-
ian) are the same histology that develops in a dif-
ferent background of gonadal pathology. These
are a premalignant condition, with a 60% chance
706
of progression to invasive GCTs. Generally,
patients with invasive GCTs are treated along the
same algorithm as patients without DSD who
have GCTs. There is a recent small series that
suggests that outcomes for those with DSD and
GCT are not as good as their non-DSD
counterparts.
54.11.7 Scrotal Violation
Trans-scrotal biopsy or orchiectomy is contrain-
dicated as it causes altered lymphatic drainage of
the testis and may be linked to increased spillage
which causes increased locoregional recurrence.
There are limited data available on this topic and
no clear guidelines exist. The most important
prognostic factor related to scrotal violation is
concomitant tumor spillage, such that without
tumor spillage or positive margins, further local
control with re-excision or adjuvant therapy is
unnecessary. Monitoring by physical exam is
important and the general consensus is an addi-
tional procedure to manage the scrotum alone
should be avoided, although it can be done with
RPLND if desired. Management of scrotal viola-
tion is conspicuously absent in guidelines and
should be the focus of further investigation.
54.11.8 Stromal Tumors
These tumors originate from Leydig and granu-
losa cells that support germ cells. These are rare
and usually present incidentally or with signs of
precocious puberty. These tumors are uniformly
benign. Juvenile granulosa cell tumor is the most
common testis mass in male infants <6mos of age
with normal tumor markers, while Leydig cell
tumors most commonly present with precocious
puberty because of testosterone production.
Unfortunately, the only sign of malignancy is
spread of disease, usually along the same pattern
as GCTs. Usually a single postoperative staging
scan is reasonable. It is highly encouraged to
enroll these patients onto the International
Ovarian & Testicular Stromal Tumor Registry
(https://www.otstregistry.org).
A. F. Saltzman and J. Ross
54.12 Outcome and Follow-up
The survival of children with testis tumors
depends on the histology of the tumor and the
presence or absence of metastatic disease.
Patients with epidermoid cysts and stromal
tumors and prepubertal patients with mature tera-
toma do well with excision alone and no long-­
term follow-up is required.
Patients with malignant tumors require a met-
astatic evaluation (CT of the chest, abdomen, and
pelvis) and serum tumor markers need to be mon-
itored post-operatively to complete staging. Any
tumor marker elevation needs to be followed at
least 5 half lives to ensure appropriate and com-
plete return to normal levels.
There are numerous cooperative groups with
guidelines on staging, management, and treat-
ment of malignant testis tumors. While each
group has some nuanced differences, it is critical
to know that post-pubertal patients should be
treated per adult algorithms to maximize onco-
logic outcomes. Post-pubertal children treated
per pediatric protocols have worse disease spe-
cific survival than their adult counterparts treated
per adult algorithms. It is imperative that the
treating surgeon/urologist and oncologist are on
the same page.
Approximately 80% of patients will have
stage 1 disease (limited to the testicle) confirmed
by a negative radiographic evaluation and nor-
malization of tumor markers. These patients are
recommended to undergo active surveillance
alone, although adjuvant therapy options are
available (generally, radiation and chemotherapy
for those with seminoma and RPLND and che-
motherapy for those with NSGCT).
The specifics of follow-up differ based on the
protocol used, but generally monitor CT imaging
and tumor markers every 2–6 mos x2y and every
6 mos until 5y. With these protocols, there are
many CT scans and thus a significant exposure to
radiation, which is not without potential side
effects. Importantly, for prepubertal males, αFP
elevation almost always occurs with radiographic
recurrence, so consideration for limiting CT
imaging and closely following αFP may be rea-
sonable. In general, for stage 1 patients, 80% will
54  Testicular Tumors
not recur and thus will not require more treat-
ment, while about 20% will recur and require
treatment intensification. The cure rate after sal-
vage remains excellent (>90%).
There is a promising advance using serum
microRNA-371 as a tumor marker for testis can-
cer. This appears to detect recurrence of disease
earlier than CT imaging and traditional serum
tumor markers, and is being studied for some
other specific scenarios, but is not clinically used
quite yet.
Patients who present with metastatic disease
are treated upfront with adjuvant treatment, usu-
ally chemotherapy, and survival is similarly
excellent for this group as well. Radiation plays
no role in the primary treatment of testis tumors
other than pure seminoma, which is rarely seen in
the pediatric population. Residual masses after
chemotherapy should be managed by RPLND for
those who are post-pubertal. While this is not an
uncommon scenario with post-pubertal patients,
it is very rare in prepubertal boys. The limited
available data suggest that prepubertally, residual
post-chemotherapy disease is usually limited to
the mass alone and not throughout the retroperi-
toneum. Thus, prepubertal patients with a post-­
chemotherapy residual mass should be considered
for excision rather than formal RPLND.
54.13 Conclusion
Diagnosis of testicular tumors involves imaging,
serum tumor markers, and surgery to determine
pathology. Decisions on surgical approach,
namely TSS vs. radical orchiectomy, are deter-
mined by the likelihood of malignancy. Several
preoperative factors are involved with this deci-
sion, namely pubertal status and serum tumor
markers. Post-pubertal patients should be
assumed to have a malignancy and undergo radi-
cal orchiectomy, while prepubertal patients with
normal serum tumor markers likely have a
benign tumor and an attempt at TSS is reason-
able. Outcomes for testicular cancer are excel-
707
lent, often curative with surgery alone. Even
those with widespread disease can achieve excel-
lent outcomes following a multi-disciplinary,
protocol-­driven approach appropriate for age.
References
Caldwell BT, Saltzman AF, Maccini MA et  al (2019)
Appropriateness for testis-sparing surgery based on
the testicular tumor size in a pediatric and adolescent
population. J Pediatr Urol 15(70):e1–70.e6
Dicken BJ, Billmire DF, Krailo M et al (2017) Gonadal
dysgenesis is associated with worse outcomes in
patients with ovarian nondysgerminomatous tumors: a
report of the Children’s Oncology Group AGCT 0132
study. Pediatr Blood Cancer 65:e26913
Dieckmann K-P, Radtke A, Geczi L, Matthies C, Anheuser
P, Eckardt U et al (2019) Serum levels of MicroRNA-­
371a-­3p (M371 test) as a new biomarker of testicular
germ cell tumors: results of a prospective multicen-
tric study. J Clin Oncol 37:1412–1423. https://doi.
org/10.1200/JCO.18.01480
Frazier L, Hale J, Rodriguez-Galindo C et  al (2015)
Revised risk classification for pediatric extracranial
germ cell tumors based on 25 years of clinical trial
data from the United Kingdom and United States. J
Clin Oncol 33(2):195–201
Gentile G, Rizzo M, Bianchi L et al (2020) Testis sparing
surgery of small testicular masses: retrospective analy-
sis of a multicenter cohort. J Urol 203(4):760–766
Nappi L, Thi M, Lum A, Huntsman D, Eigl BJ, Martin
C et al (2019) Developing a highly specific biomarker
for germ cell malignancies: plasma miR371 expression
across the germ cell malignancy spectrum. J Clin Oncol
37:3090–3098. https://doi.org/10.1200/JCO.18.02057
NCCN Clinical practice guidelines in oncology - testicu-
lar cancer. NCCN, Jenkintown, PA
NCCN Clinical practice guidelines in oncology – adoles-
cent and young adult oncology. NCCN, Jenkintown,
PA
Ross JH, Rybicki L, Kay R (2002) Clinical behavior and
contemporary management algorithm for prepubertal
testis tumors: summary of the prepubertal testis tumor
registry. J Urol 168:1675–1679
Spiekermann M, Belge G, Winter N, Ikogho R, Balks T,
Bullerdiek J et al (2015) MicroRNA miR-371a-3p in
serum of patients with germ cell tumours: evaluations
for establishing a serum biomarker. Andrology 3:78–
84. https://doi.org/10.1111/j.2047-­2927.2014.00269.x
Stephenson A, Eggener SE, Bass EB et  al (2019)
Diagnosis and treatment of early stage testicular can-
cer guideline: AUA guideline. J Urol 202:272–281
Woo L, Ross J (2019) Testicular tumors in children and
adolescents. AUA Update Series 38:134
 Part VIII
Gastrointestinal
 Esophageal Atresia
and Tracheoesophageal Fistula
55
Michael E. Höllwarth and Paola Zaupa

55.1 Introduction 55.2 History

The term congenital atresia of the esophagus The first description dates back to 1670, when
describes a large group of variant malformations Durston (1670) described thoracopagus twins
that share a defect of the esophageal continuity, with a blind upper end of the esophagus in one of
with or without a fistula, to the trachea or to the them. In his book, Gibson (1697) described “The
bronchi. It is one of the most life-threatening Anatomy of Human Bodies Epitomized” the first
anomalies in a newborn baby and the quality of case of EA with a distal trachea-­esophageal fis-
survival depends on early diagnosis and appropri- tula (TEF) (LIT). In 1861, Harald Hirschsprung
ate therapy. The history of surgical therapy of published a series of babies with EA and trachea-
babies with esophageal atresia (EA) after esophageal fistula. The first surgical attempts to
Cameron Haight is a story of success, started out correct the malformation were performed in the
with survival rates around 50%, but reaching late nineteenth century, with many more in the
nearly 100% today, when associated life-­ twentieth century (excellent overview of the his-
threatening malformations are excluded. The tory by Mazurak and Patkowski 2018).
mainstays of this success are appropriate diagno- The first classification of EA was made in 1929
sis and preoperative therapy, reconstruction of the by Vogt, a radiologist from Boston (Muensterer
esophageal continuity with closure of an existing and Berdon 2015). Dietrich Kluth, a student of
fistula, or esophageal replacement if necessary. Rehbein, published a dissertation monography,
which described variations of EA and its different
classes and subtypes (Kluth 1977).
The first successful surgical repair by an extra-­
pleural approach with ligation of the fistula and
primary anastomosis was performed on a 12-day-­
old baby by Cameron Height in 1941. Now an
adult, this patient gave birth to a newborn who
M. E. Höllwarth (*)
University Clinic of Paediatric and Adolescent also suffered from EA, which was successfully
Surgery, Medical University of Graz, Graz, Austria operated on in the same institution by Arnold
e-mail: michael.hoellwarth@medunigraz.at Coran. In October 1951, Fritz Rehbein success-
P. Zaupa fully operated on the first EA on the fifth day
University Clinic of Pediatric and Adolescent after birth in German-speaking countries. The
Surgery, Medical University of Graz, Graz, Austria procedure was performed using local anesthesia,
e-mail: paola.zaupa@medunigraz.at

© Springer Nature Switzerland AG 2023 711

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_55
712
under ether narcosis from a paravertebral inci-
sion. There was a right descending aorta. The
anastomosis was made according to Haight in the
telescope technique. The postoperative time
course was characterized by many complications,
including a recurrent fistula and anastomotic ste-
nosis. However, the follow-up control of the now
young man at the age of 16  years showed an
excellent physical and psychological develop-
ment, with minimal swallowing problems (per-
sonal information by Wolfgang Lambrecht).
55.3 Incidence
The incidence of an EA, with or without a fistula,
is 2.44 in 10,000 births, ranging from 1.77 to 3.68
according to national and international databases
for congenital anomalies, and among them are
68.6% with fistula and 25.5% without fistula
(Nassar et al. 2012). The mean incidence from a
Swedish national registry between 1987 and 2007
was 3.3 in 10,000 live births, with little or no sig-
nificant changes over time (Oddsberg et al. 2012).
According to a French national registry, the preva-
lence in 2008 and 2009 was 1.97/10,000 live
births. There is a 1.3 slightly higher incidence in
males, which is not significant (Sfeir et al. 2013).
Most cases occur sporadically, and therefore,
the etiology is likely to be multifactorial. The high
number of associated anomalies points to a very
early disturbance of the developing embryo, prob-
ably to the time point when the segmentation of
the embryo occurs. A number of risk factors, such
as environmental teratogens, are discussed in the
literature, including exposure to thalidomide,
statins, alcohol, smoking, contraceptive pills, hor-
mones, as well as endocrine diseases of the mother,
such as diabetes or higher maternal age (Oddsberg
et al. 2010; Felix et al. 2009). Recent studies show
a 4.5 times higher risk after an assisted reproduc-
tive technology (Källén et al. 2010). Additionally,
there is ample evidence that the anomaly can be
genetically determined in some cases: chromo-
somal anomalies occur in 6–10% of all cases,
including Trisomy 13, 18, 21, and 22q11 deletion
(Felix et al. 2009; Reutter and Ludwig 2013).
The recurrence risk in a second child from par-
ents with one affected child is around 0.5–2.0%,
M. E. Höllwarth and P. Zaupa
and the risk for a newborn born to an affected par-
ent is around 3.0–4.0% (Solomon et al. 2012). A
large number of different syndromes have been
reported in association with EA. Mutation of the
NMYC gene (Feingold syndrome) has a 30–40%
association with EA or SOX2 (EAG syndrome)
(Brunner and Bokhoven 2005). Further gene dele-
tions or mutations are associated with EA, such as
MID1 (Opitz syndrome), GLI3 (Pallister-Hall
syndrome), CHD7 (CHARGE association—colo-
boma, heart disease, choanal atresia, retardation,
genital hypoplasia, and ear deformities with deaf-
ness), and X-linked Opitz syndrome. However,
since most EA cases occur sporadically, there is
most likely a heterogeneous and multifactorial
pathogenesis involving different or multiple genes
and signaling pathways.
55.4 Etiopathogenesis
The normal foregut embryology is still controver-
sial. Within the fourth week of gestation, the sepa-
ration of the esophagus and the trachea takes place
by the folding of the primitive foregut. The theories
regarding the cause of EA include malformation of
a septum dividing the foregut from the airways,
deviation of the septum in one or the other direction
resulting in esophageal or tracheal atresia, and fail-
ure of esophageal recanalization or vascularization.
Recently, impaired tracheal development with the
foregut developing into the trachea rather than the
esophagus has been proposed. Experimental results
of some authors lead to the hypothesis that the dis-
tal esophago-­tracheal fistula arises as a blind end-
ing duct from a trifurcation of the trachea and only
joins the stomach anlage later, but these findings
have not been confirmed by others. Several mouse
models have shown that the dorso-ventral forma-
tion of the foregut is crucial. Precise bilateral acti-
vation and inhibition of transcription factors, such
as Nkx2.1 and SOX2, result in the separation of the
esophagus from the trachea. The transcription fac-
tors are activated and/or repressed by different pro-
teins, such as the bone morphogenic protein
(BMP4) and WNT (von Lennep et al. 2019).
Significant insights have been provided by the
Adriamycin-induced rat model of EA (Mc
Laughlin et al. 2013; Hajduk et al. 2011, 2012).
55  Esophageal Atresia and Tracheoesophageal Fistula
There is strong evidence of a close relationship
between an abnormal notochord and disturbed
somatic segmentation, finally resulting in verte-
bral anomalies, cardiac malformation, and foregut
anomalies, such as EA. More recent studies have
also shown a major role of the Sonic hedgehog
(SHH) signaling pathway (Ioannides et al. 2010;
Busa et al. 2016). It seems to be obvious that the
SHH gene and the signaling glycoprotein are
involved in the normal morphogenesis of organ
systems such as notochord, vertebra, and differen-
tiation of the trachea and esophagus (Litingtung
et  al. 1998). From these experiments, one must
conclude that there is either a primary gene-
related defect or an exogenous pathogenic insult
that occurs within the first 10 days of pregnancy,
causing notochord dysfunction and leading to the
manifestation of anomalies, such as EA or ano-
rectal atresia, renal malformations, and others.
55.5 Pathophysiology
Due to the early embryogenesis, it is only logical
that babies with EA suffer from a high number of
associated malformations, in the range of 50–80%
(Sfeir et al. 2013; Oddsberg et al. 2010; Pedersen
et al. 2012; Friedmacher et al. 2017). For a suc-
cessful treatment strategy, it is important to con-
sider a detailed diagnostic workup that has a
significant impact on the treatment strategy. The
most frequent associated anomalies are musculo-
skeletal malformations (20–70%), followed by
cardiovascular (29%), genitourinary (16%), gas-
trointestinal (16–20%), and chromosomal anoma-
lies (5%) (Bogs et  al. 2017). The wide range of
percentages given in the literature comes from
differences in the diagnostic workup. A careful
X-ray of the whole vertebral spine, counting the
ribs and vertebra in the different segments, will
show associated skeletal malformations and/or
numerical variations in up to 70% of EA patients.
The most common cardiac anomaly is the ven-
tricular septal defect (19%), which is associated
with an up to 16% mortality rate. Other common
anomalies include atrial septal defect (20%), tetral-
ogy of Fallot (5%), and coarctation (1%) or right-
descending aorta (4%). It is important to realize
713
that some of these cardiac defects lead to a clini-
cally evident heart insufficiency only a few days
after delivery. Thus, all EA patients should have an
early echocardiography, as well as ultrasound
exams of the renal tract and the brain. The most
common gastrointestinal-associated anomaly is
anorectal atresia (9%), followed by duodenal atre-
sia (5%), malrotation (4%), and other intestinal
atresias (1%). Further associated malformations
may involve nearly all organ systems, leading to
omphalocele, neural tube defects, diaphragmatic
hernia, and other anomalies. As mentioned above,
an association with at least 18 different syndromes
are described in up to 10% of patients including
Holt-Oram syndrome; DiGeorge syndrome;
Goldenhair syndrome; trisomy 13, 18, and 21; the
CHARGE association (coloboma, heart defect,
atresia of the choana, retardation, genital hypopla-
sia, and ear deformities); and many others.
The pattern of common associated anomalies
led to the creation of the acronym “VATER asso-
ciation” (vertebral anomaly, anal atresia, tracheo-
esophageal fistula, EA, and renal anomaly) or
“VACTERL association” (additionally, cardiac
anomaly and limb malformation) (Stoll et  al.
2017; Bartels et  al. 2012; Solomon 2012). The
incidence of these associations is around 20% in
the EA population, but two or more anomalies
occurring in nearly half of the patients.
55.6 Pathology
Classification is usually based on their orientation
on occurrence and type of tracheoesophageal fis-
tula. The commonly used systems are those
described by Vogt (1929) and Gross (1953). Vogt’s
extremely rare type 1 characterized by a more or
less total lack of the esophagus is not included in
Gross’ classification. An isolated tracheoesopha-
geal fistula (H-type fistula) is classified as type 4 or
E, although it may belong to a different spectrum
because the esophagus is patent. In Gross’ classifi-
cation, congenital esophageal stenosis constitutes
type F (Table 55.1, Fig. 55.1). A complete list of
all published variations of EA is summarized in
the dissertation work of Kluth in 1977.
714 M. E. Höllwarth and P. Zaupa

Table 55.1  Anatomical classification In addition to the anatomical classification,

VOGT GROSS there are risk classifications based on birth
types types % weight, cardiac anomalies, and pneumonia,
Absent esophagus 1 – –
which allows the comparison of results across
EA without fistula 2 A 8.5
different institutions. Associated cardiac malfor-
EA with fistula
   Proximal 3a B 1.0 mations still have a significant influence on the
   Distal 3b C 85.0 survival rate of patients. The best-known classi-
  Proximal and 3c D 1.5 fication is named after Waterston from 1962;
distal however, a classification adapted to the recent
H-type fistula without 4 E 4.0 progress in neonatal surgery and medicine has
atresia
been published by Spitz in 1994 (Table 55.2). In

2/A 3b/C
3a/B

3c/D 4/E

Fig. 55.1  The most common different forms of EA

55  Esophageal Atresia and Tracheoesophageal Fistula 715

Table 55.2  Risk classification

Waterstone
et al. (1962)
survival (%)
Birth weight > 2.500 g otherwise A 100
healthy
Birth weight 2.000–2.500 g or higher B 85
mild pneumonia or moderate cardiac
anomalies
Birth weight < 2.000 g or higher, severe C 65
pneumonia or severe associated cardiac
anomalies
Spitz et
al. Survival
(1994) (%)
Birth weight > 1.500 g otherwise I 97
healthy
Birth weight < 1.500 g or major II 59
cardiac anomaly
Birth weight < 1.500 g major III 22
cardiac anomaly

this classification, only birth weight and cardiac

anomalies are considered predicting factors for
the prognosis.

55.7 Diagnosis

55.7.1 Clinical Features
Prenatally, the earliest symptom of EA is a poly-
hydramnion in the second half of pregnancy. It is
an unspecific manifestation of swallowing disor-
ders or of a disturbed passage of fluid through the
uppermost part of the intestinal tract of the fetus.
Prenatal ultrasound may further reveal forward
and backward shifting of fluid in the upper pouch
and, in cases without a lower fistula, a paucity of
fluid in the stomach and small intestine. Recently,
fetal magnetic resonance imaging (MRI) has
gained more attention for the prenatal diagnosis
of congenital anomalies.
Postnatal presentation is characterized by
the typical drooling of saliva, choking, cough-
ing, and cyanotic attacks. Any feeding trial is
contraindicated with these symptoms because it
causes early aspiration and pneumonia. The
next diagnostic step is to pass a 12 F (firm and
Fig. 55.2  In this case, a very small feeding tube was used
to diagnose EA.  It resulted in a curling up in the upper
esophageal pouch mistakenly leading to the diagnosis of a
normal esophagus. Therefore, a rather firm and thicker
tube has to be used for this procedure
X-ray visible) feeding tube into the stomach. If
this is not successful, EA is almost certain.
However, small tubes must be avoided because
they may curl up in the upper pouch, thereby
giving the illusion that they have been pushed
forward into the stomach (Fig.  55.2). Very
rarely, a small tube may pass through the tra-
chea and fistula into the stomach; thus an EA is
erroneously excluded. If symptoms persist, a
contrast radiograph should be performed dem-
onstrating the atresia or even a small laryngotra-
cheal cleft. If an EA is suspected, a physical
examination of the entire body is performed to
detect further associated malformations.
716
55.7.2 Radiological Diagnosis
The next step is to perform a plain X-ray includ-
ing the neck, the thorax, and the abdomen. The
approximate length of the upper pouch can be
estimated by the length of the X-ray visible tube
in it. Air below the diaphragm can be seen in the
presence of a lower tracheoesophageal fistula, and
additional fluid levels indicate a duodenal or
intestinal atresia. A gasless abdomen indicates a
pure EA without a lower fistula (Fig. 55.3). A long
distance between the segments is to be expected,
but a tiny or secondary occluded fistula may be
present extremely rarely. The translucency of the
lungs provides the first information on whether
aspiration pneumonia—either from the saliva or
from the refluxed gastric acid through the lower
Fig. 55.3  EA without lower esophageal fistula indicating
a long distance between esophageal segments
M. E. Höllwarth and P. Zaupa
fistula—is present. Some researchers inject 0.5–
1.0 ml of water-soluble contrast material (we see
no need for barium contrast material) through the
tube into the upper pouch, under additional fluo-
roscopic control, to detect a proximal tracheo-
esophageal fistula. Cardiologic assessment,
including echocardiography, forms a part of the
routine preoperative workup in order to recognize
associated congenital cardiac anomalies or the
presence of a right-­descending aorta, which is
important for the surgeon. Abdominal and uro-
genital ultrasound is performed routinely. Finally,
one should first look for whether or not structural
or numerical anomalies exist along the vertebral
column or the ribs. Chromosomal analysis is per-
formed only if the external aspect of the baby is
suspicious of a genetic defect.
The history of a patient with an H-type fistula
is different: there is no passage problem, but the
leading symptoms are recurrent coughing and
cyanotic attacks during feeding due to aspiration
through the fistula. Presentation is usually more
protracted and sometimes delayed beyond the
first year of life.
Diagnosis is made by esophagography with
water-soluble contrast material that shows the
spillage of parts of the contrast material through
the fistula into the trachea. Tracheoscopy con-
firms the diagnosis.
55.7.3 Differential Diagnosis
EA symptoms in the newborn are quite character-
istic and a probe with the appropriate feeding
tube allows the diagnosis. However, some very
rare diseases may show similar clinical symp-
toms, such as a laryngo-tracheal-esophageal cleft
or a congenital esophageal stenosis. A trial with
contrast material is not useful because it may just
aggravate the aspiration of introduced material,
leading to pneumonia. Thus, in all unclear cases,
an immediate endoscopy is indicated with careful
inspection of the larynx and the trachea in order
to detect a cleft. In any other cases, the endoscope
is pushed forward into the esophagus to diagnose
or exclude a congenital stenosis.
55  Esophageal Atresia and Tracheoesophageal Fistula 717

55.8 Management close to the tracheal bifurcation and the infant is

ventilated manually with rather low inspiratory
55.8.1 Preoperative Management pressure and small tidal volumes.
Routine preoperative laryngo-tracheo-­
The babies are nursed in the ICU. Immediate sur- bronchoscopy using a rigid 3.5  mm endoscope
gery is rarely required; hence, all the above-­ should be standard (Taghavi and Stringer 2018).
mentioned investigations can be performed step by The trachea and the main bronchi are briefly
step. An oro- or naso-esophageal insertion of a inspected, and the fistula to the esophagus is
double-lumen Replogle tube is mandatory for con- localized, which is usually 5–7  mm above the
tinuous or intermittent aspiration of saliva in order carina. Exceptionally, it may be found at the
to prevent aspiration. The baby should be posi- carina or even in the right main bronchus indicat-
tioned upright to minimize gastroesophageal reflux ing a short lower segment, and most likely a long
into the trachea and lungs via the lower fistula. esophageal gap. The next step is to look for an
Intubation and ventilation is only necessary in upper fistula. The dorsal membranous region of
cases of respiratory distress, severe pneumonia, or the tracheal wall is inspected carefully up to the
severe associated malformations demanding respi- cricoid’s cartilage. Small upper fistulas can be
ratory therapy. In these cases, the endotracheal tube missed. To avoid this pitfall, irregularities of the
should, if possible, be positioned beyond the distal dorsal tracheal wall are gently probed with the tip
tracheoesophageal fistula to avoid insufflation of of a 3 F ureteric catheter passed through the bron-
gas into the stomach and intestine. If the latter choscope. If a fistula is present, the catheter will
problem cannot be avoided, due to a very low glide into it.
opening of the fistula into the carina or even right The goal of the surgical procedure is to
main bronchus, or due to a very large fistula, a gas- divide the fistula, to close it on the tracheal
trostomy with an underwater seal may ameliorate side, and to perform an end-to-end anastomosis
the problem. Gastrostomy and emergent surgical between the esophageal segments. The essen-
closure of the fistula by a thoracotomy is the best tial surgical part is to ensure a very careful and
strategy to avoid severe respiratory distress and is non-tissue traumatizing procedure. The stan-
also used in unstable and very premature babies. dard approach is through a right-sided extra-
Broad-­spectrum antibiotics, intravenous fluid ther- pleural latero-dorsal thoracotomy via the fourth
apy, and vitamin K analogue is administered before intercostal space. Muscle-sparing thoracotomy
surgery. If severe pneumonia exists, surgery has to between the latissimus dorsi and the anterior
be postponed until the lung recovers. In the case of serratus muscle reduces postoperative muscle
severe associated malformations (e.g., diaphrag- dysfunction and pain. If a right aortic arch is
matic hernia or cardiac malformation), a decision diagnosed preoperatively (20%), a left-sided
should first be made regarding the type of surgical approach is recommended unless there is a dou-
procedure to be performed, or whether they should ble aortic arch. However, if an unsuspected
be performed in one step. Again, a preliminary clo- right-descending aorta is encountered, the pro-
sure of the fistula saves time, but the anastomosis cedure can be continued in most cases, estab-
can be postponed if necessary. lishing the anastomosis to the right of the aortic
arch (Babu et al. 2000). While in earlier times
the procedure was performed extrapleurally by
55.8.2 Operative Management gently separating the parietal pleura from the
thoracic wall, recent progress in surgical tech-
55.8.2.1 EA with Distal niques and suture material, as well as sophisti-
Tracheoesophageal cated postoperative management, allows
Fistula (85%) operating transpleurally without any disadvan-
Surgical repair is performed under general anes- tage for the patient. Nevertheless, it is excellent
thesia with endotracheal intubation. As men- training for young colleagues to learn the subtle
tioned above, the endotracheal tube is advanced extrapleural approach.
718
The azygos vein is divided between ties and
the fibers of the vagal nerve are preserved as well
as possible. The fistula is closed near the trachea,
avoiding any narrowing of the airway. The anas-
tomosis of the esophagus is performed with 6/0
absorbable sutures. To facilitate the identification
of the upper pouch, the Replogle tube can be
pushed forward by the anesthetist. The upper
pouch mucosa often retracts and can be missed if
the surgeon does not take particular care with the
anastomosis. Once the posterior wall is sutured, a
5 F feeding tube is introduced into the stomach
with one end and back to the mouth with the
Mucosa
Fig. 55.4  The circular myotomy according to Livaditis
can lengthen the upper esophageal pouch by 0.5–1.0 cm
Fig. 55.5  If a very large upper esophageal pouch is present, a flap can be fashioned to bridge the distance to the lower
esophagus
M. E. Höllwarth and P. Zaupa
other end to allow early postoperative feeding.
Postoperatively, the feeding tube should pass
through the nose. In most cases with a distal fis-
tula, the goal of a tension-free anastomosis can
be achieved. If the tension appears to be too
much, despite extensive mobilization of the lower
esophagus and the upper pouch, further length
may be gained with a circular myotomy (Livaditis
et al. 1972; Lindahl 1987) (Fig. 55.4). However,
some authors have described the formation of
problematic pseudodiverticula after this proce-
dure (Otte et  al. 1984). Another way to reduce
inappropriate tension to the anastomosis, in some
cases, is to fashion a mucosa-muscular flap from
the upper esophageal pouch (Fig. 55.5). Extensive
mobilization of the lower esophagus can help to
achieve a safe anastomosis but it is nearly always
complicated by significant gastroesophageal
reflux problems. Schärli’s division of the lesser
curvature is rarely a useful procedure in these
cases and results in severe reflux problems
(Schärli 1992).
Some authors have published successful thora-
coscopic repair of the EA, which is necessarily
transpleural (Rothenberg and Flake 2015; Holcomb
55  Esophageal Atresia and Tracheoesophageal Fistula
et al. 2005). A recent meta-analysis including 48
articles from the literature showed no significant
differences between the two approaches in regard
to complications, anastomotic leaks or strictures,
or need for fundoplication nor the time to full
oral feeding. Only the operative time was signifi-
cantly longer with the endoscopic procedures
and conversion rate was around 10% (Yang et al.
2016; Way et al. 2019). Additionally, the thora-
coscopically performed procedure is certainly
very demanding and needs great experience in
endoscopic surgery. Thus, the carefully per-
formed open procedure is still the “gold stan-
dard” for most surgeons as it has been confirmed
by an International Survey showing that 178
contributing institutions preferred the open
(90%) and in 71% the extrapleural technique,
despite the fact that many other endoscopic pro-
cedures are performed in these institutions (Zani
et al. 2014).
At the end of the procedure, the thorax is
closed. The ribs should not be sutured together
because that causes rib fusion and leads to a long-­
term scoliosis. The use of a drain depends on the
preference of the surgeon, but it should be away
from the anastomosis to avoid any mechanical
complication.
55.8.2.2 EA with Proximal and Distal
Tracheoesophageal
Fistula (1.5%)
As mentioned above, a careful preoperative
endoscopic examination of the dorsal tracheal
wall, using a 3  F ureteric catheter, enables the
surgeon to detect an upper fistula that can be as
high as the cricoid cartilage. In these cases, it
might be better to close the fistula from the right
side of the neck (see H-type fistula). The catheter
can be left in the fistula to facilitate the
­identification when the upper pouch is separated
from the trachea. Any damage to the trachea
should be carefully avoided. The recurrent laryn-
geal nerve runs in the groove between the trachea
and the esophagus and must be identified and
protected. The upper fistula is closed flush to the
trachea and esophagus and is closed with 6/0
absorbable sutures.
719
55.8.2.3 EA with a Proximal
Tracheoesophageal Fistula
Only (1%)
In the above-mentioned case, a long gap between
the upper pouch and the lower esophagus has to be
expected; therefore, a primary thoracotomy is not
indicated. The first step is to perform a gastrostomy
and to insert a radio-opaque tube into the lower
esophagus. The X-ray shows either a rather short
lower esophageal pouch with a long distance
between the segments, or even a tiny anlage of
esophagus that cannot be used for an anastomosis.
The gastrostomy is needed for feeding the baby
and, depending on the choice of the surgeon, for the
longitudinal bougienage (see later in this chapter).
The upper fistula can be approached and iden-
tified via a right neck incision. A 3  F catheter
inserted during tracheoscopy allows easy identi-
fication of the fistula, which is then divided and
closed on both sides. The recurrent laryngeal
nerve should be identified and preserved. The
Replogle probe allows continuous suction from
the upper pouch until the final procedure will be
performed.
55.8.2.4 H-Type Fistula (4%)
The H-type fistula is identified by endoscopy and a
3 F ureteric catheter is passed across the fistula into
the esophagus. Most H-type fistulas can be
approached from the right side of the neck because
they are usually situated at, or above, the level of the
second thoracic vertebra. Palpation of the ureteric
catheter facilitates the identification of the fistula.
Again, the upper laryngeal nerve must be identified
and carefully preserved. The fistula is divided and
both sides are closed with 6/0 suture material.
55.8.2.5 Isolated EA: The Long-Gap
Problem (8%)
An airless abdomen on thoraco-abdominal X-ray
leads to suspicions of an EA without a lower fis-
tula (type A) or with an upper fistula only (type B).
An anastomosis is usually not possible in these
cases due to the long distance between the esopha-
geal pouches. Recent experience shows that
infants with type B have a shorter distance between
segments and most cases can be managed with a
720
primary anastomosis (Long et  al. 2017). In con-
trast, the surgical management of type A cases rep-
resents a major challenge due to the long gap and
is therefore extensively discussed in the literature.
The primary procedure consists of the place-
ment of a gastrostomy tube to allow early enteral
feeding. During the operative procedure, the dis-
tance between the pouches can be estimated
radiologically by inserting a radio-opaque tube in
the upper pouch and across the gastrostomy in the
lower pouch. A distance of three to four vertebral
bodies on a chest film is considered as a long gap,
and a distance of five or more vertebral bodies is
an ultra-long gap. However, there is no consensus
in the literature regarding the definition of a long
gap and the exact distance between the esopha-
geal pouches, since there are different techniques
to measure or estimate the length of the gap either
at birth, or at a later time, with or without gentle or
strong longitudinal pressure on one or both
pouches (Fig.  55.6, Table  55.3) (Bagolan et  al.
2013). Thus, the terms long gap, very long gap,
and ultra-long gap are used quite differently.
Thus, cases and results from different institutions
cannot be compared easily (Coran 1994). The use
of a retrograde endoscopy of the lower esophagus
through the stomach is useful, not only to estimate
a b
Fig. 55.6  The length of five vertebral bodies in the newborn is approximately 35 mm (left). The length of five vertebral
bodies in the 6-month-old child is approximately 47 mm (right)
M. E. Höllwarth and P. Zaupa
the length of the lower pouch, but also to exclude
the existence of a so-called nipple-like distal
esophageal remnant (Yeh et al. 2010).
Two basic surgical strategies are available in
cases of long-gap EA: either the preservation of
the patient’s own esophagus with delayed repair
or esophageal replacement. There is a general
consensus that all efforts should be directed to sal-
vage the child’s own esophagus because the ideal
graft does not exist. Concerning the preservation
of the native esophagus, several strategies exist:
1. Delayed Primary Anastomosis Until a Safe
Anastomosis Is Possible
In any case, it is essential to prove that a lower
esophageal segment truly exists and is signifi-
cantly different from a tiny nubbin at the upper
end of the stomach.
(a) Waiting for Spontaneous Growth
The basis of the strategy to wait for spontane-
ous elongation of the segments is that the swal-
lowing reflex for the upper pouch and the
gastroesophageal reflux for the lower pouch add
to the somatic growth, finally enabling a primary
55  Esophageal Atresia and Tracheoesophageal Fistula
anastomosis of the segments (Puri et al. 1992). A
recent meta-analysis of 451 cases of spontaneous
elongation reported about a 50% leak rate, sec-
ondary resections of the anastomosis due to
intensive stenosis in 21.5% and a secondary
esophageal replacement in 14% (Friedmacher
and Puri 2012). The advantage of that strategy is
the need for only one thoracotomy in nearly two-­
thirds of the patients, but the disadvantages are
Table 55.3  How long is a long gap?
2.5–4.5 cm (EA with lower fistula) >2 cm under
tension = 100% complications
Lafer and Boley (1966): >2.6 cm is long gap
and > 3.5 cm = ultra-long gap
Brown and Tam (1996): >3 cm during surgery under
normal conditions
Howard and Myers (1965): all babies without a lower
fistula
Bagolan et al. (2013): >3 cm or > 3 vertebral bodies
Spitz and Coran (2012): >4 vertebral bodies under
tension = > 4.0 cm
Any distance too wide to perform a primary repair
Fig. 55.7  Longitudinal bougienage of the upper and the lower esophageal pouch (metal probes within plastic tubes).
A near overlapping of the esophageal segments was achieved within 3 weeks
721
the long waiting time, the increasing financial
load, and the risk of aspiration pneumonia.
(b) Delayed Anastomosis with Longitudinal
Bougienage
Bougienage of the esophageal segments are
aimed to reduce the time until the anastomosis is
possible. The bougienage of the upper pouch was
first reported by Howard and Myers (1965). Many
further publications supported that technique
(Mahour et  al. 1974). Lafer and Boley (1966)
reduced the waiting period to 6 weeks by bougie-
nage of the upper and lower esophageal segments.
Recent experience shows that daily bougienage of
the upper and lower pouch for a few minutes
within the incubator, and under light sedation,
achieves a clear overlap of the segments on the
X-ray within 3–4  weeks, until the a­nastomosis
can be performed without tension (Fig. 55.7).
722 M. E. Höllwarth and P. Zaupa

Technical modifications have been published and the anastomosis can be performed by a sec-
by Hendren and Hale (1976), using electromag- ond thoracotomy. Even a thoracoscopic elonga-
netic devices for the longitudinal bougienage. tion of the esophagus was successfully performed,
Recently, good results have been published with thus avoiding the two thoracotomies (van der Zee
the use of a magnetic catheter-based system in 13 et  al. 2007). According to Foker, the technique
patients. Anastomosis was achieved between 3 successfully elongates esophageal segments
and 13 days (Slater et al. 2019). which are separated up to 12  cm. However, the
complication rate is significant: 42 patients
2. Forced Traction Methods needed additional rethoracotomies due to traction
sutures pulled out, replacement of the sutures, or
Extensive tension on the anastomosis, coupled adhesions, and in 2 out of 10 secondary resection
with the reduced microcirculation due to the of the anastomosis was needed (Foker et  al.
sutures, can lead to severe leakage or even dis- 2009). A survey of 88 international surgeons
ruption of the segments. To circumvent this prob- showed that 39% are using the Foker technique,
lem—but still be able to perform an anastomosis but 24% of those are not satisfied with the results
and preserving the child’s own esophagus— (Ron et al. 2009).
forced traction methods have been developed. In Kimura and Soper (1994), developed an
extrathoracic elongation technique of the upper
(a) Internal Traction pouch. First, a right-sided esophagostomy is cre-
ated in patients with a long-gap atresia. This
Rehbein was the first to publish a method of stoma is advanced subcutaneously after
intra-esophageal forced traction by means of sil- 2–6  months several times until the proximal
ver olives in each segment, which are pulled esophagus is long enough. In 12 patients, a final
together with a nylon thread (Rehbein and anastomosis was possible after 30 months (range
Schweder 1971). A nearly identical procedure 13–61  months); none of them needed a re-­
with Teflon balls instead of silver olives was pro- thoracotomy or secondary replacement. In three
posed by Harrison (2010). Originally, the thread patients, the procedure was performed thoraco-
was inserted into the esophagus by thoracotomy; scopically (Tamburri et al. 2009). Experiences of
later, this procedure was performed endoscopi- combining the Foker technique with Kimura’s
cally (Booss et  al. 1982). The aim of the tech- method resulted, however, in a high complication
nique was to bring the two segments together and rate (Sroka et al. 2013). Today, the Kimura tech-
to create an auto-anastomosis within 10  days. nique is rarely used.
However, the resulting stenosis needed long-term
bougienage and in secondary resection in 50% of 3. Esophageal Replacement
cases. A similar technique creating an auto-­
anastomosis by means of a thread was also suc- It is generally accepted that the preservation
cessfully applied in five patients (Stringel et  al. of the child’s own esophagus is the preferred
2010). method. With the development of sophisticated
techniques, the need to replace the esophagus is
(b) External Traction becoming rare. However, there are patients in
whom substitution of the esophagus is required,
In contrast to the internal traction, an external either because the distance between the segments
traction was introduced by Foker et  al. (1997). is too long or the primary procedure failed.
During thoracotomy, tissue-pledgeted traction For esophageal replacement, there are five
sutures are placed in the upper and lower segment options: reverse or isoperistaltic gastric tube,
and brought out to the skin below and above the colon interposition, jejunal interposition, and gas-
incision. Daily external traction of these sutures tric pull-up. Today, there is no agreement on a
brings the segments together within 14 ± 2.9 days, single organ or a single route. While gastric tubes
55  Esophageal Atresia and Tracheoesophageal Fistula
or pediculated or free transplanted jejunal tubes
are used in some centers only, more accepted
techniques are colon interposition and gastric
transposition (Spitz and Coran 2012). A detailed
overviews of the different replacement techniques
has been published (Loukogeorgakis and Pierro
2013; Shieh and Jennings 2017). A recent meta-
analysis, including 470 patients, showed that the
colon interposition is used in 73%, gastric pull-up
in 26%, and jejunal inter- position in 6% of cases
(Gallo et al. 2012). According to this analysis, the
colon interposition and the gastric pull-up are
comparable in regard to postoperative mortality (4
and 9%), anastomotic complications (16 and
18%), anastomotic leaks (17 and 31%), and graft
loss (4 and 5%), respectively. A disadvantage of
the colon interposition is the fact that the colon
has no propulsive peristalsis and the passage of
ingested food is entirely by gravity. A typical
complication is then an intrathoracic redundant
colon with delayed emptying and unpleasant sta-
sis of ingested food. Typical complications of the
gastric transposition are anastomotic leaks or
strictures and ulcers, delayed gastric emptying,
and, occasionally, dumping syndrome.
55.9 Complications
The incidence of early complications has been
reduced significantly in the last few decades
(Friedmacher et al. 2017). Today, a newborn with
EA is diagnosed in most centers shortly after birth
by the probe test, thereby preventing early pneu-
monia and aspiration after milk feeding.
Additionally, surgical techniques have been
refined and the quality of the suture material is sig-
nificantly better than it was in the past. Finally, the
progress in pre- and postoperative care, as well as
with the anesthesia techniques and pain control,
contribute largely to the excellent outcome rates in
the Waterston A and B groups (Hunt et al. 2016).
Long-term survival rates are around 95%, and
mortality is related to extreme prematurity and/or
major associated cardiac malformations. Despite
the excellent long-term survival rate, there are a
large number of early and late complications that
need special care and attention.
723
55.9.1 Early Complications
The incidence of an isolated anastomotic leak
should be lower than 10% if no tension was on
the anastomosis. Not surprisingly, they occur sig-
nificantly more often in babies with perioperative
problems and in long-gap cases with increased
tension at the anastomosis (Castilloux et al. 2010;
Mortell and Azizkhan 2009; Friedmacher and
Puri 2012).
They include minor anastomotic leaks which
occur in 6–28% in long-gap cases. An esophago-
gram with water-soluble contrast material may
show a tiny fistula from the anastomosis, indicat-
ing the anastomotic leak. If the patient’s condi-
tion is stable, oral feeding is possible because
spontaneous closure of the fistula can be expected.
The early sign of larger fistulas is the excre-
tion of saliva through the thoracic drain. In these
cases, oral feeding is postponed until the radio-
logical control shows the closure of the fistula.
Most of them will heal spontaneously. Major
leaks (3–5% up to 25% in long-gap cases) with
subtotal anastomotic insufficiency create a
­life-­threatening problem. Very early reoperation
is needed (Koivusalo et al. 2015). An early, but
fortunately rare complication, is the recurrence of
the tracheoesophageal fistula (3%), which usu-
ally occurs following a significant anastomotic
leak (Bruch et  al. 2010). In larger fistulas, air
bubbles come out of the drain and a pneumotho-
rax and a more or less extensive shadow can be
recognized on the thoracic X-ray. A spontaneous
closure cannot be expected; thus, the therapy
consists of a re-thoracotomy and surgical closure
of the fistula and the anastomosis. This procedure
is not easy because the local inflammatory pro-
cess causes edema and a reduced tissue quality.
Fibrinogen glue may be helpful, including the
use of a large pleural flap from the mediastinum
or a vascularized pericardial flap interposed
between the trachea and the esophagus.
Nevertheless, the surgical procedure is associated
with a significant morbidity and failure rate (Lal
and Oldham 2013).
Tracheoscopic strategies closing the fistula
from the internal site have gained popularity.
They are only successful in small fistulas, but
724
several endoscopic sessions are often necessary.
The strategies consist of filling up the fistula
either with synthetic tissue adhesives or fibrino-
gen. Since the adhesives easily glide into the
esophagus due to the shortness of the fistula,
additional destruction of the epithelial layer
either with diathermia or laser, mechanical abra-
sion or sclerosants, or subsequent use of tissue
glue is advised (Lal and Oldham 2013).
An anastomotic stenosis is a common finding.
The esophagogram often shows a narrow anasto-
mosis due to the differences in caliber of the
esophageal segments. This finding is different
from a true stenosis, and in most cases, oral feed-
ing is tolerated without symptoms. A true cicatri-
cial stenosis does not improve spontaneously and
causes a significant feeding problem sooner or
later. Minimal stenosis can be treated success-
fully with one to three careful dilatations—to
avoid esophageal rupture. A chronic esophageal
stenosis results often when significant tension is
on the anastomosis and it is aggravated by acidic
gastroesophageal reflux (Parolini et al. 2013). In
the first line, several dilatations as well as proton
pump inhibitor therapy (PPI) are used. However,
recent studies show that prophylactic use of PPIs
neither prevents the formation of strictures nor
reduces the number of dilatations needed
(Hagander et  al. 2012; Righini-Grundner et  al.
2019). In refractory cases, intralesional steroid
injections or local application of mitomycin addi-
tionally to the dilatations may be helpful
(Dall’Oglio et al. 2016). In the long run, a fundo-
plication is necessary in most cases. Recently,
covered stents are used, but proper location, tol-
erance by the patient, and migration of the stent
are typical problems (Levesque et al. 2013).
55.9.2 Late Complications
The most common late complication in babies
with EA is gastroesophageal reflux, causing
feeding problems, vomiting, reduced weight gain
or dystrophy, and recurrent respiratory tract
infections (Höllwarth 2012). Reflux is very typi-
M. E. Höllwarth and P. Zaupa
cal when the lower segment of the esophagus has
to be pulled up to be able to perform a primary
anastomosis. In contrast to otherwise normal
babies, there is no chance for spontaneous matu-
ration of the disturbed esophageal function due
the common dysmotility of the distal segment.
Chronic exposure of the anastomosis to refluxed
gastric acid is a well-known cause of recurrent
stenosis at the anastomosis and, on the long run,
epithelial metaplasia and even Barrett’s esopha-
gus. Since the propulsive peristalsis in the lower
segment of the esophagus is missing in most
cases, clearance time for acidic refluxes is signifi-
cantly longer when compared to children with
pathological reflux but without EA. Thus, about
33% of all patients with EA need a fundoplica-
tion, patients with type A and B in 100% and with
type C in 68% (Koivusalo et  al. 2018).
Fundoplication should be performed as soon as
significant clinical problems occur (Tovar and
Fragoso 2013).
Common long-term problems are recurrent
respiratory tract infections due to micro-­
aspiration (Kovesi 2013), or some kind of swal-
lowing disorders due to the missing or abnormal
peristaltic activity of the lower segment. However,
most patients learn to cope with this problem and
are used to drinking fluids along with their meals
as soon as they feel that a bolus does not pass eas-
ily through the esophagus.
Tracheomalacia is a common finding after EA
patients with lower fistula. The weak part of the
trachea is in the region of the former fistula. It
causes a typical barking cough and an inspiratory
stridor, which is in most cases self-limiting after
a few months. However, severe forms of tracheo-
malacia may lead to respiratory insufficiency,
apneic spells, and sudden infant death syndromes.
In these cases, an aortopexy is the most often per-
formed procedure, either by an open approach or
by thoracoscopy (Fraga et al. 2016; van der Zee
and Straver 2015). We used a Palmaz stent to sta-
bilize the trachea (Fig. 55.8) in several cases. The
stent is introduced endoscopically and the respi-
ratory problems are normalized immediately.
However, if the stent is not dilated sufficiently,
55  Esophageal Atresia and Tracheoesophageal Fistula
Fig. 55.8  A Palmaz stent used for severe tracheomalacia
with life-threatening spells. The stent stabilized the tra-
chea immediately and was removed after 8 months
recurrent granulation tissue is a typical complica-
tion due to the movements of the stent relative to
the trachea. In contrast, if the stent is in firm con-
tact with the trachea, the mucosa tends to grow
over, and removal after a few month may be dif-
ficult. Different types of stents are available
today, but experience is still poor.
55.10 Long-Term Follow-Up
Long-term morbidity involves the pulmonary, the
gastrointestinal, and the somatic development of
the child. Recurrent respiratory tract infections
are reported in 40–60% of patients and may be
associated with reflux-driven micro-aspirations
(Pedersen et al. 2017; Ijsselstijn et al. 2013; Hunt
et al. 2016).
Symptoms of dysphagia are reported in 38–85%
of patients and only 19% have been described as
free of digestive symptoms after 10–20  years
(Mahoney and Rosen 2016; Legrand et al. 2012).
Swallowing disorders are reported often, but most
patients learn to cope with this problem and become
725
used to drinking fluid together with their meals as
soon as they feel that a bolus does not pass easily
through the esophagus.
Chronic reflux is common and may recur after
fundoplication. Symptoms in adults are usually
rare and therefore chronic esophagitis may exist for
a long time. The incidence is up to 19% in patients
accepting regular endoscopy (Maynard and Bouin
2013). Chronic reflux may finally lead to intestinal
metaplasia and Barrett’s esophagus with the risk of
esophageal cancer. So far, eight cases of adenocar-
cinoma in patients with EA and distal fistula have
been described in the literature by (Rintala and
Pakarinen 2013; Jōnsson et al. 2019). Routine fol-
low-­up endoscopies with extensive biopsies are
recommended on all adult patients.
Spinal abnormalities exist in a significant
number of adult patients. Vertebral anomalies
have been detected in 45%, most often vertebral
fusions, and scoliosis in 56%, being 13-fold
higher when compared with healthy population
(Sistonen et  al. 2011). In this study, radial ray
anomalies have been found in 25%, shoulder
asymmetry in 80%, chest wall deformities in
15%, and rib fusion in 30% of patients. The latter
is caused when the ribs are sutured together at the
closure of the thoracotomy. Thus, it is advisable
not to suture the ribs together, and X-rays a week
after surgery show a perfect alignment of the ribs.
Developmental studies on the outcome at 1
and at 3 years showed that the group of 3-year-­
old infants has improved and scored within the
normal range for the age (Walker et  al. 2016).
However, the author conclude that this result
may be because of the early intervention ser-
vices for this group of children. Some children
born with EA have a risk for neurodevelopment
delay (Ijsselstijn et  al. 2016) and a multidisci-
plinary follow-up is needed. Additionally, psy-
chological support for parents should begin
during the neonatal period and should continue
during long-­term follow-up for children and par-
ents. While mental development is normal in
infancy and at preschool age, but in school-age
patients, gross motor problems as well as sus-
tained attention problems persist (Harmsen et al.
2017; Caplan 2013).
726
55.11 Conclusion
In conclusion, neonates with EA have an excel-
lent prognosis if no severe additional malforma-
tions are present. However, problems still exist in
babies with long-gap atresia. The use of the
patient’s own esophagus—e.g., with the Foker
method—causes significant long-term problems
due to the abnormal peristalsis, recurrent anasto-
motic stenosis, and severe gastroesophageal
reflux. Although esophageal replacement strate-
gies are well-established methods, they also have
a number of short- and long-term complications;
thus, the ideal method has not been found, and
more prospective studies are needed to compare
the different techniques.
References
Babu R, Pierro A, Spitz L et al (2000) The management of
EA in neonates with right-sided aortic arch. J Pediatr
Surg 35:56–58
Bagolan P, Valfrè L, Morini A (2013) Long-gap esopha-
geal atresia: traction-growth and anastomosis – before
and beyond. Dis Esophagus 26:372–379
Bartels E, Schulz AC, Mora NW et  al (2012) VATER/
VACTERL association: identification of seven new
twin pairs, a systematic review of the literature,
and a classical twin analysis. Clin Dysmorphol 21:
191–195
Bogs T et  al (2017) Esophageal atresia with or without
tracheoesophageal fistula (EA/TEF): association of
different EA/TEF subtypes with specific co-occurring
congenital anomalies and implications for diagnostic
workup. Eur J Pediatr Surg 28:176–182
Booss D, Höllwarth M, Sauer H (1982) Endoscopic
esophageal anastomosis. J Pediatr Surg 17(2):138–143
Brown AK, Tam PKH (1996) Measurement of gap length
in EA: a simple predictor of outcome. J Am Coll Surg
182:41–45
Bruch SW, Hirschl RB, Coran AG (2010) The diagnosis
and management of recurrent tracheoesophageal fistu-
las. J Pediatr Surg 45:337–340
Brunner HG, van Bokhoven H (2005) Genetic players in
esophageal atresia and tracheoesophageal fistula. Curr
Opin Genet Dev 15:341–347
Busa T et  al (2016) Esophageal atresia with tracheo-
esophageal fistula in a patient with 7q35-36.3 deletion
including SHH gene. Eur J Med Genet 59:546–548
Caplan A (2013) Psychological impact of esophageal
atresia: review of the research and clinical evidence.
Dis Esophagus 26:392–400
Castilloux J, Noble AJ, Faure C (2010) Risk factors for
short- and long-term morbidity in children with esoph-
ageal atresia. J Pediatr 156(5):755–760
M. E. Höllwarth and P. Zaupa
Coran AG (1994) Ultra long-gap esophageal atresia: how
long is long? Ann Thorac Surg 57:528–529
Dall’Oglio L et  al (2016) Endoscopic management of
esophageal stenosis in children: new and traditional
treatments. World J Gastrointest Endosc 25:212–219
Durston W (1670) A narrative of a monstrous birth in
Plymouth. Phil Trans R Soc A V:2096
Felix JF, de Jong EM, Torfs CP et al (2009) Genetic and
environmental factors in the etiology of EA and/or
tracheoesophageal fistula: an overview of the cur-
rent concepts. Birth Defects Res A Clin Mol Teratol
85:747–754
Foker JE, Linden BC, Boyle EM, Marquardt C (1997)
Development of a true primary repair for the full spec-
trum of EA. Ann Surg 226(4):533–543
Foker JE, Krosch TCK, Catton K et al (2009) Long-gap
esophageal atresia treated by growth induction: the
biological potential and early follow-up results. Semin
Pediatr Surg 18:23–29
Fraga JC et  al (2016) Pediatric tracheomalacia. Semin
Pediatr Surg 25:156–164
Friedmacher F, Puri P (2012) Delayed primary anastomo-
sis for management of long-gap EA: a meta- analysis
of complications and long-term outcome. Pediatr Surg
Int 28:899–906
Friedmacher F et  al (2017) Postoperative complications
and functional outcome after esophageal atresia
repair: results from longitudinal single-center follow-
­up. J Gastrointest Surg 21:927–935
Gallo G, Zwaveling S, Groen H et  al (2012) Long-
gap esophageal atresia: a meta-analysis of jejunal
­interposition, and gastric pull-up. Eur J Pediatr Surg
22(6):420–425
Gibson T (1697) The anatomy of human bodies epito-
mized. Awnsham & Churchill, London
Gross RE (1953) Atresia of the oesophagus. In: Gross RE
(ed) Surgery of infancy and childhood, 1st edn. WB
Saunders, Philadelphia, PA
Hagander L, Muszynska C, Arnbjornsson E et al (2012)
Prophylactic treatment with proton pump inhibitors
in children operated on for oesophageal atresia. Eur J
Pediatr Surg 22:139–142
Hajduk P, Sato H, Puri P et al (2011) Abnormal notochord
branching is associated with foregut malformations
in the adriamycin treated mouse model. PLoS One
6:e27635
Hajduk P, May A, Puri P et al (2012) The effect of adria-
mycin exposure on the notochord of mouse embryos.
Birth Defects Res B Dev Reprod Toxicol 95:175–183
Harmsen WJ et  al (2017) Developmental problems in
patients with oesophageal atresia: a longitudinal
follow-up study. Arch Dis Child Fetal Neonatal Ed
102:F214–F219
Harrison MR (2010) What if?. . . Why not? J Pediatr Surg
45:1–10
Hendren WH, Hale JR (1976) EA treated by electromag-
netic bougienage and subsequent repair. J Pediatr Surg
11(5):713–722
Höllwarth ME (2012) Gastroesophageal reflux dis-
ease. In: Coran et al (eds) Pediatric surgery, 7th edn.
Saunders, Elsevier
55  Esophageal Atresia and Tracheoesophageal Fistula
Holocomb GW III, Rothenberg SS, Bax KMA et  al
(2005) Thoracoscopic repair EA and tracheoesopha-
geal fistula. A multi-institutional analysis. Ann Surg
242(3):422–430
Howard R, Myers NA (1965) EA: a technique for elongat-
ing the upper pouch. Surgery 58:725–727
Hunt RW et al (2016) Peri-operative management of neo-
nates with esophageal atresia and trachea-esophageal
fistula. Pediatr Resp Review 19:3–9
Ijsselstijn H, van Beelen NWG, Wijnen RMH (2013) EA:
long-term morbidities in adolescence and adult- hood.
Dis Esophagus 26:417–421
Ijsselstijn H et  al (2016) Growth and development after
oesophageal atresia surgery: need for long-term mul-
tidisciplinary follow-up. Pediatr Resp Review 19:
34–38
Ioannides AS, Massa V, Ferraro E et  al (2010) Foregut
separation and trachea-oesophageal malformations:
the role of tracheal outgrowth, dorso-ventral pattering
and programmed cell death. Dev Biol 337:351–362
Jōnsson L et  al (2019) Long-term effectiveness of anti-
reflux surgery in esophageal atresia patients. Eur J
Pediatr Surg 29:521–527
Källén B, Finnström O, Lindam A et al (2010) Congenital
malformations in infants born after in  vitro fertiliza-
tion in Sweden. Birth Defects Res A Clin Mol Teratol
88:137–143
Kimura K, Soper RT (1994) Multistaged extrathoracic
esophageal elongation for long gap EA. J Pediatr Surg
29(4):566–568
Kluth D (1977) Die Missbildungen des Oesophagus
und der Trachea. Dissertation monography, Medical
University of Lübeck
Koivusalo A et al (2015) Revisional surgery for recurrent
tracheoesophageal fistula and anastomotic complica-
tions after repair of esophageal atresia in 258 infants. J
Pediatr Surg 50:250–254
Koivusalo A et  al (2018) Outcome of fundoplication in
oesophageal atresia associated gastrooesophageal
reflux. J Pediatr Surg 53:230–233
Kovesi T (2013) Long-term respiratory complications of
congenital EA with or without tracheoesophageal fis-
tula: an update. Dis Esophagus 26:413–416
Lafer DJ, Boley SJ (1966) Primary repair in EA with elonga-
tion of the lower segment. J Pediatr Surg 1(6):585–587
Lal DR, Oldham KT (2013) Recurrent tracheo-­esophageal
fistula. Eur J Pediatr Surg 23(3):214–218
Legrand C, Michaud L, Salleron J et al (2012) Long-term
out- come of children with eosophageal atresia type
III. Arch Dis Child 97:808–811
Lévesque D, Baird R, Laberge JM (2013) Refractory
strictures post-EA repair: what are the alternatives?
Dis Esophagus 26:382–387
Lindahl H (1987) EA: a simple technical detail aiding the
mobilization and circular myotomy of the proximal
segment. J Pediatr Surg 22(2):113–114
Litingtung Y et al (1998) Sonic hedgehog is essential for
foregut development. Nat Genet 20:58–63
727
Livaditis A, Radberg L, Odensjo G (1972) Esophageal
end-to- end anastomosis. Reduction of anasto-
motic tension by circular myotomy. Scand J Thorac
Cardiovasc Surg 6:206–214
Long A-M et  al (2017) Esophageal atresia with no dis-
tal tracheoesophageal fistula: management and out-
come from a population-based cohort. J Pediatr Surg
52:226–230
Loukogeorgakis SP, Pierro A (2013) Replacement surgery
for EA. Eur J Pediatr Surg 23(3):182–190
Mahoney L, Rosen R (2016) Feeding difficulties in chil-
dren with EA. PEAdiatr Respir Rev 19:21–27
Mahour GH, Woolley MM, Gwinn JL (1974) Elongation
of the upper pouch and delayed anatomic reconstruc-
tion in EA. J Pediatr Surg 9(3):373–383
Maynard S, Bouin M (2013) Follow-up of adult patients
with repaired EA: how, when and for how long? Dis
Esophagus 26:422–424
Mazurak M, Patkowski D (2018) The visionaries, the vir-
tuosos, and the history of congenital EA surgery. Ann
Thorac Surg 106:305–308
Mc Laughlin D, Hajduk P, Murphy P et  al (2013)
Adriamycin- induced models of VACTERL associa-
tion. Mol Syndromol 4:46–62
Mortell EA, Azizkhan RG (2009) EA repair with tho-
racotomy: the Cincinnati contemporary experience.
Semin Pediatr Surg 18(1):12–19
Muensterer OJ, Berdon WE (2015) From Vogt to Haight
and Holt to now: the history of EA over the last cen-
tury. Pediatr Radiol 45:1230–1235
Nassar N, Leoncini E, Amar E (2012) Prevalence of EA
among 18 international birth defects surveil- lance
programs. Birth Defects Res A Clin Mol Teratol
94(A):893–899
Oddsberg J, Lu Y, Lagergren J (2010) Maternal diabetes
and risk of EA. J Pediatr Surg 45:2004–2008
Oddsberg J, Lu Y, Lagergren J (2012) Aspects of EA in
a population-based setting: incidence, mortality and
cancer risk. Pediatr Surg Int 28:249–257
Otte JB, Gianello P, Wese FX et al (1984) Diverticulum
formation after circular myotomy for EA.  J Pediatr
Surg 19:68–71
Parolini F, Leva E, Morandi A et al (2013) Anastomotic
strictures and endoscopic dilatations following esoph-
ageal atresia repair. Pediatr Surg Int 29:601–605
Pedersen et  al (2012) EUROCAT Working Group.
Oesophageal atresia prevalence, prenatal diagnosis
and associated anomalies in 23 European regions.
Arch Dis Child 97:227–232
Pedersen RN, Markøw S, Kruse-Andersen S et al (2017)
Long- term pulmonary function in EA – a case control
study. Pediatr Pulmonol 52(1):98–106
Puri P, Ninan GK, Blake NS et al (1992) Delayed primary
anastomosis for EA: 18 months’ to 11 years’ follow-
­up. J Pediatr Surg 27:1127–1130
Rehbein F, Schweder N (1971) Reconstruction of the
esophagus without colon transplantation in cases of
atresia. J Pediatr Surg 6:746–752
728
Reutter H, Ludwig M (2013) VATER/VACTERL asso-
ciation: evidence for the role of genetic factors. Mol
Syndromol 4:16–19
Righini-Grundner F et  al (2019) Should proton pump
inhibitors be systematically prescribed in patients
with esophageal atresia after surgical repair? J Pediatr
Gastroenterol Nutr 69:45–51
Rintala RJ, Pakarinen MP (2013) Long-term outcome
of esophageal anastomosis. Eur J Pediatr Surg
23:219–225
Ron O, De Coppi P, Pierro A (2009) The surgical approach
to EA repair and the management of long- gap atresia:
results of a survey. Semin Pediatr Surg 18:44–49
Rothenberg SS, Flake AW (2015) Experience with tho-
racoscopic repair of long gap EA in neonates. J
Laparoendosc Adv Surg Tech A 25:932–935
Schärli AF (1992) Esophageal reconstruction in very long
atresias by elongation of the lesser curvature. Pediatr
Surg Int 7:101–105
Sfeir R, Michaud L, Salleron L, Gottrand F (2013)
Epidemiology of EA. Dis Esophagus 26:354–355
Shieh H, Jennings RW (2017) Long-gap esophageal atre-
sia. Semin Pediatr Surg 26:72–77
Sistonen SJ, Pakarinen MP, Rintala RJ (2011) Long-term
results of EA: Helsinki experience and review of lit-
erature. Pediatr Surg Int 27:1141–1149
Slater BJ et al (2019) Use of magnets as a minimally inva-
sive approach for anastomosis in esophageal atresia:
long-term outcome. J Laparascop Adv Surg Tech A
29:1202–1206
Solomon BD, Bear KA, Kimonis V et al (2012) Clinical
geneticists’ views of VACTERL/VATER association.
Am J Med Genet 158A:3087–3100
Spitz L, Coran AG (2012) Esophageal replacement. In:
Coran AG, Adzick NS, Krummel TM, Laberge J-M,
Shamberger RC, Caldamone AA (eds) Pediatric sur-
gery, 7th edn. Elsevier Saunders, Philadelphia
Spitz L, Kiely EM, Morecroft JA et al (1994) OEA: at-risk
groups for the 1990s. J Pediatr Surg 29:723–725
Sroka M, Wachowiak R, Losin M et al (2013) The Foker
technique (FT) and Kimura advanced (KA) for the
treatment of children with long-gap EA (LGEA): les-
sons learned at two European centers. Eur J Pediatr
Surg 23(1):3–7
M. E. Höllwarth and P. Zaupa
Stoll C et  al (2017) Associated anomalies in cases with
esophageal atresia. Am J Med Genet A 173:2139–2157
Stringel G, Lawrence C, Mc BW (2010) Repair of
long gap EA without anastomosis. J Pediatr Surg
45(5):872–875
Taghavi K, Stringer MD (2018) Preoperative laryngotra-
cheobrnchoscopy in infants with esophageal atresia:
why is it not routine? Pediatr Surg Int 34:3–7
Tamburri N, Laje P, Boblione M et al (2009) Extrathoracic
esophageal elongation (Kimura’s technique): a feasi-
ble option for the treatment of patients with complex
EA. J Pediatr Surg 44:2420–2425
Tovar JA, Fragoso AC (2013) Anti-reflux surgery for
patients with EA. Dis Esophagus 26:401–404
van der Zee DC, Straver M (2015) Thoracoscopic aorto-
pexy for tracheomalacia. World J Surg 39:158–164
van der Zee DC, Vieirra-Travassos D, Kramer WLM et al
(2007) Thoracoscopic elongation of the esophagus in
long gap EA. J Pediatr Surg 42:1785–1788
Vogt EC (1929) Congenital atresia of the esophagus. Am
J Roentgenol 22:463
von Lennep M et al (2019) Oesophageal atresia. Nat Rev
Dis Primers 5:1–21
Walker K et al (2016) Developmental outcomes at three
years of age of infants with esophageal atresia. J
Pediatr Surg 51:249–251
Waterston DJ, Carter RE, Aberdeen E (1962) Oesophageal
atresia: trachea-oesophageal fistula; a study of survival
in 218 infants. Lancet 1:819–822
Way C et al (2019) Thoracoscopy vs. thoracotomy for the
repair of esophageal atresia and tracheoesophageal
fistula: a systematic review and meta-analysis. Pediatr
Surg Int 35:1167–1184
Yang YF, Dong R, Zheng C et al (2016) Outcomes of tho-
racoscopy versus thoracotomy for esophageal atresia
with tracheoesophageal fistula repair: a PRISMA-
compliant systematic review and meta-analysis.
Medicine 95:e4428
Yeh SH, Ni YH, Hsu WM et al (2010) Use of retrograde
esophagoscopy in delayed primary esophageal anas-
tomosis for isolated EA. Eur J Pediatr Surg 20:40–44
Zani A, Eaton S, Hoellwarth ME et al (2014) International
survey on the management of EA. Eur J Pediatr Surg
24:3–8
 Achalasia
Fanny Yeung, Kenneth Wong, and Paul Tam
56.1 Introduction
Achalasia is an esophageal motility disorder char-
acterized by the absence of esophageal peristalsis,
high resting pressure of the lower esophageal
sphincter (LES), and failure of receptive relaxation
of the LES. Disease progression results in gradual
dilatation of the esophagus above the sphincter.
56.2 Historical Overview
Achalasia in pediatric population is rare and the
development of treatment modalities mainly fol-
lowed adult studies. The condition was first
described in 1674 by Willis who successfully
treated a patient by repeated esophageal bougie-
nage. Subsequently, other different methods of
esophageal dilatation had been introduced, includ-
ing mechanical, hydrostatic, and pneumatic dila-
tation (Franklin et al. 2014; Boeckxstaens et al.
F. Yeung
Queen Mary Hospital, Hong Kong, SAR, China
K. Wong
University of Hong Kong, Queen Mary Hospital,
Hong Kong, SAR, China
e-mail: kkwong@hku.hk
P. Tam (*)
Macau University of Science and Technology,
Macau, SAR, China
e-mail: paultam@hku.hk
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_56
56
2014; Kahrilas et al. 2015; Eckardt et al. 1992).
Cardiomyotomy, the basis of modern surgical treat-
ment for achalasia, was introduced by Heller in
1914. The original operation consisted of both ante-
rior and posterior myotomies on the lower esopha-
gus performed through a laparotomy. Groeneveldt
modified Heller’s original procedure by performing
only anterior myotomy on the lower esophagus in
1918. This is now known as the “modified Heller
esophagomyotomy.” Asch reported a newborn with
achalasia treated with esophagomyotomy, by far the
youngest patient described in the 1970s (Asch et al.
1974). Grosfeld also described the modified Heller
esophagomyotomy via a midline, subxiphoid inci-
sion in nine children operated between 1972 and
1979 (Ballantine et al. 1980). The first laparoscopic
esophagomyotomy in children was described a
decade later by (Holcomb et al. 1996). In 2001,
Rothenberg reviewed his experience with thoraco-
scopic and later laparoscopic Heller myotomy for
children aged from 5 to 17 years and concluded that
thoracoscopic approach required longer operating
time and hospital stay and, without a partial fun-
doplication, also may be associated with a higher
incidence of silent gastroesophageal reflux when
compared to laparoscopic approach (Rothenberg et
al. 2001). Since the first pediatric report of peroral
endoscopic myotomy (POEM) in a 3-year-old girl
with Down syndrome with satisfactory treatment
outcomes at 1-year follow-up in 2012 by Maselli,
increasing studies have found POEM to be an effec-
tive and safe alternative treatment for achalasia in
729
730
pediatric population (Maselli et al. 2012; Liu et al.
2020; Chen et al. 2015; Yeung et al. 2018; Lee et al.
2019; Khashab et al. 2020; Ren et al. 2017).
56.3 Incidence
Achalasia is a rare esophageal neurodegenerative
disorder in the pediatric population with an
annual incidence of 0.11 cases per 100,000 chil-
dren and is more prevalent in males.
56.4 Etiopathogenesis
56.4.1 Etiology
Achalasia can be classified into primary (idio-
pathic) or secondary. For primary achalasia, the
imbalance between the inhibitory and excitatory
neurotransmitters in the distal esophagus second-
ary to loss of the myenteric ganglion cells has
been proposed as the possible etiological factor.
Epidemiology studies have revealed a small inci-
dence peak in the 20–40 years age group. These,
together with the rare occurrence of familial
cases and syndromic associations, have led to the
postulation of a possible hereditary factor for the
disease (Franklin et al. 2014; Boeckxstaens et al.
2014).
For secondary achalasia, the most common
cause is Chagas’ disease, which is caused by
Trypanosoma cruzi infection. Although viral infec-
tions such as measles, VZV (varicella-zoster virus),
and HSV1 (herpes simplex virus type 1) have been
suggested, no study has so far established a causal
link. Autoimmune etiology has also been suggested
by the presence of circulating autoantibodies against
the myenteric plexus, the presence of inflammatory
T-cell infiltrates in the myenteric plexus, and the
increased prevalence of HLA class II antigens.
Pseudoachalasia may also be caused by infiltra-
tive diseases such as amyloidosis and sarcoidosis
or by neoplasms involving the esophagus (e.g.,
esophageal leiomyoma in children). An achalasic
picture may also be mimicked by complications of
surgery such as fundoplication or vagal injury.
Rarely, achalasia-like esophageal dysmotility is
found in patients with esophageal atresia.
F. Yeung et al.
A very rare cause of achalasia in children can be
part of the Allgrove’s (Four “A”) syndrome, charac-
terized by achalasia, alacrima, autonomic distur-
bance, and adrenocorticotrophic hormone-­resistant
adrenal insufficiency. Allgrove’s syndrome has
been linked to mutations in the AAAS gene on
chromosome 12q13.8. Recent observations also
suggest that achalasia is more frequent in Down’s
syndrome. Nonetheless, there is little evidence from
the studies available that achalasia is a genetic dis-
ease, except in the case of Allgrove’s syndrome.
56.4.2 Pathophysiology
After swallowing, the coordinated peristaltic
wave that moves the food bolus through the
esophagus and the subsequent relaxation of the
LES depends on coordinated interaction of enteric
nervous plexus, vagus nerve, interstitial cells of
Cajal (ICC), and gastrointestinal hormones. The
complex physiology highlights the multiple
potential pathologic targets that could give rise to
the clinical picture of achalasia.
Esophageal body aperistalsis characteristic of
achalasia occurs over time in most patients. The
pathologic characteristic of achalasia is the loss
of myenteric neurons in the LES and esophageal
body, accompanied by an inflammatory infiltra-
tion. However, a spectrum of histopathologic
changes, such as interstitial fibrosis, endoneurofi-
brosis, fibrosis of the smooth muscle layer, and
myopathic changes of the smooth muscle cells,
have also been observed. In patients with long-­
standing achalasia, myenteric ganglion cell bod-
ies are markedly diminished or absent.
Most data now strongly suggest that the major
inhibitory neurotransmitter governing relaxation of
esophageal smooth muscle is nitric oxide (NO). The
role of the loss of nitrergic neurons in the pathophys-
iology of achalasia is highlighted by studies on mice
selectively lacking neuronal nitric oxide synthase
(nNOS). These mice have manometric findings that
mimic those in achalasia in the human. Nonetheless,
the loss of nitrergic inhibitory enteric neurons is not
selective in achalasic patients and appears to occur
prior to the loss of cholinergic neurons. Other inhibi-
tory neurotransmitters including vasoactive intesti-
nal peptide (VIP) may also play a role.
56 Achalasia
Recently, immunohistochemical techniques
have consistently demonstrated the presence of a
CD8 lymphocytic infiltrate and collagen deposi-
tion within the myenteric plexus. These findings
suggest an immune-mediated destruction of
myenteric neurons. The exact stimulus initiating
this autoimmune response or the antigen targeted,
however, remains to be identified (Franklin et al.
2014; Boeckxstaens et al. 2014).
56.5 Diagnosis
The diagnosis of achalasia is suggested by clini-
cal features and confirmed by further diagnostic
tests such as barium swallow, esophagogastrodu-
odenoscopy, and manometry. These investiga-
tions are not only important in establishing the
diagnosis but are also helpful to grade the disease
severity and predict treatment outcome.
56.5.1 Clinical Features
The typical symptoms of achalasia are progres-
sive dysphagia to solids initially and to liquids
later on, regurgitation of undigested food, respi-
ratory complications (nocturnal cough and aspi-
ration), chest pain, and weight loss. In infants,
these symptoms may mimic gastroesophageal
reflux, with associated failure to thrive.
The early symptoms of achalasia may be non-
specific and diagnosis is often delayed. Indeed,
the mean duration of symptoms before diagnosis
is up to 2  years due to frequent confusion with
gastroesophageal reflux disease (GERD), low
index of suspicion of achalasia, and underuse of
esophageal manometry.
In older children, the Eckardt symptom score
(Table 56.1) is useful to quantify the severity and
frequency of symptoms. The score ranges from 0
Table 56.1  Eckardt score for the evaluation of achalasia symptoms
Score Weight loss (kg) Dysphagia
0 None None
1 <5 Occasional
2 5–10 Daily
3 >10 Each meal
731
to 12 with higher scores indicating more pro-
nounced symptoms. Generally, scores of 0–1,
2–3, 4–6, and > 6 correspond to clinical stages 0,
I, II, and III, respectively. It is particularly useful
to compare the pre- and posttreatment Eckardt
symptom scores to assess the treatment effect
(Eckardt et al. 1992; Ren et al. 2017).
56.5.2 Radiology
Chest x-ray may show an air-fluid level in the
posterior mediastinum and/or a soft tissue medi-
astinal shadow in the left hemithorax correspond-
ing to a dilated lower esophagus. Chronic
aspiration may lead to pneumonic changes. A
barium esophagogram is usually the first investi-
gation performed in the evaluation of dysphagia.
The classical appearance of achalasia on a bar-
ium study is the smooth tapering of the gastro-
esophageal junction (“bird’s beak” or “rat-tail”
deformity) with a column of contrast often mixed
with retained food in the dilated proximal esoph-
agus (Fig. 56.1).
56.5.3 Upper Endoscopy
Endoscopic evaluation is used to rule out other
conditions that may mimic achalasia. The charac-
teristic appearance is an atonic, dilated esopha-
gus with a tightly closed LES that does not open
with insufflation (Fig.  56.2) but provides little
resistance to the advancing endoscope which can
be admitted through with a “pop.” Food residues,
retained secretion and an inflamed mucosa are
often noted in the esophagus. Complications such
as ulceration and yeast esophagitis and suspicion
of malignancy are indications for esophageal
biopsies. The diagnosis of pseudoachalasia is
usually difficult and requires meticulous endo-
Retrosternal pain Regurgitation
None None
Occasional Occasional
Daily Daily
Each meal Each meal
732
Fig. 56.1  A barium esophagram showing the typical
radiological appearance of a “bird’s beak” in achalasia
Fig. 56.2  Endoscopic view of the lower esophageal
sphincter in a patient with achalasia, with characteristic
appearance of an atonic and dilated esophagus with a
tightly closed LES that does not open with insufflation
scopic inspection. In children, it is important to
exclude tracheobronchial remnants in the distal
esophagus mimicking as achalasia. Endoscopic
ultrasound examination is useful for the
differentiation.
F. Yeung et al.
56.5.4 Esophageal Manometry
In clinical practice, esophageal manometry is the
gold standard for confirming the diagnosis of
achalasia after exclusion of organic causes of dys-
phagia. Diagnosis is established manometrically
with characteristic findings of impaired or absent
esophagogastric junction (EGJ) relaxation and
esophageal aperistalsis after exclusion of pseudo-
achalasia or mechanical obstruction. In the patient
who cannot tolerate esophageal manometry,
nuclear scintigraphy is useful in evaluating esoph-
ageal transit.
56.5.5 High-Resolution Manometry
and Esophageal Pressure
Topography
In recent years, high-resolution manometry
(HRM) and esophageal pressure topography
(EPT) have largely replaced conventional
manometry as it provides a continuum of intralu-
minal pressure monitoring.
In conventional manometry, a pull-through tech-
nique is used to determine the position of the lower
esophageal sphincter (LES) pressure by identifying
a high-pressure zone. Furthermore, pressure sensors
are spaced anywhere from 3 to 5 cm apart. In HRM,
the distal end of the catheter is passed into the gas-
tric compartment below the LES and hiatal canal,
and no pull through is required. Pressure sensors of
20–36 are spaced at 1 cm intervals, decreasing the
spacing between the sensors significantly when
compared to conventional manometry. During an
HRM study, the integrated relaxation pressure (IRP,
normal value <15 mmHg) at the EGJ is measured,
and color-­coded EPT plots that allow spatiotempo-
ral representations of pressure recordings in the
esophagus are generated by computer software dur-
ing ten wet swallows to assess the peristalsis pat-
tern. This technology thus lends itself to a more
accurate and an objective assessment of esophageal
motility disorders.
56.5.5.1 Classification of Achalasia
The use of HRM and EPT not only establishes
the diagnosis but also allows classification of
achalasia into subtypes that has prognostic value
56 Achalasia
and can be used to guide the subsequent treat-
ment. Three subtypes of achalasia has been
described according to the Chicago classifica-
tion differentiated by the patterns of non-peri-
Fig. 56.3 (a) High-resolution manometric recordings of
Type I achalasia, characterized by 100% failed contrac-
tions and no esophageal pressurization. (b) Type II acha-
lasia in an 11-year-old boy (UES at 17  cm and LES at
37 cm probe marking) showing impaired OGJ relaxation
733
staltic esophageal pressurization that
accompanied the elevated IRP  >  15  mmHg
(Fig.  56.3): Type I (characterized by 100%
failed contractions and no esophageal pressur-
with median integrated relaxation pressure of 31.5 mmHg
over ten test swallows and pan-esophageal pressurization.
(c) Type III spastic achalasia, defined by the presence of
premature contractions for at least 20% of the swallows
734
Fig. 56.3 (continued)
ization), Type II (achalasia with compression,
defined by pan-esophageal pressurization occur-
ring with at least 20% of swallows and absence
of a peristaltic contraction), and Type III (spas-
tic achalasia, defined by the presence of prema-
ture contractions for at least 20% of the
swallows) (Kahrilas et al. 2015).
These three subtypes have different clinical
characteristics and treatment outcomes. Type II
achalasia is the most common subtypes and is
more prevalent in males, while Type III subtype
is the least common subtype in pediatric popula-
tion. Type I and II patients usually present with
severe dysphagia whereas patients with Type III
achalasia present more with chest pain. For treat-
ment outcome, Type II patients have the best out-
comes whereas Type III patients have the worst
response to therapy and have a significantly
greater risk of residual or recurrent symptoms
requiring therapeutic interventions than Types I
and II patients.
F. Yeung et al.
56.6 Differential Diagnosis
Patients with achalasia typically presented with
progressive dysphagia and feeding problem.
Sometimes, it may be difficult to differentiate
between gastroesophageal reflux and achalasia
especially in younger children or infants. Other
conditions that have similar symptoms to achala-
sia include diffuse esophageal spasm, paraesoph-
ageal hiatal hernia, esophageal diverticulum,
congenital esophageal stenosis, pseudoachalasia,
and gastric volvulus. Thorough clinical examina-
tion and investigations are warranted for accurate
differentiation and diagnosis.
56.7 Management
The treatment of achalasia is aimed at improving
bolus transport by reducing the pressure at the
LES. At present, established treatment options in
achalasia are pharmacotherapy, injection of botu-
56 Achalasia
linum toxin, pneumatic dilatation, and cardiomy-
otomy via surgical or endoscopic means.
56.7.1 Pharmacotherapy
Pharmacological treatments including the use of
calcium channel blockers and nitrates have been
employed with limited long-term success. These
medications are generally taken by mouth
10–30 min before a meal. As symptomatic relief
from these agents is inconsistent and short-lived,
with associated cardiovascular side effects, their
usefulness is limited.
56.7.2 Botulinum Injection
Local injection of botulinum toxin reduces pres-
sure at the LES by temporarily blocking the
release of acetylcholine from excitatory motor
neurons at the level of the LES.  Typically,
100 units of toxin are injected into the four quad-
rants of the LES (25 units/quadrant) endoscopi-
cally. While the initial rate of symptomatic
benefit may be comparable to that of pneumatic
dilatations, the effect wears off quickly in many
patients (median duration of effect of 3 months).
Also, the response seems to diminish with
repeated injections. The major advantages of this
therapy are its technical simplicity and relative
safety, and the procedure can typically be per-
formed under minimal conscious sedation.
Hence, botulinum toxin should be used only in
children who are poor candidates for either pneu-
matic dilation or surgery.
56.7.3 Pneumatic Dilatation
Pneumatic dilatation aims at disruption of the
LES circular muscle by balloon insufflation. The
procedure can be performed under sedation or
general anesthesia. A wire-guided/incorporated
balloon catheter is introduced via videoendoscope
across the LES. The maximum insufflated size of
the balloon should be no greater than the patient’s
own thumb. The balloon is inflated under fluoro-
scopic control until the “waist” disappears.
735
Follow-up studies have indicated that the best
predictor for a long-term remission is a posttreat-
ment sphincter pressure less than 10  mmHg.
Pneumatic dilatation yields good to excellent
results in 70–90% of patients, but on long-term
follow-up more than 40% of patients require addi-
tional treatment for recurrent symptoms. Pneumatic
dilatation is associated with a small but notable risk
(1–5%) of perforation. Moreover, patients who
eventually undergo Heller myotomy after previous
endoscopic therapy experience more intraoperative
complications especially esophagogastric perfora-
tion likely related to the post-dilatation fibrosis and
scarring. Those who have had prior endoscopic
therapy also experience more postoperative com-
plications and a higher failure rate than when no
preoperative therapy has been used.
56.7.4 Surgical Myotomy
Cardiomyotomy can achieve symptomatic relief
in 90–95% of patients. It also provides more
long-lasting effect than pneumatic dilatation
(Holcomb et al. 1996); Rothenberg et al. 2001).
Traditional open approach is associated with a
large incision and a prolonged postoperative stay.
Recent developments in minimally invasive tech-
niques now allow performance of cardiomyotomy
laparoscopically or with robotic assistance. The
reduction in postoperative pain and morbidity, as
well as improved cosmesis, has made this surgical
option the treatment of choice for most patients.
The advantages of laparoscopic cardiomyotomy
include excellent visualization of the distal esoph-
agus and the stomach, so that an extended gastric
myotomy and a concomitant antireflux procedure
may be performed. ­Robotic-­assisted laparoscopic
Heller myotomy in pediatric patients have been
reported, with no difference in operative time,
occurrence of complications or recurrences, and
postoperative pain control and hospital stay com-
paring with the laparoscopic approach. The major
advantages of the robotic-­assisted approach are
the magnified three-­ dimensional imaging, high
degree of freedom of movements related to the
articulated wrists, and the tremor filtration pro-
vided by the computer interface. All these factors
may contribute to the lower risk of mucosal perfo-
736
ration in robotic Heller myotomy as reported in
some centers.
The setup for laparoscopic Heller myotomy is
the same as that for a Nissen fundoplication. The
patient is placed in a modified lithotomy position
with a large-bore nasogastric tube in situ.
Standard five-port placement is the norm,
although some surgeons have reported using
fewer ports by employing the single incision lap-
aroscopic surgery (SILS) technique.
An extensive anterior and lateral hiatal and
mediastinal esophageal dissection is performed to
maximize the length of the myotomy, taking care
not to breach the pleura proximally. The gastric fun-
dus and short gastric vessels are then mobilized to
minimize tension on the subsequent fundoplication.
The anterior vagus nerve is identified and protected
from injury during the esophagogastric myotomy.
Myotomy is achieved by using the hook electrocau-
tery device. The longitudinal muscle fibers are
divided first, exposing the inner circular muscle
which is then separated until the submucosal plane
is reached. The myotomy should be carried proxi-
mally to the level of dilated esophagus (4–6 cm) and
distally 1–2 cm onto the stomach. The GE junction
is identified by the overlying fat pad and the “collar-
like” circular muscle. On completion of the myot-
omy, the esophagus is insufflated to demonstrate the
mucosal bulge. Accidental mucosal perforations
should be repaired with fine absorbable sutures.
a b
Fig. 56.4 (a) Mucosa is incised longitudinally over ante-
rior wall of the mid-esophagus at 8 cm proximal to OGJ
with the submucosal tunnel created. (b) Anterior myot-
omy cutting the circular muscle layer while preserving the
F. Yeung et al.
An antireflux procedure, either Dor (anterior
180°) or Toupet (posterior 270°) partial fundopli-
cation is usually performed to prevent postopera-
tive GERD, as well as potentially covering any
undetected mucosal injuries.
56.7.5 Peroral Endoscopic Myotomy
(POEM)
The first human POEM was performed in Japan in
2008. It is a form of natural orifice transluminal
endoscopic surgery. In pediatric population, POEM
was first successfully done in a 3-year-old girl with
Down syndrome and achalasia in 2012 (Maselli et
al. 2012). POEM is the endoscopic equivalent of
surgical myotomy.
POEM is performed under general anesthesia
and EGJ is first identified using a gastroscope
(Fig.  56.4). Saline mixed with indigo carmine
dye is injected submucosally over the anterior
wall of the mid-esophagus at 8 cm proximal to
EGJ. Mucosa is then opened longitudinally. The
submucosal tunnel is created from the mucosal
entry site distally to approximately 3 cm beyond
the EGJ, of which the position is confirmed by
identifying the palisading vessels within the
submucosal layer at the EGJ and bluish discol-
oration of the gastric cardia mucosa on retro-
flexed endoscopic view in the stomach
c
longitudinal muscle is performed. The length of incision
should start from 5 cm above the OGJ and extend to 2 cm
below the junction. (c) After completing the myotomy,
mucosal opening is closed with hemostatic clips
56 Achalasia 737

a b

c d

e f

Fig. 56.5 (a) Mucosal dissection is first performed at
8 cm above the EGJ. (b) The submucosal tunnel (arrow) is
created leaving behind the true lumen of the esophagus
(asterisk) from the mucosal entry site down and approxi-
mately 3 cm beyond the EGJ, of which the position is con-
firmed by (c) identifying the palisading vessels within the
submucosal layer at the EGJ and (d) bluish discoloration
of the gastric cardia mucosa on retroflexed endoscopic
view in the stomach. (e) Intraoperative use of EndoFLIP
(arrow) allows (f) real-time assessment of the distensibil-
ity of EGJ and immediate evaluation of the treatment
effects after POEM
738
(Fig. 56.5). Anterior myotomy cutting the circu-
lar muscle layer while preserving the longitudi-
nal muscle is performed. The length of incision
should start from 5  cm above the EGJ and
extend to 2 cm below the junction. After com-
pleting the myotomy, mucosal opening is closed
with hemostatic clips.
The accurate extent of submucosal tunneling
into the gastric cardia and the length of myot-
omy are the important determining factors to
success of POEM.  Intraoperative use of the
endoluminal functional lumen imaging probe
(EndoFLIP) which allows impedance planime-
try during volumetric distension can evaluate
the EGJ distensibility immediately before and
after the myotomy. The immediate post-myot-
omy EGJ distensibility not only helps to assess
the adequacy of myotomy but also is useful for
predicting treatment responses and gastroesoph-
ageal reflux on follow-­up. In addition, the use of
double endoscope at the end of the procedure
can help ensure complete gastric myotomy by
visualizing the transillumination of the dissect-
ing scope at the submucosal tunnel with retro-
flexed view of gastric cardia using second
endoscope. However, this double-scope transil-
lumination technique may not be feasible in
smaller children or infants as the esophageal
lumen might be too small to accommodate two
endoscopes at the same time (Yeung et al. 2018;
Khashab et al. 2020; Grimes et al. 2016).
More than 93% of pediatric patients experi-
enced improvement or resolution of achalasia
symptoms both short- and long- terms after
POEM (Liu et al. 2020; Chen et al. 2015). There
was a significant reduction in Eckardt score
and LES pressure following POEM (Ren et al.
2017; Corda et al. 2010; Tan et al. 2016; Choné
et al. 2019).  Recent meta-analysis comparing
adult patients undergoing POEM and laparo-
scopic Heller myotomy (LHM) showed that
both procedures resulted in significant
improvement in Eckhart’s score and LES pres-
sure, with POEM requiring significantly less
operative time. Another meta-analysis showed
that POEM is more effective than LHM in
relieving dysphagia in terms of short-term
results (Lee et al. 2019; Schlottmann et al.
F. Yeung et al.
2018). However, some studies raised concern
of higher incidence of pathologic reflux in
POEM compared to LHM (Inoue et al. 2020;
Hu et al. 2015; Nabi et al. 2018). More data is
still required for comparison of POEM and
LHM in pediatric population.
56.8 Complications
56.8.1 Residual Dysphagia
56.8.1.1 Inadequate Myotomy
An insufficient myotomy is one of the most
important and preventable complications of sur-
gery for achalasia and will lead to poor esopha-
geal emptying and recurrent dysphagia. Even
with an incomplete myotomy, patients often have
some symptomatic improvement in their dyspha-
gia, when compared to preoperative state. A post-
operative contrast study provides information on
the adequacy of esophageal emptying. A success-
ful myotomy will show complete/near-complete
(90–100%) emptying by the 5-min film. The most
common cause of an inadequate myotomy is fail-
ure to carry the myotomy sufficiently far onto the
stomach.
56.8.1.2 Surgical Myotomy
For patients who has recurrent dysphagia after sur-
gical myotomy, less invasive interventions such as
pneumatic dilation can be effective as secondary
treatment of residual achalasia. POEM can also be
used instead of a redo Heller myotomy by per-
forming endoscopic myotomy on the posterior
wall of the esophagus, which is at the opposite axis
to the previous myotomy. Salvage POEM has been
shown to have success rates reaching 90%.
56.8.1.3 Poem
For inadequate myotomy after POEM, repeat
endoscopic myotomy on the other side can be
performed. Heller myotomy after anterior POEM
is not effective because it can only address the
same area treated by POEM with an even shorter
length of myotomy and is also associated with
more adverse events similar to those occurring
after redo-Heller myotomy.
56 Achalasia
56.8.2 Suboptimal Myotomy Site
56.8.2.1 Surgical Myotomy
The location of the myotomy may be an important
factor due to the asymmetrical muscle thickness
around the gastroesophageal junction. Consequently,
the best location for a surgical myotomy would
seem to be on the left anterolateral side of the
esophagus extending down across the angle of His
onto the fundus of the stomach. A myotomy in this
location would disrupt the longitudinal and circular
muscular fibers of the esophagus as well as the
oblique gastric fibers while preserving the clasp
fibers along the lesser curve side of the stomach.
56.8.2.2 Poem
For POEM, anterior myotomy consists of inci-
sion of the circular muscle layer while preserv-
ing the longitudinal muscle starting from 5 cm
above the EGJ and extending to 2 cm below the
junction. The adequacy of myotomy can be
assessed intraoperatively by EndoFLIP and
double-scope transillumination technique as
described earlier.
56.8.3 Postoperative Stricture
The development of a scar band at the surgical
myotomy site may lead to impaired esophageal
emptying. It is thus important that the edges of
the surgical myotomy be widely separated.
Performing a Toupet fundoplication may aid in
distracting the muscle layers and has a theoretical
advantage in preventing this type of recurrence.
Scarring after both surgical myotomy and
POEM can be managed with pneumatic dilatation.
56.8.4 Excessive Fundoplication or
Excessively Tight Wrap
Incorporation with Surgical
Myotomy
Long-term follow-up of patients who received a
Nissen 360° fundoplication after surgical myot-
omy showed that the recurrence of dysphagia
was more common. Thus, a complete fundoplica-
tion is inappropriate.
739
56.8.5 Postoperative
Gastroesophageal Reflux
Gastroesophageal reflux can be a significant
problem in a patient who has achalasia because
the aperistaltic esophagus clears refluxed mate-
rial poorly. During the standard procedure of
Heller myotomy, adjacent natural antireflux
structures surrounding the distal esophagus are
dissected. Therefore, surgeons have adopted a
partial fundoplication in addition to the myotomy
as the standard surgical procedure for achalasia.
Another reason to add an antireflux procedure is
that it allows the myotomy to extend well onto
the stomach to be certain it is adequate.
An advantage of the Dor procedure is that
typically no short gastric vessels need to be
divided and no dissection needs to be done
around the hiatus posteriorly. Further, because
the Dor fundoplication is placed over the myot-
omy, it buttresses the mucosa and is useful should
a mucosal perforation occur during the creation
of the myotomy. The Toupet fundoplication, on
the other hand, may lower the incidence of post-
operative recurrent achalasia/stricture.
Symptomatic GERD after POEM has been
reported in 10–20% of patients (Inoue et al. 2020;
Nabi et al. 2018). Based on the current data,
GERD after POEM occurs more frequently than
after Heller myotomy plus partial fundoplication.
The excessive gastric myotomy and incision of
the collar sling fibers and oblique muscle at the
gastric cardia may be the contributive factors to
post-POEM GERD. Therefore, the length of gas-
tric myotomy is recommended to be 2–3 cm and
the direction of myotomy is recommended to be
over anterior or posterior side which could pre-
serve the oblique muscle. Most patients with post-
POEM GERD respond to proton pump inhibitor
and rarely require the need for partial fundoplica-
tion for refractory severe post-­POEM GERD.
56.8.6 Intraoperative Esophageal
Perforation
56.8.6.1 Surgical Myotomy
Perforation is one of the more common complica-
tions during surgical myotomy. The risk increases
740
with previous treatment for achalasia, such as
pneumatic dilatation or botulinum toxin injection.
Mucosal perforations or tears discovered at
the time of operation are amenable to laparo-
scopic repair using interrupted monofilament
sutures. Typically perforations occur at or near
the gastroesophageal junction. The mucosa in
this region is very thin and tears easily, so a small
perforation can quickly extend into a large one.
Mucosal integrity can be reinforced by using fun-
doplication to cover the myotomy.
56.8.6.2 Poem
For POEM, mucosal perforation during submuco-
sal tunneling should be immediately repaired with
clips to prevent leakage of esophageal contents
into the tunnel. If a full-thickness myotomy is
accidentally created at the level of the mucosal
entry, the use of over-the-scope clips has been rec-
ommended to ensure a safe and complete closure.
Postoperatively a water-soluble contrast study is
recommended before initiating oral feeding.
56.8.7 Postoperative Leak
A leak from the myotomy site should be sus-
pected if the patient develops fever, chest pain, or
clinical signs consistent with sepsis postopera-
tively. A water-soluble contrast study should be
part of the routine postoperative investigation as
described previously.
Endoscopy is a sensitive test and may be used
with caution when contrast study fails to confirm
leakage in a persistently septic patient. Most
leaks are small and can be managed with intrave-
nous antibiotics, fasting, nasogastric decompres-
sion, and parenteral nutrition. Surgical drainage
and repair may be necessary if sepsis persists
despite medical treatment.
56.8.8 Long-Term Outcome
In terms of immediate improvement, it is clear
that the treatment modalities of pneumatic dilata-
tion, botulinum toxin, and cardiomyotomy all
report early benefit in around 80–90% or more of
F. Yeung et al.
patients. However, for botulinum toxin, the relief
is temporary. This holds true despite repeated
injections. Further, the majority of patients will
relapse within 6–12 months of the first treatment.
Since both pneumatic dilatation and botuli-
num toxin injections cause inflammation and
fibrosis and have been shown to give rise to dif-
ficulties in the myotomy dissection and subse-
quent lower response rate, they should only be
offered to patients who are high risk for surgery.
Surgical and endoscopic myotomy holds the
most promise for a permanent solution to an ele-
vated LES pressure. The enthusiasm for laparo-
scopic Heller myotomy and POEM is clearly
justified by the significant reduction in morbidity
and hospitalization. Long-term results from vari-
ous studies show that about two-thirds of patients
are considered to have an excellent or good
response after 10 years (Zaninotto et al. 2018).
56.8.8.1 Management of End-Stage
Achalasia
As stated before, rates of success decrease over
time and eventually 10–15% of the patients will
evolve to progressive deterioration of their esoph-
ageal function. About 2–5% of these patients can
end up with an “end-stage” achalasia in adult-
hood, which is defined as a massive dilated and
sigmoid-like esophagus with food retention,
refractory and unresponsive reflux disease, or the
presence of preneoplastic lesions (Hu et al. 2015).
Despite a consensus on the need for esopha-
gectomy in end-stage achalasia, most adult sur-
geons nowadays emphasize that a modified
Heller myotomy should be attempted initially,
even in patients with a dilated and sigmoid-­
shaped esophagus, reserving esophagectomy for
failures. POEM has been successfully performed
in patients with sigmoid type 2 achalasia where
the lumen of the tortuous esophagus turns
upward. However, more technical challenges
should be anticipated and some surgeons recom-
mend posterior or bilateral POEM to increase the
success rates.
After esophagectomy, reconstruction can be
performed in different manners: with gastric,
colon, or jejunal interposition. At present, from
the published literature, gastric interposition is
56 Achalasia
the first choice, but vascular supply to the stom-
ach and significant reflux to the esophageal rem-
nant affect its success as an ideal transplant.
56.8.8.2 Screening for Carcinoma
A potential long-term complication of achalasia is
the development of esophageal cancer in adult-
hood. The risk of esophageal carcinoma varies
from 10 to 50 times in patients with achalasia com-
pared with the general population. This is attributed
to the increased bacterial growth and chemical irri-
tation from the retention and decomposition of
food and saliva resulting in chronic hyperplastic
esophagitis, dysplasia, and eventually malignant
transformation of esophageal epithelial cells.
However, most cancers were diagnosed in an
advanced state as symptoms of esophageal carci-
noma in achalasia patients are often misinter-
preted as merely symptoms of achalasia and
conventional white light endoscopy of the esoph-
agus is not sensitive for the detection of dyspla-
sia. The use of high-resolution endoscopy and
chromoendoscopy with Lugol’s staining has sig-
nificantly improved the sensitivity to detect pre-
malignant lesions. Hence, regular endoscopic
carcinoma surveillance using Lugol’s staining is
advised in patients with long-standing achalasia
like 10 years after initial treatment.
56.9 Conclusion
We can conclude that laparoscopic myotomy and
POEM should be the treatment of choice in
younger patients. As there is a lifelong risk of
malignancy in patients with achalasia, long-term
follow-up and transitional care by adult surgeons
are important with regular carcinoma surveil-
lance in the long term.
References
Asch MJ, Liebman W, Lachman RS, Moore TC (1974)
Esophageal achalasia: diagnosis and cardiomyotomy
in a newborn infant. J Pediatr Surg 9(6):911–912.
Ballantine TV, Fitzgerald JF, Grosfeld JL (1980)
Transabdominal esophagomyotomy for achalasia in
children. J Pediatr Surg 15(4):457–461.
741
Boeckxstaens GE, Annese V, des Varannes SB et al (2011)
Pneumatic dilation versus laparoscopic Heller’s
myotomy for idiopathic achalasia. N Engl J Med
364(19):1807–1816
Boeckxstaens GE, Zaninotto G, Richter JE (2014)
Achalasia. Lancet 383(9911):83–93
Chen WF, Li QL, Zhou PH et al (2015) Long-term out-
comes of peroral endoscopic myotomy for achalasia in
pediatric patients: a prospective, single-center study.
Gastrointest Endosc 81(1):91–100
Choné A, Familiari P, von Rahden B et  al (2019)
Multicenter evaluation of clinical efficacy and safety
of per-oral endoscopic myotomy in children. J Pediatr
Gastroenterol Nutr 69(5):523–527
Corda L, Pacilli M, Clarke S et al (2010) Laparoscopic
oesophageal cardiomyotomy without fundoplication
in children with achalasia: a 10-year experience: a
retrospective review of the results of laparoscopic
oesophageal cardiomyotomy without an anti-reflux
procedure in children with achalasia. Surg Endosc
24:40–44
Eckardt VF, Aignherr C, Bernhard G (1992) Predictors of
outcome in patients with achalasia treated by pneu-
matic dilation. Gastroenterology 103(6):1732–1738
Franklin AL, Petrosyan M, Kane TD (2014) Childhood
achalasia: a comprehensive review of disease, diagno-
sis and therapeutic management. World J Gastrointest
Endosc 6(4):105–111
Goneidy A, Cory-Wright J, Zhu L et  al (2020) Surgical
management of esophageal achalasia in pediatrics: a
systematic review. Eur J Pediatr Surg 30(1):13–20
Grimes KL, Inoue H, Onimaru M et  al (2016) Double-­
scope per oral endoscopic myotomy (POEM): a pro-
spective randomized controlled trial. Surg Endosc
30:1344–1351
Holcomb GW III, Richards WO, Riedel BD (1996)
Laparoscopic esophagomyotomy for achalasia in chil-
dren. J Pediatr Surg 31(5):716–718
Horgan S, Galvani C, Gorodner MV et  al (2005)
Robotic-­ assisted Heller myotomy versus laparo-
scopic Heller myotomy for the treatment of esopha-
geal achalasia: multicenter study. J Gastrointest Surg
9(8):1020–1030
Hu JW, Li QL, Zhou PH et al (2015) Peroral endoscopic
myotomy for advanced achalasia with sigmoid-shaped
esophagus: long-term outcomes from a prospective,
single-center study. Surg Endosc 29:2841–2850
Inoue H, Shiwaku H, Kobayashi Y et al (2020) Statement
for gastroesophageal reflux disease after peroral endo-
scopic myotomy from an international multicenter
experience. Esophagus 17:3–10
Kahrilas PJ, Bredenoord AJ, Fox M et  al (2015) The
Chicago classification of esophageal motility disor-
ders, v3.0. Neurogastroenterol Motil 27(2):160–174
Khashab MA, Sanaei O, Rivory J et  al (2020) Peroral
endoscopic myotomy: anterior versus posterior
approach: a randomized single-blinded clinical trial.
Gastrointest Endosc 91(2):288–297
Lee Y, Brar K, Doumouras AG et al (2019) Peroral endo-
scopic myotomy (POEM) for the treatment of pediat-
742
ric achalasia: a systematic review and meta-analysis.
Surg Endosc 33(6):1710–1720
Liu Z, Wang Y, Fang Y et  al (2020) Short-term safety
and efficacy of peroral endoscopic myotomy for the
treatment of achalasia in children. J Gastroenterol
55:159–168
Maselli R, Inoue H, Misawa M et al (2012) Peroral endo-
scopic myotomy (POEM) in a 3-year-old girl with
severe growth retardation, achalasia, and Down syn-
drome. Endoscopy 44(Suppl 2)
Nabi Z, Reddy DN, Ramchandani M (2018) Adverse
events during and after per-oral endoscopic myotomy:
prevention, diagnosis, and management. Gastrointest
Endosc 87(1):4–17
Ren Y, Tang X, Chen Y et al (2017) Pre-treatment Eckardt
score is a simple factor for predicting one-year peroral
endoscopic myotomy failure in patients with achala-
sia. Surg Endosc 31(8):3234–3241
F. Yeung et al.
Rothenberg SS, Partrick DA, Bealer JF, Chang JH (2001)
Evaluation of minimally invasive approaches to acha-
lasia in children. J Pediatr Surg 36(5):808–810
Schlottmann F, Luckett DJ, Fine J et  al (2018)
Laparoscopic Heller myotomy versus peroral endo-
scopic myotomy (POEM) for achalasia: a systematic
review and meta-analysis. Ann Surg 267:451–460
Tan Y, Zhu H, Li C et  al (2016) Comparison of peroral
endoscopic myotomy and endoscopic balloon dilation
for primary treatment of pediatric achalasia. J Pediatr
Surg 51:1613–1618
Yeung F, Wong IYH, Chung PHY et al (2018) Peroral endo-
scopic myotomy with EndoFLIP and double-­endoscope:
novel techniques for achalasia in pediatric population. J
Laparoendosc Adv Surg Tech A 28(3):343–347
Zaninotto G, Bennett IC, Boeckxstaens G et  al (2018)
The 2018 ISDE achalasia guidelines. Dis Esophagus
31:1–29
 Esophageal Perforations
and Caustic Injuries in Children
Shilpa Sharma and Devendra K. Gupta
57.1 Introduction
Perforation of the esophagus is a life-threatening
condition and requires timely diagnosis and man-
agement. The complications of mediastinitis can
be prevented if a conservative approach is
adopted to keep the esophagus dry and drain the
chest collection. Caustic injuries require the child
to be nil per mouth, and, in most cases, a gastros-
tomy is required for drainage and feeding.
57.2 Esophageal Physiology
and Anatomy
The esophagus is derived from the foregut during
the fourth week of development. It comprises of
longitudinal and circular muscle fibers sur-
rounded by collagen and elastic fibers. The
“pinch-cock” action of the cricopharyngeal is
pronounced in the newborn. The vascular supply
is derived from branches of the thyroid arteries,
aorta, bronchial arteries, left gastric artery, and
left inferior phrenic artery.
S. Sharma (*)
Department of Pediatric Surgery, All India Institute of
Medical Sciences, New Delhi, India
D. K. Gupta
Super Speciality Pediatric Hospital and Post Graduate
Teaching Institute, Noida, India
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_57
57
The esophagus is located to the left of the mid-
line at the level of the first thoracic vertebra, right
of the midline at the level of the sixth thoracic
vertebra, and left of the midline again at the level
of the tenth thoracic vertebra. Thus, the esopha-
gus makes a reverse “S” shape in front of the ver-
tebral column. In almost everyone, there are four
classic narrowings in the esophagus. These
include (a) the beginning point at the oropharyn-
geal muscle [body of sixth cervical vertebra]
called the upper esophageal sphincter, (b) the
plane that is located at the superior border of the
sternum, (c) crossing of the aortic arch over the
esophagus [fourth thoracic vertebra], and (d)
crossing point of the esophagus and left main
bronchium [fifth thoracic vertebra].
57.3 Etiology of Esophageal
Perforation
The various causes of esophageal perforation
include:
1. Spontaneous—Boerhaave’s syndrome
2. Iatrogenic
3. Esophagoscopy with dilatation
4. Button battery ingestion
5. Foreign body ingestion such as coin
ingestion
6. Nasogastric tube placement
743
744
57.4 Esophageal Perforation
in the Newborn
Perforation of the esophagus is common in the
newborn, especially in premature babies (Maeda
et al. 2019). The most common cause is acciden-
tal during insertion of nasogastric tube. The
injury can also be in the pharyngeal area, and oral
inspection is important to identify the injury. It is
important to examine the skiagram chest care-
fully for any suspicion of pharyngeal or
­esophageal injury (Wolf et  al. 2018). The most
important finding is that the feeding tube does not
reach the stomach. The normal position of an
orogastric or nasogastric tube is towards the left
of the trachea, with a curve to the left at the lower
mediastinum and no air leaks (Wolf et al. 2018).
On the other hand, if there is an injury, the feed-
ing tube is in an abnormal position being right to
the trachea with a straight termination and curve
to the right towards the right pleural space. There
may also be evidence of air in the mediastinum
and right pneumothorax (Wolf et al. 2018).
There are three types of injury: (1) posterior
pharyngeal rupture, (2) noncomplicated esopha-
geal injury with esophageal false lumen, and (3)
complicated esophageal rupture with penetration
into the right pleural space (Wolf et  al. 2018;
Krasna et  al. 1987; Hesketh et  al. 2015). The
involvement of the right pleural space is second-
ary to the close proximity of the azygoesopha-
geal recess and the right medial parietal pleura.
In rare instances, in cases of undiagnosed tra-
cheal agenesis, attempted intubation may also
lead to perforation of the esophagus.
Fortunately, most of the cases can be man-
aged by nonoperative management (Krasna
et  al. 1987; Onwuka et  al. 2016). However,
timely identification is essential. It helps if oro-
gastric or nasogastric insertion of a feeding tube
is successful. A word of caution applies;
although the nonoperative management may be
a safe initial strategy in the neonatal setting,
more aggressive interventions may ultimately
be required (Hesketh et al. 2015).
S. Sharma and D. K. Gupta
57.5 Boerhaave’s Syndrome
Spontaneous rupture of the esophagus occurs
when intraesophageal pressure increases against
a closed cricopharyngeus during excessive retch-
ing or vomiting. The incidence of Boerhaave’s
syndrome is extremely low in children
(Mohammed et al. 2016; Janjua 1997; Chikkappa
et  al. 2009). The 28th case was reported in a
4-month-old infant, the youngest patient to have
been diagnosed with Boerhaave’s syndrome in
the literature and managed conservatively by use
of underwater seal drainage (Janjua 1997). The
pathophysiology involves esophageal barotrauma
resulting from increased intraluminal pressure
during excessive emesis, which is transmitted to
the esophagus against a closed cricopharyngeus
(Sabanathan et  al. 1994). The same perforation
could happen after significant blunt trauma to the
chest wall, especially after crushing accidents.
The spontaneous nature of rupture following
vomiting is characteristic. The most common site
of rupture is the left posterolateral wall of the
lower third of the esophagus, followed by the
subdiaphragmatic area. Early detection of the
respiratory embarrassment is critical.
The clinical presentation is described by
Mackler’s triad of symptoms of vomiting, chest
pain, and subcutaneous emphysema associated
with clinical and radiological findings of pneu-
mothorax and hydropneumothorax (Janjua
1997). However, history may not always be reli-
able, especially in children.
Although rare, if left untreated, it can cause
life-threatening complications such as mediasti-
nitis and sepsis (Mohammed et al. 2016).
Fluid resuscitation and broad-spectrum antibi-
otics are the mainstay of management. Most
require a surgical intervention with primary
repair of the esophageal tear especially in com-
plicated cases with pyomediastinum or sepsis.
Children tend to do well compared with adult
counterparts. Boerhaave’s syndrome has a 70%
survival in surgically treated cases and mortality
of 100% in untreated cases (Curci and Horman
57  Esophageal Perforations and Caustic Injuries in Children
1976). Mortality and morbidity are increased in
direct proportion to the time between diagnosis
and appropriate surgical intervention. Sepsis,
hypovolemia, and shock are the predominant
causes of morbidity and mortality in Boerhaave’s
syndrome (Curci and Horman 1976).
57.6 Button Battery Ingestion
With advances in the digital world, it has become
very common to encounter button battery inges-
tion. These are corrosive and damage the esopha-
geal mucosa.
The authors encountered 20 cases of accidental
ingestion of button battery over a 3-year period
from 2014 to 2016. The mean age at presentation
was 47  months (13  months–8  years). The mean
duration of exposure was 30 h (from ingestion until
removal). The common symptoms included dys-
phagia and vomiting after feeds. The common
sources included the television remote and toys.
Most were witnessed accidents and patients were
brought to hospital early. However, even discharged
batteries can lead to charring of the esophageal
mucosa. We encountered burns and charring were
found in the esophagus in 70% of cases. Most bat-
teries had a diameter of 20 mm. The step pattern of
the battery helps to distinguish it from a coin.
The upper two-thirds of the esophagus is the
most common site for impaction of these batter-
ies. The percentage of the batteries that were
impacted in the anterior wall is 85%. When the
accident is not witnessed, the duration of expo-
sure becomes longer. We had a case presenting
even after 14 days of exposure. In such cases, seri-
ous complications such as esophageal perforation
may be encountered. We had such a complication
in three patients. Two of these had succumbed to
major complications such as a tracheoesophageal
fistula and aorto-esophageal fistula. Symptoms
like cough and mild pain may occur during swal-
lowing after endoscopic removal.
A follow-up contrast study is important to
document a normal esophagus without any perfo-
ration or stricture.
745
57.7 Diagnosis of Esophageal
Perforation
The diagnosis can be made on the basis of his-
tory, clinical symptoms, and radiological
imaging.
In the neonate, excessive salivation, respira-
tory distress, and, occasionally, cervical swelling
and crepitus may occur. Failure of passage of the
nasogastric tube could be the first manifestation
of cervical perforation.
In cases of thoracic perforations, the patient
usually presents with respiratory distress, tachy-
cardia, and sweating. Tension pneumothorax is a
clinical diagnosis and should be relieved with a
chest tube immediately. If not promptly relieved,
shock and death may occur.
A chest x-ray is essential to rule out any perfo-
ration after any dilatation procedure for stricture
esophagus. In cases of perforation, there may be
pneumo-mediastinum or pneumothorax
(Fig.  57.1). After routine dilatation, the patient
should be observed for at least 6 h, and if there is
any deterioration in their general condition, a
delayed perforation should be suspected and
repeat rapid x-ray of the chest should be obtained.
Similarly, after removal of difficult impacted for-
eign bodies such as button batteries, one should
get a check x-ray (Fig. 57.2a and b). If there is
Fig. 57.1  Pneumothorax due to esophageal perforation
746
a b
Fig. 57.2 (a) Eroded esophagus with embedded button battery. (b) Lateral skiagram showing the step pattern of the
button battery. (c) Contrast esophagogram outlining the leak
Fig. 57.3  Perforation of the esophagus with contrast in
the thoracic cavity
S. Sharma and D. K. Gupta
c
any doubt, a contrast esophagogram should be
done (Fig. 57.2c). A widened mediastinum on the
chest radiograph is also very suspicious of esoph-
ageal perforation. Sometimes the first sign is cer-
vicofacial emphysema.
Upper esophageal perforations occur on the
right side and may lead to a right hydropneumo-
thorax. Mid and lower esophageal perforations
present on the left side. After dilatation, there
may be small linear mucosal tears, which might
not be symptomatic, but with insufflations of air
during endoscopy, the tear gets bigger and trans-
mural perforation occurs into the thoracic cavity.
A simple chest x-ray will show the pneumo-
thorax. After insertion of a chest tube and
improvement in the patient’s condition, a water-­
soluble radio-opaque contrast may be instilled in
the esophagus to identify the precise location of
the leak (Fig. 57.3).
Often, the minor leak depicted by moderate
pneumothorax heals well and a contrast study
outlines the healed esophagus (Fig. 57.4a and b).
57  Esophageal Perforations and Caustic Injuries in Children 747

a b

Fig. 57.4 (a) Mild pneumothorax due to esophageal perforation. (b) Contrast study outlining the healed esophagus

57.8 Differential Diagnosis A passage of the nasogastric tube to the stom-

Tracheobronchial injury and lung parenchymal
disease remain the two differential diagnoses.
57.9 Management of Esophageal
Perforation
The first step in the management is local exami-
nation. In premature infants, the injury is usually
in the lower pharynx or cervical esophagus.
Conservative management, guided by the
patient’s clinical course, with antibiotics and
nutritional support is a safe and effective treat-
ment for esophageal perforations in children
(Garey et al. 2010).
ach under fluoroscopic guidance is one of the
important steps in all esophageal perforations. It
allows deflation of the stomach, thereby decreas-
ing gastroesophageal reflux. Later feedings could
be started through the tube with elevation of the
patient’s head. If reflux of enteral nutrition occurs
and comes through the chest tube, then advanc-
ing the enteral feeding tube to the duodenum or
the first part of the jejunum should be performed
(Hamza 2009). An intercostal chest drain should
be placed on the side of the leak that is usually
the right side.
Chest drainage and nasogastric tube in the
stomach are usually sufficient to heal most perfo-
rations in the neonates or prematures. Cervical
injuries rarely need drainage and heal within
748
7–10  days. Cervical injury generally does not
lead to any esophageal stenosis and further fol-
low-­up is not needed.
Injuries to the thoracic esophagus after instru-
mentation or trauma also heal after chest drain-
age (Hamza 2009). In cases of tension
pneumothorax, a chest tube should be inserted
immediately followed by reassessment of the
patient.
In children, primary repair with tissue rein-
forcement is unnecessary. However, in some
traumatic cases, exploration and surgical closure
of the tear should be done if the patient presents
early enough and has a large perforation. In cases
after esophageal dilatation, the tear is usually
­linear and healing occurs after 10–15  days. In
late presenting cases or after failure of conserva-
tive measures, exclusion of the esophagus by cer-
vical esophagostomy and a gastrostomy should
be done. A lateral esophagostomy for drainage of
the saliva and a jejunostomy for feeding or a gas-
trostomy are other alternatives that could pre-
serve the native esophagus.
Total parenteral nutrition is started early to
take of caloric needs till the time of enteral feed-
ing is reestablished.
If the perforation occurs during a dilatation,
dilatations should not be resumed until there is
verification of healing by esophageal contrast
study. Dilatation could be restarted, if needed
after a period of healing, to the esophagus and
with very careful under image control dilatation.
Some of these cases will eventually need esopha-
geal replacement.
In cases associated with dangerous mecha-
nism of trauma, severe injuries, and a tracheo-
esophageal communication, a diversion is
preferable followed by esophageal replacement
later (Sharma et al. 2018).
57.10 Caustic Injuries
of the Esophagus
Caustic ingestion by children leading to caustic
injuries of the esophagus and stomach is still a
serious social issue and prevalent in the develop-
S. Sharma and D. K. Gupta
ing countries due to lack of awareness. Ingestion
of caustic agents leads to injuries in the esopha-
gus, pharynx, larynx, and mouth, which often
lead to scarring, stricture formation adverse
effects on the gastrointestinal and respiratory
tracts and even death. The most common cause of
caustic injuries is accidental.
57.11 Epidemiology
Caustic ingestion is still prevalent in some coun-
tries such as India, Egypt, Turkey, and South
Africa and some eastern European countries. The
most common age group is less than 5 years of
age, although we have seen this in several adoles-
cents too due to carelessness. The majority of
cases occur in children between 1 and 4 years of
age. The injury can also rarely occur in adoles-
cents due to suicidal attempts.
It is more common in boys than girls, with an
incidence of 3:1, and is due to accidental swal-
lowing of caustic material due to its resemblance
to milk or water (Hamza et al. 2003).
Caustic materials are widely used for house-
hold or industrial cleaning. When the concentra-
tion of their active components exceeds 10% (or
2% by weight of free chemical), they become
dangerous and should have child-resistant
packaging.
Most accidents are due to alkaline agents used
to clean the house, such as bleach products, oven
cleaners, and soda (Rafeey et al. 2016). The new
generation of cleaning materials used in dish-
washers, with very high pH levels, is especially
destructive to the esophageal wall. In cases of
caustic ingestion, pooled data suggest that house-
hold bleach accounts for 30–40% of caustic
ingestions and laundry detergents for 20%, and
acids and alkalies from products such as oven,
toilet, tile, drain, and various cleaners account for
50% or more.
The causative agents are either alkalies or
acids and they are usually in liquid form, but
occasionally, we have seen some injuries due to
swallowing of crystal forms. The alkalies are
usually potassium hydroxide (Potash) or sodium
57  Esophageal Perforations and Caustic Injuries in Children 749

hydroxide. The acids are usually hydrochloric 57.12 Pathology

acid or sulfuric acid. Drain cleaners, which are
commonly available, also contain acids. When
these substances are not placed in childproof
containers, they are extremely dangerous.
Lithium batteries can also cause injuries to the
esophagus, especially if they are big enough to
lodge in the cervical esophagus. If not promptly
removed, it may lead to leakage of contents that
usually causes severe esophageal injury with
perforation.
As small children start to become ambulatory,
they are curious to explore the surroundings. As a
result, they gain access to cabinets or areas where
cleaning products are stored for an ingestion to
occur. Although they account for 51% of toxic
exposures, children less than age 6 years account
for only 2.3% of fatalities due to toxic exposures
primarily due to the accidental nature of their
ingestions. In more than 90% of cases, the expo-
sure is to a single substance and the most expo-
sures occur at home.
The injury depends on the amount swallowed, the
concentration, and the type of agent. Alkalies
cause liquefactive necrosis of the esophagus, and
it usually most profoundly affects the areas of the
esophagus that have slower transit of the bolus
during swallowing. These include the cervical
part, at the level of the carina, and just above the
diaphragm. The injury may penetrate through the
muscle layer and extend into the mediastinum,
especially when high concentrations of the caus-
tic agent are ingested. Healing occurs by resorp-
tion of the necrotic tissue and fibrosis. This leads
to stricture formation, which is completed at
around 6 weeks following the injury (Fig. 57.5a).
Acids are usually neutralized by the saliva and
pass rapidly into the stomach. The acids may
induce severe pyloric spasm that retains the
ingested material in the antral area. This then
leads to coagulative necrosis. Very severe injuries
lead to perforation and necrosis, and even milder

a b

Fig. 57.5 (a) Caustic esophageal stricture. (b) Pyloric obstruction due to acid injury
750
acid ingestions may lead to pyloric and antral
obstruction usually within 2–3 weeks (Fig. 57.5b).
Acids may also lead to esophageal injury but this
is less frequent than injuries to the stomach.
Ingestion of very concentrated acids leads to
necrosis of the whole esophagus and stomach.
57.13 Diagnosis
There is usually a positive history of corrosive
intake from the parents. If it is witnessed, details
of the agent may be available. If no history is
available but there is hyperemia of the mouth
with excessive salivation, suspicion of caustic
ingestion should be placed at the top of a differ-
ential diagnosis. In cases of crystal ingestion, the
crystals may adhere to the mucus membrane of
the mouth resulting in ulcers and bleeding.
Airway injuries may also occur in some severe
cases and tracheostomy may be needed for air-
way stabilization. First responders should avoid
dilution of the caustic agent by giving the patient
water. Furthermore, any attempts to induce vom-
iting may expose the esophagus to a second
injury. Caustic agents may induce pylorospasm
and thus augment the risk of vomiting and further
esophageal injury.
Intravenous fluids, antibiotics, antacids, and
urgent endoscopy are important tools as first aid
measures. Steroids do not prove to be effective in
the management of acute injuries of the esopha-
gus, but it is very useful in airway edema and
injuries. Fiber-optic endoscopy gives an idea
about the degree of injury and the extent of
lesions; some centers manage without endoscopy
in mild alkali cases, as the management will not
differ in mild and superficial injuries.
57.14 Management
Acids usually cause the most severe cases, and if
endoscopy shows necrosis of the esophagus and
stomach, then total esophagogastrectomy is
needed urgently but fortunately; this is very rare.
For the usual injuries, patients are kept on I.V.
fluids until they can swallow fluids orally, which
S. Sharma and D. K. Gupta
usually occurs after healing of the mouth lesions.
Children who cannot swallow are kept on TPN
for 2–3 weeks, and then a gastrostomy is estab-
lished for enteral feeding and for retrograde
esophageal dilatation.
In cases of pyloric obstruction, several meth-
ods are available: antrectomy, pyloroplasty, and
gastro-jejunostomy depending on the lesion and
extent of fibrosis.
A dilatation program using bouginage remains
the mainstay of treatment. Dilatation of caustic
strictures can be started as early as the third week,
but better results have been achieved if dilata-
tions start after the sixth week when complete
healing is established. Balloon dilators, retro-
grade Tucker dilators through the stomach, or
antegrade Savary-Gilliard dilators have all been
used successfully, but the best results in our
hands has been with Savary antegrade dilators.
Pneumatic balloon dilators and stenting proce-
dures are also in vogue recently.
Dilatation is repeated every 2 weeks for a total
of 6 weeks, and then the patient is followed up by
a contrast study. If swallowing of meat and solid
food is satisfactory and the contrast esophago-
gram findings are encouraging, then the spacing
of the dilatation procedures can be done and
complete cure can occur in about 6  months to
1 year. In cases of localized strictures, local injec-
tions of triamcinolone may lead to less frequent
dilatation. Likewise, application of Mitomycin C
(an inhibitor of fibroblast proliferation) has also
been used with success. Intralesional steroid or
Mitomycin C may decrease the dilatations
required for severe strictures, although long-term
effects are unknown (Millar et al. 2021). Antacids
(proton pump inhibitors) should be used in these
patients as injury of the cardiac sphincter with
reflux usually occurs due to the caustic effect.
Medical therapies such as oral antibiotics, sys-
temic steroids, and H2 blockers; surgical therapy,
such as esophageal dilatation; and conservative
treatments were used.
Contraindications to dilatation include tortu-
ous and very long strictures or multiple strictures
and cases of tracheoesophageal fistula.
Esophageal replacement is required when
there is failure of dilatation because there is no
57  Esophageal Perforations and Caustic Injuries in Children
improvement in swallowing. Discontinuation of
dilatation is indicated when re-stricture forma-
tion is too rapid, there is significant proximal
dilatation of the esophagus with pouch forma-
tion, and in cases of perforation.
Esophageal sleeve resection can be done for
very localized strictures, but this is very uncom-
mon as most of them respond to dilatation.
Replacement of the esophagus is indicated in
cases of failure or contraindication of dilatation.
The stomach, gastric tube, jejunum, and the colon
are all substitutes for the injured esophagus, but
the colon has proved to be better in cases of cor-
rosive injuries. Replacement is done when the
child’s physiological condition permits. The less
delay the better as swallowing solid food is
essential for mouth and teeth development in
children.
The following effects have been seen:
1. Esophageal damage
2. Gastroesophageal reflux
3. Gastric ulceration
4. Aspiration pneumonia and laryngeal sore in
the respiratory system
5. Malignant transformation reported during
long-term follow-up
6. Esophageal stricture
57.15 Prevention of Caustic
Injuries
The following are recommended to prevent caus-
tic injuries (Rafeey et al. 2016):
1. Large amounts of detergent must not be kept
at the home.
2. Cleaning agents should be kept away from the
kitchen.
3. Chemical substances should be placed at a
safe place, in a locked cabinet, and in the
upper shelves.
4. Chemicals should be immediately put in its
place after utilization.
5. Chemicals should be kept in the labelled con-
tainers with tight door.
751
6. Chemicals should not be stored in food
containers.
7. Concentrated chemicals should be kept in
children-safe packaging.
57.16 Conclusion
Esophageal perforation and caustic injury to the
esophagus are life-threatening conditions. Timely
diagnosis and immediate management are essen-
tial to avoid any catastrophes. Caustic ingestion
continues to be a significant problem worldwide,
especially in developing countries and particu-
larly in children under the age of 6  years.
Increased public awareness of the problem of
caustic ingestion by wide publicity in the media,
reporting instances of ingestion, and developing
educational packages for schools and workplaces
should prevent these injuries.
References
Chikkappa MG, Morrison C, Lowe A, Gorman S, Antrum
R, Gokhale J (2009) Paediatric Boerhaave’s syndrome:
a case report and review of the literature. Cases J 2:1–4
Curci JJ, Horman MJ (1976) Boerhaave’s syndrome: the
importance of early diagnosis and treatment. Ann Surg
183:401–408
Garey CL, Laituri CA, Kaye AJ et al (2010) Esophageal
perforation in children: a review of one institution’s
experience. J Surg Res 164:13–17
Hamza AF (2009) Chapter 36: Esophageal perforations
and caustic injuries in children. In: Puri P, Höllwarth
M (eds) Pediatric surgery: diagnosis and management.
Springer-Verlag, Berlin Heidelberg, pp  353–359.
https://doi.org/10.1007/978-­3-­540-­69560-­8_36
Hamza AF, Abdelhay S, Sherif H et  al (2003) Caustic
esophageal strictures in children: 30 years experience.
Overseas guest lecture. J Pediatr Surg 38:828–833
Hesketh AJ, Behr CA, Soffer SZ, Hong AR, Glick RD
(2015) Neonatal esophageal perforation: nonoperative
management. J Surg Res 198:1–6
Janjua KJ (1997) Boerhaave’s syndrome. Postgrad Med J
73(859):265–270
Krasna IH, Rosenfeld D, Benjamin BG, Klein G, Hiatt M,
Hegyi T (1987) Esophageal perforation in the neonate:
an emerging problem in the newborn nursery. J Pediatr
Surg 22:784–790
Maeda T, Tachibana T, Harada T (2019) Esophageal per-
foration with orogastric tube malposition in low-birth-­
weight infant. Clin Case Rep 7:1794–1795
752
Millar AJW, Numanoglu A, Cox S (2021) Caustic inju-
ries of the esophagus. In: Puri P (ed) Pediatric surgery.
Springer
Mohammed A, Saneeq S, Kolakkadan K, Amrutha K
(2016) Paediatric Boerhaave’s syndrome: a rare but
crucial diagnosis. Egypt J Otolaryngol 32:335
Onwuka EA, Saadai P, Boomer LA, Nwomeh BC (2016)
Nonoperative management of esophageal perforations
in the newborn. J Surg Res 205:102–107
Rafeey M, Ghojazadeh M, Sheikhi S, Vahedi L (2016)
Caustic ingestion in children: a systematic review and
meta-analysis. J Caring Sci 5:251–265
S. Sharma and D. K. Gupta
Sabanathan S, Eng J, Richardson J (1994) Surgical man-
agement of intrathoracic oesophageal rupture. Br J
Surg 81:863–865
Sharma S, Mishra B, Gupta A, Soni KD, Aggarwal R,
Kumar S (2018) Challenges in management of pedi-
atric life-threatening neck and chest trauma. J Indian
Assoc Pediatr Surg 23:10–15
Wolf JA, Myers EH, Remon JI, Blumfield E (2018)
Imaging findings of iatrogenic pharyngeal and esopha-
geal injuries in neonates. Pediatr Radiol 48:1806–1813
 Gastroesophageal Reflux Disease
Michael E. Höllwarth and Valeria Solari
58.1 Introduction
Gastroesophageal reflux (GER) is the unrecog-
nized backflow of gastric contents into the esoph-
agus. It is a physiological phenomenon and
occurs in persons of all age groups, more fre-
quently, after the ingestion of liquid food (soup)
or other fluids (soft drinks, coffee, tea, milk, alco-
hol, etc.). A pathological reflux or gastroesopha-
geal reflux disease (GERD) is considered to exist
when the number and/or duration of reflux events
exceeds normal ranges and causes troublesome
symptoms or complications. Recurrent and
chronic exposure of the esophageal epithelia to
the strong acidity of the refluxed material may
cause esophagitis, esophageal stenosis, or meta-
plasia within the mucosa, sometimes leading to
Barrett’s esophagus, a premalignant disease of
esophageal adenocarcinoma. If the refluxed
material reaches the pharynx, it causes laryngitis
and can be recurrently aspirated and lead to reac-
tive airway diseases. Furthermore, gastroesopha-
geal reflux is commonly seen in children with
neurologic dysfunction. These children often
M. E. Höllwarth (*)
University Clinic of Paediatric and Adolescent
Surgery, Medical University of Graz, Graz, Austria
e-mail: michael.hoellwarth@medunigraz.at
V. Solari
Department of Paediatric Surgery, Klinik Donaustadt,
Vienna, Austria
e-mail: valeria.solari@gesundheitsverbund.at
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_58
58
have associated swallowing disorders, failure to
thrive, recurrent aspiration, spasticity, increased
intraabdominal pressure, and central mechanisms
for inducing gagging and retching. Some of these
symptoms reflect pain due to severe esophagitis.
Therefore, early diagnosis of a GERD and age-­
specific appropriate therapy is essential to pre-
vent further complications.
58.2 History
Spitting up and vomiting are well known for cen-
turies as typical symptoms of reflux in neonates
and infants. Since it resolves spontaneously in
most babies, it was regarded as harmless or even
beneficial: there is an old German saying
“Speihkinder sind Gedeihkinder,” which means
that spitting up infants develop very well. In the
last 100 years, there has been enormous progress
in the understanding of the physiological and
pathological mechanisms of gastroesophageal
function (Modlin et  al. 2004; Dent 2006). In
1828, Charles-Michel Billard published the first
case of esophagitis in a child after autopsy
(Stylopoulos and Rattner 2005). Significant
progress was possible at the beginning of the
twentieth century, after the introduction of X-ray
imaging and barium contrast studies. The physi-
ologist Walter B.  Cannon showed for the first
time in history, in esophageal X-ray experiments
with cats, that repeated regurgitations of contrast
753
754
material occur from the stomach into the esopha-
gus and are followed by a peristaltic wave push-
ing the barium back into the stomach, “… the
cardia again relaxed and it rushed out, only to be
restored to the stomach by another peristaltic
wave. Thus, the process continued. The peristal-
tic wave was seldom started by voluntary degluti-
tion, but was of secondary order, stimulated by
the presence of material in the esophagus”
(Cannon 1911). Cannon also described a cardiac
sphincter, “The thickened band of circular
smooth muscle at the junction of the esophagus
with the stomach—the cardiac sphincter, or car-
dia—has the function of preventing the passage
of material from the stomach back into the esoph-
agus.” However, the existence of a cardiac
sphincter was denied for decades, mainly due to
the inability to show a sphincter-like structure at
the lower end of the esophagus in humans.
Finally, in fine and detailed anatomical studies,
Liebermann-Meffert demonstrated the presence
of a true sphincter at the lower end of the esopha-
gus (Liebermann-Mefffert et al. 1979). Palmer in
1968 and Cohen in 1975 showed that pathologi-
cal reflux is not connected with a hiatal hernia but
with an insufficient reflux barrier (Palmer 1954;
Cohen 1975).
With the introduction of intraluminal manom-
etry in the 1950s, Code was able to prove a high-­
pressure zone at the lower esophagus in adults
(Code et  al. 1956; Ingelfinger 1958). Combined
cineradiographic and manometric studies in six
infants confirmed the presence of a high-pressure
zone between the stomach and the esophagus
(Carré and Astley 1958). The early results of
manometric studies in newborn and infants
showed a very low lower esophageal sphincter
(LES) tone, which seemed to be consistent with
the high incidence of regurgitation and vomiting
in this age group. However, more sophisticated
investigations in the late 1970s, using a low com-
pliance manometric device, showed that the LES
tone is normal even in newborns, and the higher
incidence of reflux episodes is not caused by a
missing sphincter tone but by transient LES relax-
ations (TLESR), resulting in a common space
between the stomach and the esophagus (Höllwarth
1979). Furthermore, these studies supported previ-
M. E. Höllwarth and V. Solari
ous descriptions by Cannon from his animal
experiments that the refluxed material is pushed
back into the stomach by a secondary peristaltic
wave generated in the esophagus. Later, it was also
confirmed in adults that virtually all reflux epi-
sodes are caused—or made possible—by TLESR
(Dent et al. 1980). Today, 24-h pH monitoring and
impedance studies allow a sophisticated evalua-
tion of the esophageal function, and the barium
swallow a detailed anatomic study. Endoscopy
with biopsy and histology allows the diagnosis if a
reflux pattern already causes esophagitis.
58.3 Incidence
The incidence of GER among the population is
difficult to determine, because the majority of
refluxes are asymptomatic, and a GERD becomes
clinically evident only when complications and
their accompanying symptoms occur. Epidemio-
logic data from a large survey with more than
71.000 adult participants show that 44.1%
reported having had GERD symptoms in the past
and 30.9% reported having GERD symptoms in
the last week (Delshad et al. 2020). In childhood,
prevalence numbers are more difficult because
regurgitation is a common phenomenon in babies,
but usually disappears by 1  year of age, most
probably due to the change of diet from a liquid
to more solid food (Poddar 2019). A French
nationwide cross-sectional observation study on
children and adolescents (>10.000 aged 0–17
years, mean 3.8 ± 5.6 years) showed a prevalence
of 10.3 for GER and 6.2 for GERD (Martigne
et al. 2012). In an Italian study, 12% of more than
2000 infants (less than 1  year old) had clinical
symptoms of reflux, but only 1 of 210 infants
older than 24 months of age developed reflux dis-
ease (Campanozzi et al. 2009).
58.4 Etiopathogenesis
Basically, the esophagus is a relatively simple
transport organ without digestive or absorptive
functions but pushes swallowed content from the
pharynx by means of a propulsive peristalsis into
58  Gastroesophageal Reflux Disease
the stomach (primary peristalsis because it is ini-
tiated by swallow). The esophageal lumen is cov-
ered with a non-keratinizing, stratified squamous
epithelium that is resistant against a variety of
physical and chemical trauma associated with
normally ingested solid food and drinks.
Submucosal glands lubricate the mucosa. At the
esophageal-gastric junction, the stratified epithe-
lial layer changes irregularly to a monolayer
columnar epithelium, the so-called cardia
­epithelium (ora serrata or Z-line), which defines
the inner junction between the esophagus and the
stomach. The muscular coat of the esophagus
consists of an inner circular and an outer longitu-
dinal layer, which are arranged towards the lower
end in a mixed spiral way, a theory which would
explain that the opening of the sphincter is caused
by contraction and shortening of the muscle
fibers and therefore by a shortening of the esoph-
agus (Stelzner and Lierse 1968). At the upper and
lower ends of the esophagus are sphincters—the
upper (UES) and the lower esophageal sphincter
(LES). Pressure values in the UES (40–80 mmHg)
are much higher than those in the LES, which
indicates that reflux into the esophagus may
occur more easily than reflux into the pharynx or
the oral cavity. The UES relaxes during the pro-
pulsion of food from the oral cavity, and strong
contraction of the hypopharynx passes the bolus
with great velocity into the esophagus. However,
even when reflux approaches from below, the
UES might relax and reflux can reach the hypo-
pharynx, larynx, and eventually the mouth.
The LES lies within the diaphragmatic tunnel.
Manometric studies show that there is definitely
no part of the esophagus in the abdomen, but the
LES belongs exactly with its lower part to the
abdominal cavity and with its upper part to the
thorax (Fig.  58.1). The tone of the LES lies
between 15 and 25 mmHg not only in adult per-
sons but also in neonates and infants. The length
in newborn babies is around 1.0 cm and in adults
3.0–5.0 cm. The LES maintains its basic tone but
relaxes with the ongoing peristaltic wave of the
swallow and closes with a brief increase in tone
up to 35–40 mmHg.
755
Fig. 58.1  Manometric pull-through of the LES.  Its
lower part belongs to the abdomen, the upper part to the
thoracic cavity (M, stomach; DU, pressure inversion;
Oe, esophagus). With permission from Höllwarth
M.  Gastroösophagealer Reflux und Erkrankungen des
Magens. In von Schweinitz, Ure (eds) Kinderchirurgie.
SpringerVerlag Heidelberg Berlin 2009
Spontaneous transient relaxation of the LES
occur frequently (TLESR), even in the absence of
any other esophageal activity in healthy individu-
als and create a common cavity (CCP) between
the abdomen and the intraluminal esophagus,
which can be seen by manometric studies
(Fig.  58.2). Manometric studies, simultaneous
with a radiological barium investigation, show
that refluxed material is rapidly pushed back to
the stomach by a secondary propulsive peristal-
sis (evoked not by a swallow but by the reflux
driven dilatation of the organ)  =  volume clear-
ance. In contrast, combined manometric and pH
metric studies show that the pH is not immedi-
ately normalized, but is only stepwise neutralized
by alkaline saliva during subsequent acts of
deglutition = acid clearance (Fig. 58.3). Due to
the fact that during sleep the frequency of swal-
lows is much lower compared to the awake sta-
tus, the acid clearance time is much longer during
sleep phases. Therefore, volume clearance is
rather fast, while neutralization of acidic reflux
may take between 20 s and several minutes, the
latter especially during sleep.
756 M. E. Höllwarth and V. Solari

Fig. 58.2 Transient
spontaneous relaxation of the
LES, terminated by a secondary
peristalsis. The common cavity
phenomenon is shown by the
increase of pressure to the
stomach level and typical
stomach curve (M, stomach;
Oe1, upper esophagus;
Oe2, lower esophagus). With
permission from Höllwarth
M. Gastroösophagealer Reflux
und Erkrankungen des Magens.
In von Schweinitz, Ure (eds)
Kinderchirurgie. SpringerVerlag
Heidelberg Berlin 2009

Fig. 58.3  Combined pH and

manometric study showing a
TLES relaxation with reflux
and terminated by a secondary
peristalsis. The pH gets
normalized later by
swallowing saliva. Patient has
a short cough during reflux
indicating that it reaches the
pharynx (M, stomach;
C, cardia; Oe2, lower
esophagus; Oe1, upper
esophagus; S, swallow;
R, reflux; hustet, cough). With
permission from Höllwarth
M. Gastroösophagealer Reflux
und Erkrankungen des
Magens. In von Schweinitz,
Ure (eds) Kinderchirurgie.
SpringerVerlag Heidelberg
Berlin 2009
58  Gastroesophageal Reflux Disease 757

58.5 Pathophysiology required to detect thickening of this muscle at the

esophagogastric transition (Liebermann-Meffert
The diaphragmatic hiatus is formed mainly by et al. 1979). We have shown that the higher inci-
the right crus of the diaphragm. The esophagus is dence of GER in newborn and small infants is not
anchored within the diaphragm by the phrenico-­ caused by a low or missing LES tone but by an
esophageal membrane. The medial wall of the increased frequency of TLESR’s (Fig. 58.2). The
esophagus continues directly into the lesser cur- LES pressure is between 12 and 15 or 25 mmHg,
vature of the stomach, while the lateral wall not only in adults but also in neonates and infants
forms a type of incisure on the lateral circumfer-
ence in the direction of the fundus—the so-called
angle of His. A rather sharp angle of His is
­supposed to prevent reflux. However, in contrast
to older children and adults, this angle tends to be
flat on the X-ray of a newborn because of the
inferior position of the diaphragm and reflux may
occur more frequently (Fig. 58.4).
Within the lumen of the esophagogastric junc-
tion, a mucosal fold exists at the angle of His,
which is clearly visible during endoscopy, a so-­
called flutter valve.
This mucosal fold is passively pressed against
the esophagus when the gastric fundus is filled
and thus can prevent reflux (Fig. 58.5) (Edwards
1982). However, a flatter angle of His, as it is in
newborns and small infants or typical for obese
persons, the less developed is the flutter valve
mechanism and reflux can occur more easily.
Backward flow of the contents of the stomach
into the esophagus is prevented by the LES.  It
maintains a tone and acts as a barrier between the
stomach and the esophagus. While the pressure
zone can easily be demonstrated by manometry, a Fig. 58.5  Typical flatter valve on the left side of the
special anatomical technique of preparation is hiatus

1 mo 6 mo 3 years

Fig. 58.4  Typical change of the angle of His with age

758 M. E. Höllwarth and V. Solari

(Höllwarth 1979, 2012; Höllwarth and Uray and in infants who are mainly milk fed, the pH of
1985). The length of the LES in the newborn is the reflux in the first 2  h after a remains meal
about 1.0  cm and in adults about 2.5–3.5  cm nearly neutral or increases the pH in the esopha-
(Table 58.1). gus but becomes more acidic later, as evidenced
Most of TLESR are accompanied by reflux of by long-term pH monitoring (Fig. 58.7).
gastric content into the esophagus; however, they
remain unrecognized because the refluxed mate-
rial reaches only the lower esophagus. A recent
study by 24-h impedance manometry in 15
healthy adult volunteers showed on average 4.2/h
reflux episodes, mostly postprandial (42.7%) and
during daytime (31.5%). More than 80% were of
the TLESR were associated with reflux (He et al.
2019). Rarely, refluxed material reaches the phar-
ynx and only then one can realize that a reflux
happened. Depending on the actual pH, the
refluxed material can be highly acidic if it con-
sists of stomach juice only (Fig. 58.3). In infants,
the pH of the reflux reflects more or less the aver-
age pH of the ingested food. Thus, refluxes may
even increase the pH in the esophagus (Fig. 58.6),

Table 58.1  LES tone and length in the first month of life
Age LES tone LES length
n (days) (mmHg) (mm)
Prematures 7 7–28 23.0 ± 3.6 1 0.0 ± 1.1
Newborns 24 1–10 20.4 ± 8.0 10.7 ± 0.8
Fig. 58.6  Spontaneous reflux with an increase of the pH
Newborns 19 11–28 21.8 ± 10.0 11.0 ± 0.5
in the esophagus in a milk nutrition (M, stomach; Oe1,
Infants 20 ˃28 18.0 ± 7.0 11.3 ± 1.1 upper esophagus; Oe2, lower esophagus)

Fig. 58.7  1-year-old child on milk meals: most of the under 4 and even down to pH 1 (N, nutrition; S, sleep; in
refluxes within the first 2 h after nutrition show an increase black, refluxes shorter than 15  s; in red, refluxes longer
of the pH above 4. But after 2 h, the pH of refluxes drops than 15 s up to 5 min during sleep)
58  Gastroesophageal Reflux Disease 759

58.6 Pathology of 1–2 years with the change of nutrition to more

solid food, and it looks like they no longer have
Parasympathetic innervation of the esophagus is by pathological reflux. However, we and others have
the vagus nerve, which courses along the esopha- shown that the disappearance of clinical signs of
gus and sends its fibers to the muscles. Sympathetic reflux does not necessarily mean that the number
innervation is achieved by postganglionic neurons of TLESR and reflux episodes are reduced, and a
of the sympathetic trunk. The myenteric plexus and GERD may still persist (Orenstein et  al. 2006;
the submucosal plexus contain non-adrenergic and Martin et  al. 2002; Pesendorfer et al. 1993).
non-cholinergic nerves and initiate the complex Nearly half out of a cohort of adult patients diag-
activity of the esophagus through a number of neu- nosed with GERD in childhood suffered from
rotransmitters. Central regulation is governed by weekly heartburn or regurgitation and the major-
the deglutition center located in the brain stem and ity of them were taking proton pump inhibitors
core areas of ­peristalsis connected in a serial fash- (PPI) (El-Serag et  al. 2007). There is clear evi-
ion. In infants with central sleep apnea and in cere- dence that, in some children, GERD can persist to
bral disabled patients, disorders in the central adulthood (Winter et al. 2011). Therefore, a 24-h
coordination of motor functions of the esophagus pH monitoring or impedance manometry has to
may play a major role in the etiology of reflux be performed in all children at the age of 1 or
(Landler et al. 1990). 2  years, even when no more clinical symptoms
The contraction and relaxation of the LES is exist, in order to prove if abnormal reflux episodes
triggered by cholinergic and non-adrenergic non-­ persist. Our experience shows that a pathological
cholinergic inhibitory nerves, as well as by nitric reflux pattern, which is still present at the age of
oxide in non-adrenergic non-cholinergic inhibi- 5  years will persist into adulthood (Pesendorfer
tory nerves (Hirsch et  al. 1998; Tomita et  al. et  al. 1993). A systematic review showed that
2003). Experiments with porcine LES stripes based on the current literature, it is not yet possi-
showed that long-chain fatty acid receptors medi- ble to identify children at risk for unfavorable out-
ate relaxation of the LES (Tsai et  al. 2019). come of GERD with regard to the symptoms or
Glutamate is a different transmitter that has been endoscopic findings (Singendonk et al. 2018).
shown to inhibit TLESR (Frisby et  al. 2005). It is generally believed that the luminal acid
Recently, it has been shown that prostaglandin E2 mediates an injury to the superficial cells of the
plays a role in the pathogenesis of gastroesopha- mucosa firstly leading to erosive esophagitis.
geal reflux disease (Bai et al. 2019). Longitudinal However, recent research shows that reflux
esophageal shortening and distension occurs dur- causes a more general esophageal inflammation
ing swallowing resulting in LES relaxation through a cytokine-mediated mechanism starting
(Pandolfino et al. 2006). A recent study showed a in the basal layer of the mucosa (Blevins et  al.
role for cholecystokinin (CCK) by inducing con- 2018; Souza et al. 2009).
traction of the esophageal muscle together with
LES relaxation and crural diaphragm inhibition
(Babaei and Mittal 2018). 58.7 Diagnosis
It is well known that any kind of ingested fluid
triggers TLESR, thereby increasing reflux epi- 58.7.1 Clinical Symptoms
sodes from the stomach to the esophagus. New-
borns and small babies are fed milk; therefore Symptoms of otherwise asymptomatic reflux in
the increased incidence of GER due to TLESR newborns and small infants are common with a
may be caused by their liquid diet shown by prevalence between 40 and 70% up to 6 month of
increased incidences of spitting up, flaccid leak age and a decline to less than 5% after 1 year of age
out, or occasional vomiting immediately after (Moussa and Hassan 2017; Poddar 2019). Typical
being fed, suggesting a GERD.  Characteristi- symptoms include postprandial flaccid leak out,
cally, these symptoms tend to disappear at the age regurgitation, mild spitting up, spilling, or some-
760
times even minimal vomiting. A wet pillow after
sleeping is another sign that reflux reached the
mouth. As mentioned above, reflux is not due to
low LES tone but by TLESR, which might be
caused by factors such as liquid food and/or flat
angle of His without a flutter valve. In contrast,
GERD in that age group is characterized as more
severe recurrent vomiting resulting in constant irri-
tability, failure of thrive, recurrent respiratory
infections due to microaspiration, and restless sleep
interspersed by waking phases and unexplained
crying during sleep (Rosen et al. 2018).
Symptoms beyond infancy are recurrent epi-
gastric pain, effortless vomiting, recurrent respi-
ratory infections, and hoarseness, sometimes
sour mouth odor and gurgling sounds in the chest
after eating (Chang et al. 2006).
A very rare, but characteristic, symptom of
reflux is the Sandifer syndrome, which is accom-
panied by a recurrent tic-like sideways inclina-
tion of the head to the left; this may be interpreted
as support of passive prevention of reflux (Lee
et al. 1999; Rosen et al. 2018).
Adolescents may suffer from similar symp-
toms to adults, such as retrosternal pain and
heartburn. Hypopharyngeal refluxes at nighttime
with microaspiration may cause laryngitis, laryn-
geal pseudopolyps (Fig.  58.8), and chronic
cough. Asthmatic symptoms may also be caused
by chronic microaspiration (Thakkar et al. 2010).
Fig. 58.8  Pseudopolyps on both vocal cords due to recur-
rent acidic reflux into the pharynx and upper airways. With
permission from Höllwarth M. Gastroösophagealer Reflux
und Erkrankungen des Magens. In von Schweinitz, Ure
(eds) Kinderchirurgie. SpringerVerlag Heidelberg Berlin
2009
M. E. Höllwarth and V. Solari
The primary complication of GER is esophagi-
tis, which is caused by frequent and excessively
long relaxations of the LES with recurrent acid
exposure of the esophagus. The resulting inflam-
mation leads to microhemorrhage of mucous mem-
branes and, in chronic cases, hemorrhagic anemia.
A significant aspect of the course of the disease is
the fact that inflammation of the esophagus sets a
vicious circle into motion, characterized by a dis-
turbance of secondary peristalsis and consequent
prolongation of acid clearance. Furthermore, the
inflammation causes more numerous relaxations of
the LES due to reflux. Subsequently, the inflamma-
tion spreads to deeper layers of the wall, eventually
leading to stenosis because of simultaneous scar
formation (Fig.  58.9). The symptoms of evident
esophagitis in children include, in addition to recur-
rent pain in the upper abdomen, frequent bouts of
vomiting partly tinged with blood. Chronic blood
loss might lead to anemia, and the presence of a
significant stenosis can cause swallowing difficul-
ties and regurgitation of food.
Fig. 58.9  Esophageal stenosis due to chronic reflux with
esophagitis. With permission from Höllwarth M. Gastroö-
sophagealer Reflux und Erkrankungen des Magens. In
von Schweinitz, Ure (eds) Kinderchirurgie. SpringerVer-
lag Heidelberg Berlin 2009
58  Gastroesophageal Reflux Disease
Such grave complications in children and ado-
lescent are rare; however, children with mental
disabilities constitute an exception. Grave
sequelae of reflux are often discovered late in
these children because the patients are unable to
articulate their pain. Investigations have shown
that the symptom of autoaggression may be an
important sign of painful esophagitis in these
children and should always be taken as a reason
to investigate the presence of reflux (Goessler
et al. 2007).
58.7.2 Diagnostic Investigations
Several diagnostic tests are available to investi-
gate whether a pathological reflux exists and/or
which complications of a GERD are already veri-
fiable. Each test has its advantages and disadvan-
tages and usually several tests have to be used in
combination to assess the severity or grade of
reflux. A recent systematic review compared dif-
ferent diagnostic tests (Zhang et al. 2019).
58.7.2.1 Upper Gastrointestinal
X-Ray Passage
X-rays with contrast material are mainly used to
investigate the morphology and peristaltic func-
tion of the esophagus. The primary benefits of
an X-ray investigation include visualization of
the esophagogastric junction or evidence of a
sliding or fixed hiatal hernia, assessment of the
angle of His, the presence of an orderly or path-
ological pharyngoesophageal act of swallowing,
the course of esophageal peristalsis, and any
irregularities of the epithelium as signs of
inflammation. Evidence of aspiration of contrast
medium during the investigation is another
important finding. Additionally, different
pathologies can be visualized or excluded such
as pylorus hypertrophy, antral or duodenal web
or malrotation, or other pathologies in the upper
GI tract. Actual evidence of reflux episodes, on
the other hand, is of lesser significance because
the duration of a barium esophagogram is short
and an obvious reflux may occur or not occur
during the few minutes of fluoroscopy. However,
761
indirect signs of a GERD are several: frequent
air reflux during the investigation and a positive
water siphon test (reflux after drinking a large
sip of water to provoke reflux), and in the case
of a reflux events, it will be recorded to which
height these refluxes reach in the esophagus
(lower/middle/upper esophagus) (Fotter et  al.
1985).
58.7.2.2 24-Hour pH Monitoring
This technique has long been the gold standard to
evaluate the function of the esophagus related to
reflux events. Glass or antimony electrodes are
introduced through the nose and pH values are
recorded continuously on a battery-operated
recorder. At the end of the 24-h investigation, the
results are evaluated by means of suitable soft-
ware. Ideally, multichannel probes are used to
evaluate pH levels in the stomach, the lower
esophagus, and the upper esophagus. Falling pH
levels in the esophagus can be correlated with the
pH of the stomach, and the number of refluxes
that reach the esophagus are analyzed (Semeniuk
and Kaczmarski 2007). All drops in pH below 4,
with a duration of at least 15 s (number of signifi-
cant refluxes), and the time taken for pH levels to
normalize or rise back to pH 4 (reflux clearance),
the number of refluxes with clearance time longer
than 5 min, the longest reflux time, and the reflux
index (RI  =  ratio of total number of significant
refluxes with a pH <4 and the recording time—
24  h) are analyzed. The type and the ingestion
time of food, as well as the length of time the
person lies in supine position or is upright, are
also registered. The limitation of the pH monitor-
ing comes from the fact that a large number of
refluxes may not change the pH or even slightly
increase it and therefore cannot sufficiently dif-
ferentiated by this technique (Vandenplas et  al.
2007; Moussa and Hassan 2017). Furthermore,
laboratories use different cutoff values, which are
partly influenced by standard values in adults
(Lupu et al. 2016). The lower threshold of the RI
at 3% in largely milk-fed infants reflects the buff-
ering of milk and the consequent lower numbers
of acidic refluxes (accompanied by a pH drop to
below 4) (Table 58.2).
762
Table 58.2  Cutoff values of the RI in our lab (the lower
values for children <1 year of age are the result of the fact
that the gastric acidity is neutralized in young infants as
long as they are mainly milk fed)
Age Low grade Intermediate High grade
(year) % % %
<1 3–5 5–7 >7
>1 5–7 7–10 >10
58.7.2.3 pH-Impedance Monitoring
(pH/MII)
This 24-h multichannel method of impedance
measurement, combined with pH monitoring
(pH/MII), by which voltage variations during
distension of the esophagus are recorded at mul-
tiple recording points, does not merely show the
pH of refluxes but also the direction of the bolus.
Thus, it not only permits measurement of acid
refluxes but also of all neutral and alkaline
refluxes over a period of 24 h and is the new gold
standard (Fig. 58.10). Thus, the informative value
is greatly enhanced compared to simple pH mon-
itoring (Vandenplas et al. 2007; Safe et al. 2016;
Rossi et  al. 2018). The combined pH/MII
­monitoring is clearly superior to pH monitoring
alone, especially because, in young infants,
symptoms are more frequently associated with
weakly acidic or neutral refluxes. In one pub-
lished study, acid-related parameters were sig-
nificantly related to MII baselines, although they
did not correlate exactly with the endoscopic find-
ings (van der Pol et al. 2013). However, the results
of this study might have been influenced by the dif-
ficult macroscopic estimation of a grade of esopha-
gitis, especially with lower grades. Recurrent
cough due to microaspiration of nonacidic refluxes
may play a significant causative role in respiratory
tract infection and is a diagnostic domain of pH/
MII (Blondeau et al. 2011). Furthermore, pH/MII
results have an influence on the choice of medical
treatment, since it is possible to discriminate
patients with primarily acid refluxes from those
with nonacid refluxes needing a different neuro-
modular therapy (Jodorkovsky et al. 2014). There
are still a number of limitations, such as limited
defining references, different interpretations among
users, and the availability of the technology in all
medical centers. Reflux-like symptoms during a
M. E. Höllwarth and V. Solari
Fig. 58.10  Combined impedance and pH monitoring. It
shows a short reflux episode with decrease of the pH and
a small increase of the pH with the next swallow. With
permission from Höllwarth M.  Gastroösophagealer
Reflux und Erkrankungen des Magens. In von Schweinitz,
Ure (eds) Kinderchirurgie. SpringerVerlag Heidelberg
Berlin 2009
pH/MII investigation (irritability, bradycardia,
desaturation) do not correlate with reflux events
(Funderburk et al. 2016). However, the recent pub-
lished guidelines support the use of pH/MII in chil-
dren to differentiate patients with nonerosive
reflux, eosinophilic esophagitis (EoE), and/or func-
tional heartburn, to determine the efficacy of acid
suppression therapy and the role of acid and non-
acid refluxes in the etiology of esophagitis (Rosen
et al. 2018). Recently, an in vivo mucosal imped-
ance (MI) catheter system was developed that is
supposed to instantly detect changes in the mucosal
barrier function, corresponding to a reduced tran-
sepithelial resistance during endoscopy over a long
segment of the esophagus allowing the identifica-
tion of patients with GERD, EoE, or non-GERD
(Patel et al. 2019).
58.7.2.4 Manometry
Manometric investigation of esophageal function
was a diagnostic method introduced in the late
1960s, primarily to record the pressure within the
LES.  The general assumption at that time was
58  Gastroesophageal Reflux Disease
that primarily a low LES pressure is the respon-
sible factor allowing reflux to occur. Initial mano-
metric investigations in newborns and infants
appeared to confirm that the LES pressure in this
age group is extremely low. However, only when
low-compliance perfusion pumps were used,
investigators were able to register relevant pres-
sures values. As mentioned earlier, investigations
in the 1970s already showed that normal LES
pressures are present in neonates and infants
(Höllwarth 1979). Furthermore, manometry is
highly suitable to analyze the motor functions
and peristalsis of the esophagus and is the gold
standard investigation for esophageal motility
disorders (Rosen et al. 2018). It should be consid-
ered in children with GER symptoms refractory
to acid suppression therapy (Hsing et al. 2019).
Manometry shows that reflux of gastric contents
is rendered possible not due to low pressures but
due to spontaneous transient relaxations of the
LES (Fig. 58.2). These occur much more often in
patients with GERD and persist for a signifi-
cantly longer time. The refluxed volume is
returned to the stomach by secondary peristalsis,
whereas the drop in pH is normalized in a step-
wise manner by swallowing saliva (Fig.  58.3).
High-resolution manometry together with simul-
taneous pH monitoring permits demonstration of
the pH of refluxes being nonacidic, neutral, or
alkaline and allows the investigator to draw con-
clusions similar to those obtained by impedance
measurement. However, the drawback of manom-
etry is the fact that it is motion dependent, in con-
trast to pH monitoring and pH/MII.
58.7.2.5 Endoscopy and Histology
Investigations with flexible fiberoptic endoscopes
are part of the standard investigation of GERD. It
allows the evaluation of erosive changes of the
esophageal mucosa, as well as taking biopsies
and differentiating other conditions, such as ste-
noses, Barrett’s esophagus, and hiatal hernia.
The device is introduced into the esophagus
under direct vision, usually down to the duode-
num—esophagogastroduodenoscopy (EGD).
Biopsy specimens are taken from the duodenum
and the antrum of the stomach. In the stomach,
the tip of the device is inverted to view the
763
esophagogastric junction from below. In normal
cases, the device is closely encircled by the
esophagus. At the lateral circumference, one
sees the flap valve mentioned above (Fig. 58.5).
In contrast, in a hiatal hernia, the diaphragmatic
pinch is a little open and one obtains a clear
upward view towards the higher positioned
LES.
By withdrawing the device further, one is able
to inspect the esophagogastric junction and the
Z-line with the cardia monolayer epithelium
more closely. Within the esophagus, the endosco-
pist evaluates erosive esophagitis: the typical
esophageal squamous epithelium is smooth and
of milky red color (Grade 0). According to
Savary-Miller, the grades in the presence of
esophagitis are as follows: Grade 1, single ero-
sion or exudative lesion on only one longitudinal
fold; Grade 2, multiple erosions on more than
one longitudinal fold; Grade 3, circular erosion
or exudative lesion; Grade 4, ulcers, strictures, or
short esophagus; and Grade 5, Barrett’s epithe-
lium. The more sophisticated Los Angeles
Classification of the severity of reflux esophagitis
can also be used in children; it describes four
grades of mucosal damage, depending on the
extension between mucosal folds.
However, it must be emphasized that the mac-
roscopic grading, especially with exudative
lesions of the Grades 1 and 2, is affected by the
impression of the observer and does often not
correlate with the histological results. Therefore,
it is mandatory to take several biopsy specimens,
starting 1–2  cm proximal to the Z-line and
extending upward into the upper esophagus. It is
advisable to place the biopsy specimens on a
piece of cork, in appropriate direction, immedi-
ately after taking the specimen, and then insert-
ing it in formalin. An optimal specimen consists
of the entire layer of epithelium, including the
basal cell layer. Histologic specimen is evaluated
in regard to basal cell hyperplasia, papillary elon-
gation, intraepithelial eosinophils, neutrophil and
mononuclear cell number, necrosis/erosion and
healed erosions, and dilated intracellular space
(Schneider et al. 2014). Histologic signs of ero-
sions and ulcerations are, by their very nature,
signs of severe chronic esophagitis. However, the
764 M. E. Höllwarth and V. Solari

latter is not always associated with unequivocal the lung leading to a flatten diaphragm and loss
symptoms in children. A recent study investigat- of the angle of His. Investigations on this subject
ing symptom severity and histological grade in have shown that 25–75% of children with asthma
children and adolescents did not show a positive actually have a pathological reflux, but only 50%
correlation, even in adolescent with heartburn or of them have corresponding symptoms such as
chest pain (Quitadamo et al. 2015). The presence belching, vomiting, or pain in the upper abdo-
of more than 20 eosinophils per “high-power” men. It is assumed that microaspiration and even
field, on the other hand, is a sign of another non-­ acidification of the esophagus in asthmatics may
reflux-­related allergic or atopic disease, i.e., so-­
trigger hyperreactivity of the respiratory tract.
called eosinophilic esophagitis (EoE) (Furuta pH/MII is reasonable to detect all reflux events
et al. 2007). and to estimate whether they reach the upper
esophagus or not. In patients suspected of having
58.7.2.6 Scintigraphy GERD, empiric therapy with proton pump inhib-
Scintigraphic investigation methods permit the itor (PPI) is now considered as a first diagnostic
observation of the moving bolus through the step (Naik and Vaezi 2015). A positive effect
esophagus, any aspiration of tracer into the lungs would explain why drug-based reflux treatment
due to reflux, and measurement of gastric empty- reduces the symptoms of asthma or causes the
ing time after standardized meals. The latter con- patient to require lower doses of asthma therapy.
sist of a solid and a liquid portion, such as a meal 24-Hour scintigraphy may show that refluxed
consisting of egg and water. 99mTc sulfur colloid material was aspirated into the airway tract.
is used as tracer. The act of swallowing is visual- Recently, salivary pepsin was shown to be a
ized on dynamic sequences. After setting regions valuable intrinsic marker for gastroesophageal
of interest (ROI), refluxes are registered, and the reflux-­related disorders (Wang et  al. 2020).
mean gastric emptying rate is documented on a Surgical treatment of reflux is rarely needed but
time/activity curve. Investigations after 24 h per- can lead to a clinically demonstrable improve-
mit the detection of tracer in the lung, provided ment of asthma symptoms in 85% of cases (Field
aspiration has occurred (Ravelli et al. 2006; Bar-­ et al. 1999).
Server 2017).
58.7.3.2 Reflux and Apnea Syndrome
Sudden infant death syndrome (SIDS) is the most
58.7.3 Differential Diagnosis common cause of death in infants; its incidence is
especially high between the age of 2 and
There are a number of diseases that have relations 4  months. In the last few years, the association
to GER, either as a side effect of the underlying between SIDS and episodes of apnea, on the one
disease or as a complication of a GERD. hand, and gastroesophageal reflux, on the other
hand, was investigated. Diverse results were
58.7.3.1 Reflux and Asthma reported (Slocum et  al. 2007). Some authors
Symptoms of GERD are frequently seen in demonstrated acid reflux as the cause of apnea
asthma, but the true incidence of reflux in asth- episodes, whereas others were unable to show
matic children is not known (Tolia and this association. Combined investigations con-
Vandenplas 2009). A prevalence study on more sisting of manometry and pH monitoring in
than 2000 students (age 13–14 years) in Seattle infants with pathological sleep apnea showed fre-
middle schools showed that symptoms of GERD quent episodes of apnea and reflux but did not
were significantly higher (19.3%) in children demonstrate acid refluxes as the direct cause of
with asthma compared to those without asthma apnea. Landler observed a marked delay in the
(2.5%) (Debley et  al. 2006). While recurrent growth and development of the esophageal motor
aspiration may trigger asthmatic symptoms in functions of the esophagus in children with
contrast, chronic asthma causes hyperinflation of “near-miss” events (Landler et al. 1990). An ear-
58  Gastroesophageal Reflux Disease 765

lier study from the same group found that infants and allergic rhinitis. Moreover, in 50% of fami-
with pathological sleep apnea also have patho- lies exists an atopic disorder (Reed and Dellon
logical reflux in many cases, whereas infants 2019; Alterio et al. 2019). The decisive aspect of
with primarily pathological reflux do not experi- differential diagnosis and its delineation from
ence episodes of apnea to a remarkable extent reflux disease are the findings of a normal 24-h
(Kurz et al. 1985). The investigation confirms the pH monitoring and no response to treatment with
hypothesis that infants with sleep apnea syn- acid blockers. X-ray investigation of esophageal
drome (or near miss events) suffer from delayed passage reveals a circumscribed stenosis in about
regulatory function of some centers in the brain 6% of children; in most cases, the stenosis does
stem, because of which the higher centers respon- not require dilatation. The endoscopic investiga-
sible for esophageal motor function are also tion frequently shows concentric rings of mucosa
involved. On the other hand, delayed maturation (similar to the trachea) and whitish, partly flaky
of esophageal motor functions is not necessarily exudates, thickened mucosal folds, and, in dis-
associated with disorders of deeper respiratory tension of the esophagus, longitudinal folds over
regulation centers. Thus, the rather frequent coin- longer distances. The diagnosis can be estab-
cidence of reflux and sleep apnea syndrome lished when, in the presence of a normal 24-h pH
appears to be caused by underlying retarded mat- monitoring, the histological investigation shows
uration in both centers in the brain stem. In older more than 15–20 eosinophils per “high-power”
children and adults, there is a strong evidence of field (Furuta et  al. 2007). In recent published
a bidirectional relationship between GERD and guidelines, the use of pH/MII in children is sup-
sleep disturbances, including sleep fragmentation ported to differentiate patients with nonerosive
and early morning awakenings (Fujiwara et  al. reflux, eosinophilic esophagitis (EoE), and/or
2013). PPI therapy in children with sleep apnea functional heartburn, to determine the efficacy of
has been shown to reduce the obstructive events acid suppression therapy and the role of acid and
significantly in some patients indicating a rela- non-acid refluxes in the etiology of esophagitis
tionship between apnea and reflux (Wasilewska (Rosen et al. 2018). As regards therapy, primarily
et al. 2012). foodstuffs identified as allergens on testing must
be avoided. However, symptoms improve only
58.7.3.3 Eosinophilic Esophagitis after 2–3 weeks in many cases. Initial choice for
Eosinophilic esophagitis is an immune-antigen-­ pharmacologic therapy is proton pump inhibitors
mediated disease in which antigens in certain (PPI). If the response is insufficient, further rec-
foods elicit a T-helper cell cascade leading to ommended drug therapies include topically or
eosinophilia inflammation and fibrosis within the systemically administration of corticosteroids.
mucosa (Blevins et al. 2018). There is increasing On average, 50% of children with eosinophilic
incidence and prevalence of this condition in esophagitis responded to PPI therapy indepen-
children and adults affecting the esophagus with dent of the pH monitoring results (Gutiérrez-­
symptoms similar to those of reflux (Alterio et al. Junquera et  al. 2016). Therefore, the beneficial
2019). In children and adults, typical symptoms use of PPI in children with eosinophilic esopha-
are a refusal to eat, eating slowly, extensive food gitis needs further evaluation (Dranove et  al.
chewing, avoidance of troublesome food and 2009).
dysphagia, and pain in the upper abdomen. With
advancing age, the individual may experience 58.7.3.4 Barrett’s Esophagus
other symptoms similar to reflux, such as regur- In 1950, Norman Barrett reported in detail
gitation of food, vomiting, dystrophy, and regarding a columnar-lined esophagus in patients
retrosternal pain (Reed and Dellon 2019). Quite that is associated with GER and chronic esopha-
often, eosinophilic esophagitis occurs in combi- gitis (Barrett 1950).
nation with other atopic diseases, children have a The histological feature of Barrett’s esopha-
prevalence of 30–50% and 50–75% for asthma gus (BE) is replacement of the layered squamous
766
Fig. 58.11  Histology of a biopsy of Barrett’s esophagus
with intestinal metaplasia and typical goblet cells. With
permission from Höllwarth M.  Gastroösophagealer
Reflux und Erkrankungen des Magens. In von Schweinitz,
Ure (eds) Kinderchirurgie. SpringerVerlag Heidelberg
Berlin 2009
epithelium of the esophagus by the metaplastic
epithelium, either from the cardia or from the
stomach or intestinal tract. According to the
American Gastroenterological Association Tech-
nical Review, a distinction is made between three
forms. Intestinal metaplasia indicates the pres-
ence of a duodenogastroesophageal reflux, and
intestinal epithelium has replaced the squamous
epithelium of the esophagus. In the strict sense of
the term, only the third type is known as BE
(Fig. 58.11).
(a) The metaplastic cylindrical epithelium of the
cardia composed almost exclusively of
mucus secreting cells.
(b) Gastric-fundic-type epithelium with mucus
secreting cells, parietal cells, and chief cells.
(c) Intestinal-type epithelium (intestinal meta-
plasia) containing goblet cells, associated
with a higher risk of carcinoma.
• Subclassification:
–– c/a: non-dysplastic BE
–– c/b: BE with high-grade dysplasia
–– c/c: adenocarcinoma
–– d/a: BE >3 cm above the gastroesopha-
geal junction
–– d/b: BE <3 cm above the gastroesopha-
geal junction
–– e/a: BE circumferential extend
–– e/b: BE maximum extend
M. E. Höllwarth and V. Solari
On endoscopy, the mucosa is rather flesh-­
colored, slightly bleeding with irregular changes.
These may be insular in shape or may cover wide
areas in a circular fashion. By injecting Lugol’s
solution or indigo carmine solution, the mucosal
changes can be seen better and biopsied more
accurately. The histological investigation permits
differentiation of the more harmless condition of
cardiac or gastric epithelium from intestinal
metaplasia with typical goblet cells. In cases of
proven intestinal metaplasia, one must obtain
four-quadrant biopsy specimens at intervals of
2 cm along the esophagus.
BE is the major precursor lesion for most
cases of esophageal adenocarcinoma, with an
increasing incidence. The prevalence of BE in the
general population is around 1–2% and the
annual risk of BE converting to esophageal ade-
nocarcinoma is estimated at 0.12–0.5% (Maes
et al. 2016). There are multiple predictors of pro-
gression to high-grade dysplasia or adenocarci-
noma in BE (Whitson and Falk 2015). Regular
endoscopic screening of patients with GERD at
risk for BE is recommended for male >5  years
GERD or > weekly GERD symptoms and at least
two more risk factors: age > 50, central obesity,
smoking history, Caucasian, and first-degree rel-
ative with BE or esophageal adenocarcinoma.
The guidelines suggested by the American
College of Gastroenterology for GERD at risk for
BE (intestinal metaplasia) recommend the fol-
lowing monitoring strategy:
(a) In the case of dysplasia not demonstrated on
two occasions: control endoscopies every
2–3 years.
(b) In the case of low-grade dysplasia demon-
strated on two occasions: control endosco-
pies every year.
(c) In the case of high-grade dysplasia: selective
resection or control endoscopies at 3-month
intervals.
Today, intensive clinical research is focused
on risk stratification to perform screening endos-
copy in the appropriate patient’s group (Steele
et al. 2019). Analysis of the results of esophago-
gastroduodenoscopies for uncomplicated GERD
showed that only 5.6% had suspected BE of any
58  Gastroesophageal Reflux Disease
length and 1.4% has suspected BE >3 cm length
(Lin et al. 2019).
Raicevic and Saxena performed a systematic
review in children and reported that out of 130
cases of BE, 61.5% were associated with GERD,
10% with esophageal atresia and reflux, and the
other patients with a variety of different diseases
(Raicevic and Saxena 2018). Adenocarcinoma
after BE is rare in children; however, individuals
born preterm or small for gestational age have a
threefold increased risk for developing BE as
adults (Forsell et al. 2013). The presence of high-­
grade dysplasia or adenocarcinoma is rare in
children and has been reported in only a few iso-
lated cases. Radiofrequency ablation is highly
effective and safe for treatment of BE with dys-
plasia or early stage cancer.
58.7.3.5 Hiatal Hernia
Any upward shift of portions of the stomach into
or beyond the esophageal hiatus is referred to as
a hiatus hernia (HH). A distinction is made
between axial and paraesophageal hiatal hernias.
The axial HH is either a sliding phenomenon or
is fixed in the chest region (Fig. 58.12). A sliding
Fig. 58.12  Typical hiatal hernia seen during endoscopy.
The lower “ring” corresponds with the diaphragm; high
up one can see the LES. (With permission from Höllwarth
ME: Gastroesophageal reflux disease. In Coran A (editor
in chief) Pediatric Surgery, seventh edition 2012 by
Saunders, an imprint of Elsevier)
767
HH is much rarer in children than in older adults
(it occurs in approximately 60% of adults). A fixed
HH is occasionally seen in children after correc-
tion of an esophageal atresia and/or diaphragmatic
hernia, in children with mental disability, congeni-
tal short esophagus, and esophagitis-related steno-
sis with a shortening of the esophagus due to
scarring. In infants or young children with reflux
disease and HH, the condition cannot be expected
to resolve spontaneously. Therefore, conservative
therapy has no chance of success in this setting and
surgery is indicated in all cases.
Paraesophageal hernias are a typical postoper-
ative complication after fundoplication (4.5%) and
occur less often after suture closure of the hiatus
(3%) than without (10%) (Fig.  58.13) (Avansino
et al. 1999). A part of the stomach slips through the
hiatus into the chest sideways, from the esophago-
gastric junction located in normal position. When a
paraesophageal hernia is not combined with a path-
ological recurrent reflux and no further symptoms
are present (such as a sensation of postprandial
pressure or pain, dysphagia, vomiting, or gastric
obstruction), surgical correction is not required in
every case. The congenital upside-down stomach is
an extreme form of a paraesophageal hernia and
must be corrected surgically.
Fig. 58.13  Typical endoscopic view of a paraesophageal
hernia. (With permission from Höllwarth ME: Gastro-
esophageal reflux disease. In Coran A (editor in chief)
Pediatric Surgery, seventh edition 2012 by Saunders, an
imprint of Elsevier)
768
58.8 Management
58.8.1 Conservative Treatment
The North American Society for Pediatric
Gastroenterology, Hepatology, and Nutrition
(NASPGHAN) along with the European Society
for Gastroenterology, Hepatology, and Nutrition
(ESPGHAN) published 2018 clinical practice
guidelines for children with GER, in regard to
diagnostic and therapeutic strategies. The follow-
ing management strategies are based on these
recommendations. Additionally, patient and
parental education, guidance, and support are
always required as a part of the treatment of
GERD (Rosen et al. 2018; Poddar 2019). In this
age group, it is important to avoid the use of dis-
ease labels because they may promote overtreat-
ment (Vandenplas and Hauser 2015). Parents of
healthy children should be assured that symp-
toms of GER are common in babies while on liq-
uid nutrition and may resolve spontaneously after
a year. However, we always tell the parents that a
pathological reflux can still persist, but the
refluxed material doesn’t reach the mouth. Thus,
a pH/MII control is advisable at the age of 1 or
2  years in order to prove that no pathological
reflux pattern is going on.
58.8.1.1 Babies and Small Infants
Conservative therapy is the primary strategy
focused on feeding habits in this age group.
A reduction of the ingested volume per feed-
ing can already avoid an overloading of the
stomach, but at the same time, the frequency of
feedings is increased to allow the adequate and
age-appropriate nutrition. Additional thickening
of feedings with cereals improves visible regur-
gitation. Despite the safety concern of the FDA
due to elevated levels of inorganic arsenic in all
forms of rice, rice gruel still has the advantage
of a good dissolving thickener and it is afford-
able. Of course, breastfeeding should be sup-
ported at least for the first 6 months. In breastfed
babies with significant GER, the breast milk can
be thickened with carob bean too, which is
approved for infants after 42 weeks of gestation.
Feeding modifications are recommended as the
M. E. Höllwarth and V. Solari
first therapeutic strategy before other costly or
risky interventions.
Investigations have shown that lying in prone
position, possibly with a raised torso, prevents
reflux most effectively. However, it also involves
a much greater risk of sudden infant death, due to
vomiting in the sleep and apnea events. Therefore,
the supine sleeping position is recommended.
Infants with an allergy to cow’s milk protein
(CMPA) often show regurgitation and vomiting
typical of GERD.  Vomiting frequency signifi-
cantly decreases within 2  weeks of cow’s milk
protein elimination from the diet. Thus, infants
with suspected CMPA should receive formula
feeding with hydrolyzed protein.
58.8.1.2 Conservative Therapy
in Older Children
and Adolescents
As the child’s food starts to resemble that of
adults, there is an increasing quantity of acid
secretion from the stomach after meals. Therefore,
the majority of postprandial refluxes are now
accompanied by a pH reduction to below 4, and,
in cases of GERD, the possibility of esophagitis
is increased.
Recommendations for lifestyle and behavioral
changes include avoiding food and beverages (caf-
feine, chocolate, carbonated beverages, spicy or
acid food) that trigger reflux symptoms. A meta-
analysis showed that weight loss and head of the
bed elevation and left lateral decubitus are the only
effective lifestyle interventions (Kaltenbach et al.
2006). However, obesity is still a matter of discus-
sion. A retrospective study did not show that obe-
sity is a risk factor in 738 children with reflux
esophagitis (Elitsur et al. 2009), while other stud-
ies found an association between childhood obe-
sity and the risk for developing a GERD (Koebnick
et al. 2011; Quitadamo et al. 2012).
58.8.1.3 Pharmacologic Treatment
The aim of drug therapy is to reduce acid expo-
sure to the esophagus and thus avoid or treat
esophagitis. A distinction should be made here
between drugs that protect the mucosal surface
and those that reduce or hinder the production of
gastric juices.
58  Gastroesophageal Reflux Disease 769

Alginates acid or volume clearance takes a very long time.

Alginates are aluminum, magnesium, or calcium This happens in children who have undergone
salts of alginic acid which are used to treat acid-­ surgery for esophageal atresia, especially when
related disorders such as heartburn or dyspepsia. the lower segment has to be mobilized exten-
Alginates are often combined with sodium/potas- sively, in children who have experienced dia-
sium bicarbonate. They buffer gastric acid and phragmatic hernia, in those with an incompetent
improve the symptoms of esophagitis and reduces esophagogastric junction, as well as children
signs of inflammation. Short-term treatment with with severe mental disorders.
alginates seems to have no side effects, but long-­ Children frequently require a much higher PPI
term use of aluminum-containing alginates dose compared to adults (0.7–3.5  mg/kg/day
(sucralfate) may lead to increased aluminum BW). The efficacy of gastric acid blockade must
plasma concentration in infants. NICE (National be checked by performing pH monitoring about
Institute for Health and Care Excellence) guide- 2 weeks after the start of therapy, as the dose may
lines recommended alginates as an alternative to need to be modified significantly in individual
feed thickeners in breastfed infants or as a trial incases. Investigations after long-term treatment
infants in whom symptoms persist despite con- with PPIs in adults revealed a 30% reduction in
servative measures (Davies et al. 2015). vitamin B12 levels, atrophic gastritis, and several
weeks of acid hypersecretion after discontinua-
Proton Pump Inhibitors (PPI) tion of the medication. As mentioned earlier,
PPIs are the most potent pharmacologic agents long-term therapy with acid inhibitors in children
suppressing acid production by inhibiting H +/K is indicated only in exceptional cases, although
+
-ATpase in parietal cells within the stomach. In positive results have been reported in the pub-
fact, PPIs reduce the acidity of gastric contents lished literature in this regard (Hassall et al. 2007;
and refluxes, thereby avoiding erosions of the Tolia and Boyer 2008). Side effects are rare and
esophageal mucosa but does not necessarily mostly mild (headache, constipation, diarrhea,
reduce the total number of refluxes. As long as nausea, hypomagnesemia) (Carroll and Jacobson
this is the case, the vicious cycle between severe 2012). However, long-term treatment with PPIs
esophageal inflammation leading to a high num- is associated with a higher risk due to the pro-
ber of TLESR exists. Thus, respiratory tract-­ longed acid suppression and hypochlorhydria,
related symptoms due to recurrent aspiration which impairs vitamin B12, calcium, and iron
during sleep may still be present. absorption. Furthermore, long-term acid suppres-
Infants: Most milk-fed infants usually do not sion may lead to a significant change of the small
require a drug-based treatment of reflux. The bowel microbiota and overgrowth with patho-
above-mentioned guidelines do not recommend genic bacteria causing gastroenteritis, increased
PPI for crying distress and also not for the treat- risk of community-acquired pneumonia, and
ment of visible regurgitation in otherwise healthy Clostridium difficile infections (Moussa and
infants. However, if recurrent unquietness, dis- Hassan 2017; Nylund et  al. 2014; Berni and
turbed sleep, and pain are the symptoms of a Terrin 2010; Lo and Chan 2013; Poddar 2019).
GERD, PPI therapy is indicated. Studies in
infants <1  year of age with pain-related symp- Histamin-2 Receptor Antagonists (H2RAs)
toms have shown that PPI reduced the number of H2RAs reduce acid secretion and pepsin by
acid refluxes significantly with a relative increase blocking the histamine H2 receptor on the surface
of the total number of non-acidic refluxes. Pain-­ of parietal cells in the stomach but might fail in
related symptoms improved significantly, but the presence of gastric juice secretion induced by
there was no effect on vomiting and recurrent meals. Studies have shown superior efficacy of
respiratory infections (Castellani et al. 2014). PPI H2Ras over placebo in improvement of symp-
therapy is useful for infants in which the peristal- toms, as well as endoscopic features (Carroll and
sis of the esophagus is massively disturbed, and Jacobson 2012). However, today the efficacy of
770
H2RAs is considered to not be as effective as PPI
therapy in GERD children (Savarino et al. 2009;
Ummarino et  al. 2012). A recent observation
cohort study with several thousand patients in
each group showed more unfavorable complica-
tions, mostly gastroenteritis cases after H2Ras
when compared to PPI therapy (Ruigómez et al.
2017).
Prokinetics
Baclofen, a gamma-aminobutyric acid receptor
agonist (GABA), is a promising therapeutic
option for the pharmacologic inhibition of tran-
sient sphincter relaxation (Lidums et al. 2000).
Experience reported with baclofen in adults and
children with GERD has demonstrated a signifi-
cant reduction of LES relaxations and reflux
(Dent 2009). Baclofen reduced significantly
upright refluxes as well as belching and regurgi-
tation in adult patients (Cossentino et al. 2012).
In children with refractory reflux, baclofen
together with PPI significantly reduced the clin-
ical symptoms in two-thirds of patients
(Vadlamudi et  al. 2013). Adverse side effects
may be drowsiness, dizziness, and fatigue.
Other prokinetic drugs were used in the past to
improve peristalsis and cleansing functions of
the esophagus. However, experience has shown
that none of these drugs (metoclopramide, cis-
apride, bethanechol, erythromycin) had a con-
firmed reflux-preventive effect, and the side
effects may superimpose any positive effects.
Therefore, currently there are little or no indica-
tions for the use of these drugs in reflux disease
(Rosen et al. 2018).
58.8.2 Surgical Treatment
Antireflux surgery is usually performed when
other options failed or in complicated GERD
cases, such as severe esophagitis with stenosis, in
children with esophageal dysfunction due to
other congenital malformations (esophageal atre-
sia, diaphragmatic hernia, congenital distal
esophageal stenosis), or in children with mental
disorders and a severe GERD. Today, fundoplica-
tion is the gold standard of surgical treatment.
M. E. Höllwarth and V. Solari
58.8.2.1 Fundoplication
The strategic principles of surgical reflux treat-
ment was introduced by Nissen in 1956 and are
based on the creation of an intra-abdominal por-
tion of the esophagus and complete plication (dor-
sally Toupet, ventrally Thal) of the gastric fundus
around the esophagus (Nissen 1979). Later,
Toupet recommended Atlas Series. Springer
GmbHended a partial plication with a dorsal part,
and Thal a partial plication with a ventral part of
the fundus (Toupet 1963; Thal et al. 1965).
The first step requires removal of the esopha-
gus from the hiatus in order to obtain a longer
intra-abdominal portion. Therefore, approximat-
ing the crura with one or two sutures may be
required additionally to reduce the risk of wrap
herniation into the thorax (Fig.  58.14). For the
Nissen procedure, a part of the fundus is pushed
behind the esophagus from the left to right and
sutured ventrally with the other part of the fundus
in a floppy way (Fig.  58.15) (Slater and
Rothenberg 2017). The selection of fundoplica-
tion depends on the surgeon’s choice. Favorable
results have been achieved with all three ­methods.
Fig. 58.14  Approximation of the diaphragmatic crura.
With permission from Georgeson KE and Höllwarth ME.:
Gastroesophageal Reflux and Hiatus Herna. In Puri P,
Höllwarth M. (eds) Pediatric Surgery Springer GmbH
Germany
58  Gastroesophageal Reflux Disease
Fig. 58.15  Pull-through of a part of the fundus behind
the esophagus and closure ventrally. With permission
from Georgeson KE and Höllwarth ME.: Gastroesophageal
Reflux and Hiatus Hernia In Puri P, Höllwarth M. (eds)
Pediatric Surgery Springer GmbH Germany
However, a prospective randomized study with
200 adult patients comparing Nissen with Toupet
fundoplication showed that 85% of patients have
been satisfied with the result, but dysphagia was
significantly more common following Nissen and
did not correlate with preoperative motility prob-
lems. The authors conclude that the Toupet pro-
cedure is the better operation (Strate et al. 2008).
Another study of the Nissen vs Thal procedure
showed similar results but a higher risk of reop-
eration for recurrent symptoms after the Thal
technique (Du et al. 2017). The current standard
is laparoscopic surgery or robot-­assisted laparo-
scopic fundoplication. However, the latter does
not provide any substantial clinical advantage
(Rosen et  al. 2018; Hambraeus et  al. 2013).
According to Nissen, the gastric fundus should
be closed very loosely in a circular fashion
around the esophagus with just 2–3 sutures so
that passage is not hindered, and spontaneous
vomiting is still permitted. Tovar has convinc-
ingly shown that fundoplication with a floppy
wrap is a powerful means of controlling reflux in
appropriately selected children in whom medical
treatment fails, or in symptomatic patients with
specific comorbidities such as neurologic disor-
771
ders or congenital anomalies (Tovar et al. 2007).
Systematic reviews of randomized controlled tri-
als of laparoscopic versus open fundoplication
showed a significant lower operative morbidity,
shorter postoperative stays, and a less prolonged
sick leave following the endoscopic approach
(Rhee et al. 2011; Zhang et al. 2016). In a ran-
domized multicenter trial with a 5-year follow-
­up, antireflux surgery was more effective than
omeprazole with respect to failure rates. However,
when the dose of omeprazole was adjusted in
case of recurrent symptoms, the failure rate still
revealed the superiority of surgery, but the differ-
ence was not significant (Lundell et al. 2001).
The highest population-based procedure rate
in the United States (45%) was performed in
infants younger than the age of 1  year (Lasser
et al. 2006). Given the option of spontaneous mat-
uration of esophageal function in infants younger
than the age of 1  year, concerns have been
expressed about the necessity of antireflux proce-
dures. Hassal discussed failure rates and results of
fundoplication in children and recommended sur-
gery only in selected patients (Hassal 2005).
58.8.2.2 Other Surgical Techniques
A rarely used method is Collis gastroplasty, dur-
ing which the fundus is separated parallel to the
esophagus in the angle of His by using a stapler,
and the fundoplicate is then fixed to a kind of
elongated esophagus, either by the Nissen or the
Toupet/Thal procedure. The procedure is mainly
used in cases of primary or secondary short
esophagus (esophageal atresia or chronic esopha-
gitis) to elongate the distance and perform the
fundoplication within the abdomen (Nakahara
et al. 2012).
A number of new endoscopic transoral techni-
cal methods have been introduced in the last few
years for the treatment of reflux in adults. The
oldest is the radiofrequency technique (Stretta)
and the TIF (transoral incisionless fundoplica-
tion) with the EsophyX-Z device. Long-term
follow-up studies and FDA approval exist for
both techniques. More recent techniques are the
use of Ultrasonic Surgical Endostapler and
GERDx device and the antireflux mucosectomy
(ARMS) procedure. A new, not endoscopic, tech-
772
nique which has raised interest is the LINX
device by which a ring of magnetic beads is
placed around the esophagus during laparoscopy.
Another non-endoscopic innovation is the
implantation of an electrical LES stimulator
which is supposed to improve the LES pressure
(for a review, see Nicolau et  al. 2018). For all
these techniques is a lack of sufficient experience
regarding the use of these methods in children.
In children with mental disabilities, one
should always try to create conditions for oral
(although modest) nutrition. In the case of insuf-
ficient intake of food, the fundoplication should
be combined with a gastrostomy button so that
food may be supplemented via gastrostomy. In
cases of severe mental disability, the patient may
occasionally develop numerous recurrent refluxes
due to extensive rumination. In order to preserve
the option of oral feeding, it would be advisable
to perform the esophagogastric dissection proce-
dure (Bianchi 1997). The stomach is severed
from the esophagus, and closed, and a gastros-
tomy with pyloroplasty is performed. An esoph-
agojejunostomy (as a Roux loop) is created to
preserve continuity. Thus, food taken orally
reaches the jejunum directly, and recurrent
refluxes or rumination are rendered impossible.
Long-term studies show nutritional and meta-
bolic complications, including dumping syn-
drome and chronic digestive malabsorption
(Tanaka et al. 2018).
58.8.2.3 Complications After
Fundoplication
A prospective multicenter study with 25 children
showed that laparoscopic fundoplication reduced
reflux symptoms, total acid exposure time, and
the number of acidic, as well as weakly acidic
reflux episodes, and the LES pressure increased,
but esophageal function and gastric emptying
were not affected. However, three patients had
persisting reflux, and in a further three, there was
a new-onset dysphagia present (Mauritz et  al.
2017). Thus, the most common complication is
recurrent reflux with an incidence as high as
10–15% and occurs with all three surgical proce-
dures but is associated with a higher periopera-
M. E. Höllwarth and V. Solari
tive risk. This complication is partly related to
natural motions in this region due to the motility
of the diaphragm during respiration and because
the esophagus is shortened during every act of
swallowing for about 5  cm in adults (Dodds
1974). Further known risk factors for recurrent
reflux are being aged below 6 years, the presence
of hiatus hernia, and mental disabilities accompa-
nied by frequent choking, belching, rumination,
and preoperative dystrophy; these phenomena
are frequently observed in children with mental
disabilities (Ngerncham et  al. 2007). It is of
upmost importance that children with mental dis-
abilities and marked catabolism should be fed
additionally with parenteral calories until they
clearly have an anabolic metabolism before per-
forming the surgical procedure in order to avoid
unnecessary complications and/or recurrences
(Weber 1995). Epilepsy, however, is not an addi-
tional risk factor for recurrent reflux in mentally
disabled patients (Goessler et  al. 2006; Vernon-­
Roberts and Sullivan 2007). Further causes of
recurrences are inadequate closure of the dia-
phragmatic crura, short esophagus, or inadequate
mobilization of the esophagus. Common late
complications are gas-bloat syndrome (mostly as
a consequence of the smaller gastric volume),
dysphagia, postoperative pain, and diarrhea.
However, most of the problems resolve spontane-
ously after a few months. After a Nissen fundo-
plication, a temporary inability to vomit may
occur. This complication is rare after the Toupet
procedure, during which the fundus is placed
about 270° dorsally around the esophagus, as
well as after the Thal procedure.
58.9 Conclusion
Gastroesophageal reflux (GER) in early infancy
is not caused by an insufficient LES but can be
seen as a delay of the motor development of the
esophagus with increased number and prolonged
LES relaxations. Spontaneous maturation usually
occurs within the first year and conservative
treatment is the appropriate therapy for most of
the cases. In contrast, chronic reflux leading to
58  Gastroesophageal Reflux Disease
symptoms such as recurrent spitting up, vomit-
ing, and failure to thrive, and recurrent exposure
of the esophageal mucosa to gastric acid is
defined as GERD.  Esophagitis is common and
replacement of the esophageal mucosa by gastric
or intestinal epithelia may lead to BE. Other dis-
eases, such as chronic respiratory tract infections
and asthma, may be related to GER. Eosinophilic
esophagitis is an atopic disease but often has
symptoms similar to a GERD.  Conservative
treatment with PPIs is used to prevent acidic
reflux. However, the number of reflux episodes is
usually not reduced and secondary symptoms
such as recurrent aspiration at nighttime and
respiratory infections are common. Babies with
esophageal atresia, diaphragmatic hernia, or
severe cerebral retardation suffer often from
severe reflux and may have additionally a hiatal
hernia. Surgical therapy is indicated in most of
these patients. Fundoplication by laparoscopy is
today the best choice of treatment for GERD in
children and adults. In recent years, a number of
endoscopic antireflux procedures have been pro-
posed. Despite early promising results, long-term
quality does not withstand the established clini-
cal efficacy of conservative treatment with PPI or
conventional fundoplication for children.
References
Alterio T et  al (2019) Eosinophilic esophagitis in chil-
dren: current knowledge to open new horizons. Scand
J Gastroenterol 54:822–829
Avansino JA et  al (1999) Characterization and manage-
ment of paraesophageal hernias in children after anti-
reflux operation. J Pediatr Surg 34:1610–1614
Babaei A, Mittal R (2018) Cholecystokinin induces
esophageal longitudinal muscle contraction and tran-
sient lower esophageal sphincter relaxation in healthy
individuals. Am J Physiol Gastrointest Liver Physiol
315:G734–G742
Bai X et  al (2019) Involvement of different receptor
subtypes in prostaglandin E2-induced contraction
and relaxation in the lower esophageal sphincter
and esophageal body. Eur J Pharmacol 857:172405.
https://doi.org/10.1016/j.ejphar.2019.172405
Barrett NR (1950) Chronic peptic ulcer of the oesophagus
and ‘oesophagitis’. Br J Surg 38:175–182
Bar-Server Z (2017) Scintigraphic evaluation of gastro-
esophageal reflux and pulmonary aspiration. Semin
Nucl Med 47:275–285
773
Berni CR, Terrin G (2010) Gastric acidity inhibi-
tors and the risk of intestinal infections. Curr Opin
Gastroenterol 26:31–35
Bianchi A (1997) Total esophagogastric dissociation: an
alternative approach. J Pediatr Surg 32:1291–1294
Blevins CH et  al (2018) The esophageal epithelial bar-
rier in health and disease. Clin Gastroenterol Hepatol
16:608–617
Blondeau K et al (2011) The relationship between gastro-
esophageal reflux and cough in children with chronic
unexplained cough using combined impedance-pH-­
manometry recordings. Pediatr Pulmonol 46:286–294
Campanozzi A et  al (2009) Prevalence and natural his-
tory of gastroesophageal reflux: pediatric prospective
study. Pediatrics 123:779–783
Cannon WB (1911) Conditions affecting the activities of
the cardia. In: Arnold E (ed) The mechanical factors of
digestion. Longmans, Green& Co., New York
Carré IJ, Astley R (1958) The gastro-oesophageal junc-
tion in infancy. A combined cineradiographic and
manometric study. Thorax 13:159–164
Carroll MW, Jacobson K (2012) Gastroesophageal reflux
disease in children and adolescents. Pediatr Drugs
14:79–89
Castellani C et al (2014) Omeprazole for reflux in infants:
effectiveness determined by impedance-pH monitor-
ing. Pediatr Surg Int 30:381–385
Chang AB et  al (2006) Gastro-oesophageal reflux treat-
ment for prolonged nonspecific cough in children and
adults. Cochrane Database Syst Rev 4:1–46
Code CF et  al (1956) The gastroesophageal sphinc-
ter in healthy human beings. Gastroenterologia 86:
135–150
Cohen S (1975) Recent advances in management of gas-
troesophageal reflux. Postgrad Med 57:97–102
Cossentino MJ et al (2012) Randomized clinical trial: the
effect of baclofen in patients with gastro-oesophageal
reflux  – a randomized prospective study. Aliment
Pharmacol Ther 35:1036–1044
Davies I et  al (2015) Gastro-oesophageal reflux in chil-
dren. NICE guidance. BMJ 350:g7703
Debley JS et  al (2006) Prevalence and impact of gas-
troesophageal reflux in adolescents with asthma. A
population-­based study. Pediatr Pulmonol 41:475–481
Delshad SD et al (2020) Prevalence of gastroesophageal
reflux disease and proton pump inhibitor-refractory
symptoms. Gastroenterology 158:1250–1261
Dent J (2006) Review article: from 1906–2006 – a cen-
tury of major evolution of understanding of gastro-­
esophageal reflux disease. Aliment Phamacol Ther
24:1269–1281
Dent J (2009) Landmarks in the understanding and
treatment of reflux disease. J Gastroenterol Hepatol
24:S5–S14
Dent J et al (1980) Mechanism of gastroesophageal reflux
in recumbent asymptomatic human subjects. J Clin
Invest 65:256–267
Dodds WJ (1974) Effect of esophageal movement
on intraluminal esophageal pressure recordings.
Gastroenterology 67:1059–1063
774
Dranove JE et al (2009) Predictors of response to proton
pump inhibitor therapy among children with signifi-
cant esophageal eosinophilia. J Pediatr 154:96–100
Du X et  al (2017) Laparoscopic Nissen (total) versus
anterior 180ο fundoplication for gastro-esophageal
reflux disease: a meta-analysis and systematic review.
Medicine (Baltimore) 96:e8085
Edwards DAW (1982) The anti-reflux mechanism, its
disorders and their consequences. Clin Gastroenterol
11:479–496
Elitsur Y et al (2009) Obesity is not a risk factor in chil-
dren with reflux esophagitis: a retrospective analysis
of 738 children. Metab Syndr Relat Disord 7:211–214
El-Serag HB et al (2007) Determinants of gastroesopha-
geal reflux disease in adults with a history of childhood
gastroesophageal reflux disease. Clin Gastroenterol
Hepatol 5:696–701
Field SK et  al (1999) The effects of antireflux surgery
on asthmatics with gastroesophageal reflux. Chest
116:766–774
Forsell L et al (2013) Increased risk of Barrett’s esopha-
gus among individuals born preterm or small for ges-
tational age. Clin Gastroenterol Hepatol 11:790–794
Fotter R et al (1985) Correlation between manometric and
roentgenologic findings of diseases of the esophagus
in infants and children. Prog Pediatr Surg 18:14–21
Frisby CL et  al (2005) Inhibition of transient lower
esophageal sphincter relaxation and gastroesophageal
reflux by metabotropic glutamate receptor ligands.
Gastroenterology 129:995–1004
Fujiwara Y et  al (2013) Gastroesophageal reflux and
sleep. Gastroenterol Clin N Am 42:57–70
Funderburk A et al (2016) Temporal association between
reflux-like behaviors and gastroesophageal reflux in
preterm and term infants. JPGN 62:556–561
Furuta GT et  al (2007) Eosinophilic esophagitis in
children and adults: a systemic review and consen-
sus recommendation for diagnosis and treatment.
Gastroenterology 133:1342–1363
Goessler A et  al (2006) Does epilepsy influence the
outcome of antireflux procedures in neurologically
impaired children. Pediatr Surg Int 22:485–490
Goessler A et  al (2007) Gastroesophageal reflux and
behavior in neurologically impaired children. J Pediatr
Surg 42:1486–1490
Gutiérrez-Junquera C et  al (2016) High prevalence of
response to proton-pump inhibitor treatment in chil-
dren with esophageal eosinophilia. JPGN 62:704–710
Hambraeus M et  al (2013) A literature review of the
outcomes after robot-assisted laparoscopic and con-
ventional laparoscopic Nissen fundoplication for
gastro-esophageal reflux disease. Int J Med Robot
9:428–432
Hassal E (2005) Outcomes of fundoplication: causes for
concern, newer options. Arch Dis Child 90:1047–1052
Hassall E et al (2007) Characteristics of children receiv-
ing proton pump inhibitors continuously for up to 11
years duration. J Pediatr 150:262–267
He S et al (2019) 24-hour monitoring of transient lower
esophageal sphincter relaxation events by long-term
M. E. Höllwarth and V. Solari
high-resolution impedance manometry in normal vol-
unteers: the “mirror phenomenon”. Neurogastroenterol
Motil 31:e13530. https://doi.org/10.1111/nmo.13530.
Epub 2019
Hirsch DP et  al (1998) Involvement of nitric oxide
in human transient lower esophageal sphincter
relaxations and esophageal primary peristalsis.
Gastroenterology 115:1374–1380
Höllwarth ME (1979) Entwicklung der Speiseröhrenfunk-
tion bei Neugeborenen – eine manometrische Untersu-
chung. Z Kinderchir 27:201–215
Höllwarth ME (2012) Gastroesophageal reflux disease.
In: Coran AG (ed) Pediatric surgery, 7th edn. Elsevier
Höllwarth ME, Uray E (1985) Physiology and pathophys-
iology of the esophagus in childhood. Pediatr Surg Int
18:1–13
Hsing T-Y et  al (2019) Role of esophageal manometry
in children with refractory gastroesophageal reflux
symptoms. Pediatr Int 61:807–811
Ingelfinger FJ (1958) Esophageal motility. Physiol Rev
38:533–584
Jodorkovsky D, Price JC, Kim B et al (2014) Multichannel
intraluminal impedance-pH testing is clinically useful
in the management of patients with gastroesophageal
reflux symptoms. Dig Dis Sci 59:252–255
Kaltenbach T et  al (2006) Are lifestyle measures effec-
tive in patients with gastroesophageal reflux dis-
ease? An evidence-based approach. Arch Intern Med
166:965–971
Koebnick C et  al (2011) Extreme childhood obesitiy is
associated with increased risk for gastroesophageal
reflux disease in a large population-based study. Int J
Pediatr Obes 6:e275–e263
Kurz R et al (1985) Combined disturbance of respiratory
regulation and esophageal function in early infancy.
Prog Pediatr Surg 18:52–61
Landler U et  al (1990) Esophageal function of infants
with sudden infant death-risk. Klin Padiatr 202:37–42.
(German)
Lasser MS et  al (2006) National trends in the use
of antireflux procedures for children. Pediatrics
118:1828–1835
Lee WS et  al (1999) Gastro-oesophageal reflux: clini-
cal profiles and outcome. J Paediatr Child Health
35:568–571
Lidums I et al (2000) Control of transient lower esophageal
relaxations and reflux by the GABAβ agonist baclofen
in normal subjects. Gastroenterology 118:7–13
Liebermann-Meffert D et al (1979) Muscular equivalent
of the lower esophageal sphincter. Gastroenterology
76:31–38
Lin EC et al (2019) Low prevalence of suspected Barrett’s
esophagus in patients with gastroesophageal reflux
disease without alarm symptoms. Clin Gastroenterol
Hepatol 17:857–863
Lo W-K, Chan WW (2013) Proton pump inhibitor use and
the risk of small bacterial overgrowth: a meta-analysis.
Clin Gastroenterol Hepatol 11:483–490
Lundell L et  al (2001) Continued (5 year) follow-up of
a randomized clinical study comparing antireflux sur-
58  Gastroesophageal Reflux Disease
gery and omeprazole in gastroesophageal reflux dis-
ease. J Am Coll Surg 192:172–181
Lupu VV et  al (2016) Correlation between the different
pH-metry scores in gastroesophageal reflux disease in
children. Medicine (Baltimore) 95:e3804. https://doi.
org/10.1097/MD.0000000000003804
Maes S et al (2016) Review: surveillance of patients with
Barrett esophagus. Best Pract Res Clin Gastroenterol
30:901–912
Martigne L et  al (2012) Prevalence and management of
gastroesophageal reflux disease in children and ado-
lescents: a nationwide cross-sectional observation
study. Eur J Pediatr 171:1767–1773
Martin AJ et al (2002) Natural history and familial rela-
tionship of infants spilling to 9 years of age. Pediatrics
109:1061–1067
Mauritz FA et al (2017) Effect and efficacy of laparoscopic
fundoplication in children with GERD; a prospective,
multicenter study. Surg Endosc 31:1101–1110
Modlin IM et al (2004) Gastroesophageal reflux disease.
Then and now. J Clin Gastroenterol 38:390–402
Moussa H, Hassan M (2017) Gastroesophageal reflux dis-
ease. Pediatr Clin N Am 64:487–505
Naik RD, Vaezi MF (2015) Extra-esophageal gastro-
esophageal reflux disease and asthma: understand-
ing this interplay. Expert Rev Gastroenterol Hepatol
9:969–982
Nakahara Y et  al (2012) Modified Collis-Nissen proce-
dure for pure long-gap esophageal atresia. J Pediatr
Surg 47:262–266
Ngerncham M et al (2007) Risk factors for recurrent gas-
troesophageal reflux disease after fundoplication in
pediatric patients: a case-control study. J Pediatr Surg
42:1478–1485
Nicolau AE et  al (2018) New minimally invasive endo-
scopic and surgical therapies for gastroesophageal
reflux disease (GERD). Chirurgia 113:70–82
Nissen R (1979) Fundoplikation und Gastropexie bei
Refluxkrankheit und Hiatushernie. G. Thieme Verlag,
New York
Nylund CM et al (2014) Association of Clostridium dif-
ficile infections with acid suppression medications in
children. J Pediatr 165:979–984
Orenstein SR et al (2006) Natural history of infant reflux
esophagitis: symptoms and morphometric histol-
ogy during one year without pharmacotherapy. Am J
Gastroenterol 101:628–640
Palmer ED (1954) Hiatus hernia: the problem of diagno-
sis. J Thorac Surg 27:271–276
Pandolfino JE et al (2006) Transient lower esophageal
sphincter relaxations and reflux: mechnistic analysis
using concurrent fluoroscopy and High-resolution
manometry. Gastroenterology 131:1725–1733
Patel et  al (2019) Development and validation of a
mucosal impedance contour analysis system to dis-
tinguish esophageal disorders. Gastoenterology
156:1617–1626
Pesendorfer P et al (1993) Long-term follow-up in infants
with pathological gastroesophageal reflux. Klin
Padiatr 205:363–366
775
Poddar U (2019) Gastroesophageal reflux disease (GERD)
in children. Pediatr Int Child Health 39:7–12
Quitadamo P et al (2012) Total and abdominal obesity are
risk factors for gastroesophageal reflux symptoms in
children. JPGN 55:72–75
Quitadamo P et  al (2015) Gastroesophageal reflux in
young children and adolescents: is there a relation
between symptom severity and esophageal histologi-
cal grade? JPGN 60:318–321
Raicevic M, Saxena AK (2018) Barrett’s esophagus
in children: what is the evidence? World J Pediatr
14:330–334
Ravelli AM et al (2006) Pulmonary aspiration shown by
scintigraphy in gastroesophageal reflux-related respi-
ratory disease. Chest 130:1520–1526
Reed CC, Dellon ES (2019) Eosinophilic esophagitis.
Med Clin North Am 103:29–42
Rhee D et  al (2011) Population based comparison of
open vs laparoscopic esophagogastric fundoplication
in children: application of the Agency for Healthcare
Research and Quality pediatric quality indicators. J
Pediatr Durg 46:648–654
Rosen R et  al (2018) Pediatric gastroesophageal reflux
clinical practice guidelines: joint recommenda-
tions of the North American Society for Pediatric
Gastroenterology, Hepatology, and Nutrition and the
European Society for Pediatric Gastroenterology,
Hepatology, and Nutrition. JPGN 66:516–554
Rossi P et al (2018) Combined multichannel intraluminal
impedance and pH monitoring is helpful in managing
children with suspected gastro-oesophageal reflux dis-
ease. Dig Liver Dis 50:910–915
Ruigómez A et al (2017) Utilization and safety of proton-­
pump inhibitors and histamin-2 receptor agonists
in children and adolescents: an observational cohort
study. Curr Med Res Opin 33:2201–2209
Safe M et  al (2016) Combined multichannel intralumi-
nal impedance and pH measurement in detecting
gastroesophageal reflux disease in children. JPGN
63:e98–e106
Savarino V et al (2009) Proton pump inhibitors in GORD
an overview of their pharmacology. Pharmacol Res
59:135–153
Schneider N et al (2014) Validation study of the Esohisto
consensus guidelines for recognition of micro-
scopic esophagitis (histoGERD trial). Hum Pathol
45:994–1002
Semeniuk J, Kaczmarski M (2007) 24-hour esopha-
geal pH-monitoring in children suspected of gastro-
esophageal reflux disease: analysis of intraesophageal
pH monitoring values recorded in distal and proxi-
mal channel at diagnosis. World J Gastroenterol
13:5108–5115
Singendonk MMJ et al (2018) Pediatric gastroesophageal
reflux disease: systematic review on prognosis and
prognostic factors. JPGN 66:239–243
Slater BJ, Rothenberg St S (2017) Gastroesophageal
reflux. Semin Pediatr Surg 26:56–60
Slocum C et al (2007) Infant apnea and gastroesophageal
reflux: a critical review and framework for further
investigation. Curr Gastroenterol Rep 9:219–224
776
Souza RF et al (2009) Gastroesophageal reflux might
cause esophagitis through a cytokine-mediated mech-
anism rather than caustic acid injury. Gastroenterology
137:1776–1784.
Steele D et  al (2019) Evolving screening and surveil-
lance techniques for Barrett’s esophagus. World J
Gastroenterol 25:2045–2057
Stelzner F, Lierse W (1968) The angiomuscular dilatation
closing the terminal esophagus. Langenbecks Arch
Chir 321:35–64
Strate U et al (2008) Laparoscopic fundoplication
Stylopoulos N, Rattner DW (2005) The history of hiatal
hernia surgery. From bowditch to laparoscopy. Ann
Surg 241:185–193
Tanaka Y et  al (2018) Indications for total esophago-
gastric dissection in children with gastroesophageal
reflux. Surg Today 48:971–977
Thakkar K et  al (2010) Gastroesophageal reflux and
asthma in children: a systematic review. Pediatrics
125:e925–e930
Thal AP et al (1965) New operation for distal esophageal
stricture. Arch Surg 90:464–472
Tolia V, Boyer K (2008) Long-term proton pump inhibitor
use in children: a retrospective review of safety. Dig
Dis Sci 53:385–393
Tolia V, Vandenplas Y (2009) Systemic review: the
extra-oesophageal symptoms of gastro-oesophageal
reflux disease in children. Aliment Pharmacol Ther
29:258–272
Tomita R et  al (2003) Physiological studies on nitric
oxide in the lower esophageal sphincter of patients
with reflux esophagitis. Hepato-Gastroenterology
50:110–114
Toupet A (1963) Technique d’esophago-gastroplastie
avec phreno-gastropexie dans la cure radicale des
hernies hiatales et comme complement de l’operation
de Heller dans les cardiospasmes. Mem Acad Chir
89:394–399
Tovar JA et al (2007) Pediatric surgeons and gastroesoph-
ageal reflux. J Pediatr Surg 42:277–283
Tsai C-C et  al (2019) Long-chain fatty acid receptors
mediate relaxation of the porcine lower esopha-
geal sphincter. Front Physiol 10:676. https://doi.
org/10.3389/fphys.2019.00676
Ummarino D et  al (2012) Impact of antisecretory treat-
ment on respiratory symptoms of gastroesophageal
reflux disease in children. Dis Esophagus 25:671–677
M. E. Höllwarth and V. Solari
Vadlamudi NB et al (2013) Baclofen for the treatment of
pediatric GERD. JPGN 57:808–812
van der Pol RJ et al (2013) Outcomes of endoscopy and
novel pH-impedance parameters in children: is there a
correlation? JPGN 56:196–200
Vandenplas Y, Hauser B (2015) An updated review on
gastro-esophageal reflux in pediatrics. Expert Rev
Gastroenterol Hepatol 9:1511–1521
Vandenplas Y et al (2007) Gastro-oesophageal reflux dis-
ease: oesophageal impedance versus pH monitoring.
Acta Paediatr 96:956–962
Vernon-Robert A, Sullivan PB (2007) Fundoplication ver-
sus post-operative medication for gastro-­oesophageal
reflux in children with neurological impairment
undergoing gastrostomy. Cochrane Database Syst Rev
1:1–13
Wang Y-J et  al (2020) Salivary pepsin as an intrinsic
marker for diagnosis of sub-types of gastroesopha-
geal reflux disease and gastroesophageal reflux
disease-­related disorders. J Neurogastroenterol Motil
26:74–84
Wasilewska A et al (2012) Respiratory response to proton
pump inhibitor treatment in children with obstructive
sleep apnea syndrome and gastroesophageal reflux.
Sleep Med 13:824–830
Weber TR (1995) A prospective analysis of factors influ-
encing outcome after fundoplication. J Pediatr Surg
30:1061–1064
Whitson MJ, Falk GW (2015) Predictors of progression to
high-grade dysplasia or adenocarcinoma in Barrett’s
Esophagus. Gastroenterol Clin North 44:299–315
Winter HS et al (2011) Review of the persistence of gastro-
esophageal reflux disease in children, adolescents and
adults: does gastroesophageal reflux disease in adults
sometimes begin in childhood? Scand J Gastroenterol
46:1157–1168
Zhang P et al (2016) Laparoscopic vs. open Nissen’s fun-
doplication for gastro-oesophageal reflux disease in
children: a meta-analysis. Int J Surg 34:10–16
Zhang M et al (2019) Assessing different diagnostic tests
for gastroesophageal reflux disease: a systematic review
and network meta-analysis. Ther Adv Gastroenterol
12. https://doi.org/10.1177/1756284819890537
 Esophageal Replacement
Shilpa Sharma and Devendra K. Gupta
59.1 Introduction
The esophagus in children may need to be
replaced following any congenital anomaly or
acquired disease, such as corrosive ingestion.
Surgical interventions may be needed for dilata-
tion, stricture, and reflux. Long-term complica-
tions of colonic interposition include reflux,
halitosis, colonic segment dilatation, and anas-
tomotic stricture. As the stomach has a good vas-
cular supply, the gastric tube and gastric
transposition are both in vogue with many sur-
geons. Gastric transposition involves only a sin-
gle anastomosis in the neck and no suture lines in
the thorax or the abdomen. However, the imme-
diate postoperative period may be rough, espe-
cially in infants, and require constant monitoring
and artificial ventilation (Sharma and Gupta
2019). A gastric tube involves a technically
demanding procedure to fashion a watertight
tube. There is a long suture line in the chest,
which is associated with serious early complica-
tions such as a prolonged leak in the neck or
mediastinum, graft necrosis, or ischemia leading
to stricture of the tube. However, the lumen
S. Sharma (*)
Department of Pediatric Surgery, All India Institute of
Medical Sciences, New Delhi, India
D. K. Gupta
Super Speciality Pediatric Hospital and Post Graduate
Teaching Institute, Noida, India
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_59
59
matches that of the esophagus and the long-term
results are favorable. Jejunal interposition is not
very popular and is technically demanding due to
its precarious blood supply to achieve the length
of the jejunum to reach the neck, requiring the
expertise for the microvascular surgery. Chances
of graft loss are high if not performed by the
experts in the field.
59.2 Historical Overview
In 1894, Bircher first devised a neoesophagus by
creating a subcutaneous skin tube to fill the gap
between the proximal esophagus and the stom-
ach. While this procedure allowed patients to eat
as normally as possible, the frequency of fistulae
and strictures was high (Sherman and Waterston
1957).
The colon was the first organ to be used for
esophageal replacement. Lundblad had used the
colon for reconstructing the esophagus in a
3-year-old child suffering from a corrosive stric-
ture. Years later in 1934, he also showed that such
an esophagus grew with the child.
Since then, various methods have been tried
both clinically and experimentally. These have
included the use of stomach, skin tubes, jejunum,
colon, aorta, inert tubes, etc. Sandblom (1948)
was the first to use the colon for reconstruction in
atresia of the esophagus, and it was his work that
encouraged Waterston to attempt this method
777
778
using a slightly different technique (Sherman and
Waterston 1957). In 1957, Sherman and
Waterston modified this technique, creating a
transthoracic interposition graft. In 1982,
Freeman and Case modified the technique by
using a posterior route through the esophageal
hiatus.
The stomach was also used as an esophageal
substitute by creating a gastric tube or by trans-
posing the entire stomach. Described by both
Gavriliu and Heimlich in the mid-1950s, the
reversed gastric tube approach was later adapted
for children by Burrington, Anderson, and
Randolph (Azizkhan 2014). The key proponents
in the use of the entire stomach for esophageal
replacement were Spitz and Coran.
The fourth esophageal replacement technique,
jejunal interposition, was performed by Roux in
1907 and Herzen in 1908 and adapted for use in
children by Leven and Varco in the 1950s. Jejunal
interposition has been popular with some for its
narrow lumen (Azizkhan 2014). However, it has
not been used extensively worldwide due to pre-
carious vascularity. The long-term results are
good in expert hands.
The follow-up of esophageal replacement is
essential to understand the outcomes and the
complications encountered (Sharma and Gupta
2017). Postoperative complications include gas-
tric stasis, bile reflux, restricted growth, and
decreased pulmonary functional capacity.
59.3 Indications for Esophageal
Replacement
The indications for esophageal replacement in
children are given in Table  59.1 (Sharma and
Gupta 2017).
An ideal esophageal substitute should be tech-
nically simple, reproducible, and adaptable for
small children with minimal complications and
good long-term results. It should grow with the
age of the child and last for a lifetime with mini-
mal gastric acid reflux. It should allow normal
swallowing without causing mediastinal com-
pression or compromise to the respiratory or car-
S. Sharma and D. K. Gupta
Table 59.1  Indications for esophageal replacement
Congenital
   Pure esophageal atresia
  Esophageal atresia with tracheoesophageal fistula
with a long gap
  Esophageal atresia with tracheoesophageal fistula
with failed repair
   Long esophageal stenosis
   Congenital short esophagus
   Neuromotor abnormalities
Acquired
  Stricture of the esophagus (alkali ingestion—
liquid or solid pellets)
   Button battery injury to esophagus
   Acid esophageal burns
   Perforation of the esophagus
   Long esophageal stenosis
   Epidermolysis bullosa
   Reflux esophagitis with stricture
   Indolent stricture of the esophagus
   Advanced obliterative esophagitis
diac function. The vascularity should be good to
prevent major postoperative complications. It
should not become tortuous or redundant and
should not have increased malignant potential
(Soccorso and Parikh 2016).
59.3.1 Age for Esophageal
Replacement
The appropriate time for undertaking esophageal
replacement is after infancy as the vascularity and
lengths of the neo conduits are reliable. At this
age, the child can sit and eat without much risk of
aspiration from the conduit. However, in patients
with pure esophageal atresia or following leak for
esophageal atresia, replacement has also been
performed by the authors in the neonatal period
(Gupta et al. 2007; Gupta and Sharma 2011).
59.4 Routes for Esophageal
Replacement
The route to place the graft is a major decisive
factor in planning the procedure as surgeons
would prefer to use a shorter route. The
59  Esophageal Replacement 779

retrosternal route was found to be 1.9 cm longer

than the transhiatal route on autopsy studies
(Sharma and Gupta 2017). On the contrary, it
was demonstrated that the retrosternal route is
about 3 cm shorter than the posterior mediasti-
nal route when the stomach is used for the
esophageal ­ replacement (Chen et  al. 2009).
Thus, the retrosternal route is shorter if the stom-
ach is used and the transhiatal route becomes
shorter if the colon is to be used (Sharma and
Gupta 2017).
Currently, there are two popular routes:

1. Substernal (retrosternal): The substernal route

can be created easily by blunt finger dissec-
tion, starting just behind the xiphisternum up
to the neck in front of the trachea with an ade- Fig. 59.1  The patient is placed in a supine position
quately created wider space to house the with a sand bag under the shoulder and the neck turned
conduit. to the opposite side from the site of the cervical esopha-
gostomy. Reproduced with permission from Springer,
2. Mediastinal (transhiatal, native, orthotopic):
Gupta DK, Sharma S.  Colonic interposition for
This route is made by digital dissection pro- Esophageal replacement. 2nd Edition Pediatric Surgery
gressively upwards from the hiatus strictly in (Springer Surgery Atlas Series) Prem Puri & Hollwarth
front of the vertebral body. A retro tracheal M.  Germany Springer 2019. Chapter 9: 59–65
https://doi.org/10.1007/978-3-662-56282-6_9
digital dissection is gently directed down-
wards, within the posterior mediastinum in
front of the prevertebral facia. The transhiatal
Esophagus
and cervical operator’s fingers should get
together and finalize a wide and comfortable
passage.

59.5 Mobilization of the Cervical

Esophagostomy

59.5.1 Operative Procedure

This step is the same for all esophageal replace-

ment procedures.
The patient is placed in a supine position with
a sand bag under the shoulder to extend the neck.
The neck is turned to the opposite side from the
site of the cervical esophagostomy. The cervical
esophagostomy is mobilized, taking care to pre-
serve the recurrent laryngeal nerve during mobi-
lization. The esophagus is mobilized for 2–3 cm.
The blood supply to the cervical esophagus is
submucosal and usually not affected by extensive
mobilization in the neck (Figs. 59.1 and 59.2). A
soft red rubber catheter is placed into the esopha-
Strap muscle
Recurrent laryngeal nerve
Sternocleidomastoid
muscle
Fig. 59.2  The cervical fascia is opened along the anterior
border of the sternomastoid. Reproduced with permission
from Springer, Gupta DK, Sharma S. Colonic interposi-
tion for Esophageal replacement. 2nd Edition Pediatric
Surgery (Springer Surgery Atlas Series) Prem Puri &
Hollwarth M. Germany Springer 2019. Chapter 9: 59–65
https://doi.org/10.1007/978-3-662-56282-6_9
780
gus through the mouth to allow identification of
the esophagus during neck dissection. A right
transverse supraclavicular incision is made for
about 3–4 cm around the stoma.
The esophagostomy (following diversion for
complicated esophageal atresia) is mobilized
with an elliptical incision, after applying stay
sutures around it. The cervical fascia is opened
along the anterior border of the sternomastoid.
As the deeper dissection continues, the strap
muscles are either divided or retracted in the
line of incision. The carotid sheath, with the
internal jugular vein and common carotid
artery, is gently retracted laterally. The recur-
rent laryngeal nerve is identified and retracted
medially.
59.6 Colon Interposition
A colon segment of adequate length is chosen,
depending upon the suitable vessel found at sur-
gery. It could be iso- or antiperistaltic. The length
of the graft must be measured with all due pre-
caution, keeping a few centimeters extra on either
side. The graft usually shrinks once disconnected
from the mainline. It is not the total length of the
graft that is important but the length that reaches
the neck without tension to the vessels. Usually,
it is 2/3 for the chest and 1/3 from the vessel base
for the abdominal side.
The middle colic artery, branch of the superior
mesenteric artery, has two branches: right and
left, which anastomose freely with the adjoining
vessels to supply the right, transverse, and the left
colon. Normally, the left branch is more promi-
nent and supplies the larger segment of the trans-
verse colon. However, in about one-quarter of
cases, the right branch is more prominent. Colon
interposition is based on middle colic branch of
the superior mesenteric artery or the left ascend-
ing colic branch of the inferior mesenteric artery,
passing behind the stomach, with a cologastric
anastomosis with the posterior wall of the stom-
ach in the abdomen and an esophago-colonic
anastomosis in the neck.
S. Sharma and D. K. Gupta
59.6.1 Advantages of Colonic
Interposition
1. Reliable vascular pedicle due to the marginal
artery
2. Adequate length of graft
3. Feasibility to use in both iso- and antiperistal-
tic segments
4. Easily courses over the liver edge
5. Relatively safe and easy to perform
6. Less infrequent and less serious complica-
tions compared to those seen with other pro-
cedures (Sharma and Gupta 2017)
The left conduit is preferred over the right
colon because it is less bulky and has a more reli-
able blood supply. A good vascular supply and
isoperistaltic loop have been suggested to be ben-
eficial in the prevention of anastomotic leakage
and intestinal ischemia (Sharma and Gupta
2017). The optimal artery of colic segment dur-
ing replacement is the left colic followed by the
middle colic artery. With the middle colic artery,
its right colic artery is variable; thus the use of
ascending branch of the left colic artery is more
dependable for the blood supply to the transverse
colon when planned for replacement of esopha-
gus in an isoperistaltic fashion.
59.6.2 Operative Procedure
A midline incision is made over the abdomen
from the xiphisternum to the umbilicus and
extended down around the umbilicus if need be.
The transverse colon is mobilized from the
ascending to the descending colon and exterior-
ized to assess the vascular supply. The gastros-
tomy, when present, is assessed for the site and
the need for closure. The colonic graft is chosen
based on the blood supply. Division of the middle
colic vessels will be required. Before dividing the
major vessels, it is mandatory to use a bulldog
clamp to verify adequate circulation to the pro-
posed segment of colon (Fig.  59.3) (Gupta and
Sharma 2019).
59  Esophageal Replacement
Resection
line
Fig. 59.3  A midline incision is made over the abdomen
from the xiphisternum to the umbilicus and extended
down around the umbilicus if need be. A bulldog clamp is
used to verify adequate circulation to the proposed seg-
ment of colon. Reproduced with permission from
The left triangular ligament of the liver is
incised. The esophagus is then dissected all
around at the hiatus after the division of the
phrenoesophageal ligament (Fig. 59.4). The right
and the left vagi are divided. The esophagus is
freed as high as possible. The blunt dissection is
continued from above and below until the esoph-
agus is freed completely. It is essential to avoid
aggressive dissection in the region of the aortic
arch and to stay close to the esophageal wall. If
the esophagus is badly adherent and impossible
to mobilize safely, it should be left in situ without
causing unnecessary injury to neighboring struc-
tures. In that case, one can switch over to the sub-
sternal (retrosternal) route for the replacement.
Once the esophagus has been freed from all its
attachments, an esophagectomy is performed by
dividing the esophagus in the neck and pulling it
down to the abdomen (Fig. 59.5). The gastric end
is closed in two layers. A red rubber catheter with
the silk suture attached is left in the mediastinum,
to be used for guidance for the later passage of
the colon through the hiatus. The colon is reeval-
781
Middle colic artery
Springer, Gupta DK, Sharma S. Colonic interposition for
Esophageal replacement. 2nd Edition Pediatric Surgery
(Springer Surgery Atlas Series) Prem Puri & Hollwarth
M. Germany Springer 2019. Chapter 9: 59–65 https://doi.
org/10.1007/978-3-662-56282-6_9
uated and the pulsation of the marginal artery is
examined carefully. The exact measurement of
the colon is reassessed after the esophageal resec-
tion and the colon is divided between noncrush-
ing intestinal clamps. The isolated colonic
segment should be kept warm and moist to keep
the vessels and graft supple. It should be cleaned
on the table with normal saline and diluted povi-
done (1:100) iodine until clear. After the wash,
the colon graft is kept open on both sides without
using any clamps (Fig. 59.6).
The adequacy of the colon segment length is
reassessed and the colon is placed in the desired
place without any tension (Fig. 59.7). The colon
graft on its pedicle is passed behind the stomach
to the desired route. The excess of the pulled
colon, if any, is trimmed and the vascularity is
again reassessed.
A passage is created behind the stomach,
through which the colon is taken up in an isoperi-
staltic manner (Fig. 59.8). Care should be taken
to avoid causing any torsion of the pedicle or
traction of the vessels. Viability of the graft is
782 S. Sharma and D. K. Gupta

Esophagus

Stomach
Liver

Anterior vagal nerve

Fig. 59.4  The esophagus is dissected all around at the

hiatus after the division of the phrenoesophageal liga-
ment. Reproduced with permission from Springer,
Gupta DK, Sharma S.  Colonic interposition for
Esophageal replacement. 2nd Edition Pediatric Surgery
(Springer Surgery Atlas Series) Prem Puri & Hollwarth
M.  Germany Springer 2019. Chapter 9: 59–65
https://doi.org/10.1007/978-3-662-56282-6_9

confirmed by noting the bleeding from its cervi-
cal end. If the esophagus has not been resected
and a colon bypass procedure is planned, then a
retrosternal tunnel is made by blunt dissection,
dividing the endothoracic fascia at the upper end
from the neck incision after detaching the strap
muscles at the suprasternal notch and at the lower
end by separating the diaphragm and the rectus
abdominis muscle slips from the xiphoid carti-
lage and the lower sternum. The tunnel is enlarged
using fingers, remaining close to the under sur-
face of the sternum and taking care not to injure
the pleura. A red rubber catheter with a long silk
suture is passed through the tunnel.
The anastomosis in the neck is usually per-
formed first (Fig. 59.9). If there is not much dis-
Fig. 59.5  Once the esophagus has been freed from all
its attachments, esophagectomy is done by dividing the
esophagus in the neck and pulling it down to the abdo-
men. Reproduced with permission from Springer,
Gupta DK, Sharma S.  Colonic interposition for
Esophageal replacement. 2nd Edition Pediatric Surgery
(Springer Surgery Atlas Series) Prem Puri & Hollwarth
M.  Germany Springer 2019. Chapter 9: 59–65
https://doi.org/10.1007/978-3-662-56282-6_9
parity in size, a wide and single-layer end-to-end
esophago-colonic anastomosis is performed,
using absorbable interrupted sutures. The colon
is fixed with the prevertebral or deep cervical
facia to prevent sagging. If the colonic end is
slightly bigger, the esophagus is spatulated pos-
teriorly to accommodate the large size of the
colon for the anastomosis. If the esophagus is
59  Esophageal Replacement
Fig. 59.6  The exact measurement of the colon is reas-
sessed after the esophageal resection and the colon is
divided between noncrushing intestinal clamps.
Reproduced with permission from Springer, Gupta DK,
Sharma S.  Colonic interposition for Esophageal
replacement. 2nd Edition Pediatric Surgery (Springer
Surgery Atlas Series) Prem Puri & Hollwarth
M.  Germany Springer 2019. Chapter 9: 59–65
https://doi.org/10.1007/978-3-662-56282-6_9
markedly smaller in diameter than the colon, an
end-to-side esophago-colonic anastomosis is
made, with closure of the colonic stump. In a
border line tension of the colon graft to reach
the neck, multiple relaxing transverse incisions
may be given on the taenia coli to elongate the
graft. This may add 1.5–2 cm additional length
of the graft (Sharma and Gupta 2017). Fixation
of the colon to the neck muscles is done to avoid
traction on the neoanastomosis. Suturing of the
strap muscles is important to support the neck
783
Fig. 59.7  Assessment of the length of colon to be used as
replacement
during swallowing. The wound is closed in lay-
ers, leaving a drain in place.
In cases of caustic strictures reaching high to
the pharynx, an end-to-side pharyngocolic anas-
tomosis to the wall of the pharynx is created, lat-
eral to the thyroid cartilage. First, a hockey-stick
type of incision should extend up from the lower
neck to the angle of the mandible. Then, the dis-
section should go deep to the wall of the pharynx,
which is opened on stay sutures posterior to the
thyroid cartilage. A healthy mucous membrane
should be available for the pharyngo-colic anas-
tomosis. The colonic graft should be long enough
to reach the pharynx. A wide, single-layer, end-­
to-­side anastomosis is made without any tension.
A nasogastric tube is left as a splint for 7–10 days.
784
Fig. 59.8  A passage is created behind the stomach,
through which the colon is taken up in an isoperistaltic
manner. Reproduced with permission from Springer,
Gupta DK, Sharma S.  Colonic interposition for
Esophageal replacement. 2nd Edition Pediatric Surgery
(Springer Surgery Atlas Series) Prem Puri & Hollwarth
M. Germany Springer 2019. Chapter 9: 59–65 https://doi.
org/10.1007/978-3-662-56282-6_9
The neck wound is closed with a soft drain
around the anastomosis. The native diseased and
strictured esophagus in patients with corrosive
injuries should be removed. The esophageal
mucosa at least should be removed in cases of
corrosive injuries to avoid any risk of malig-
nancy. It is challenging and unsafe to utilize the
S. Sharma and D. K. Gupta
Fig. 59.9  Anastomosis in the neck. Reproduced with
permission from Springer, Gupta DK, Sharma S. Colonic
interposition for Esophageal replacement. 2nd Edition
Pediatric Surgery (Springer Surgery Atlas Series) Prem
Puri & Hollwarth M. Germany Springer 2019. Chapter 9:
59–65 https://doi.org/10.1007/978-3-662-56282-6_9
transhiatal approach in children with past history
of major leaks after tracheoesophageal repair
with a scarred mediastinum (Gupta and Sharma
2019).
The ends of the divided colon are then anasto-
mosed in two layers, in front and inferior to the
pedicle of the isolated colon graft. The distal end
of the graft is anastomosed to the posterior (rarely
anterior) wall of the stomach in two layers in the
mid body (Fig. 59.10). The colon should be fixed
to the edges of the diaphragm at the lower end of
the tunnel in cases of retrosternal colon and to the
edge of the hiatus if the posterior mediastinal
route is used for the colonic replacement. The
liver is hitched behind the xiphisternum. The
pleura, if opened accidentally, is closed after
inflating the lung.
The colon should be decompressed to avoid
colonic dilatation and also to drain the swallowed
saliva. Most series have used the retrosternal
approach with a posterior cologastric anastomosis
to prevent the gastrocolic reflux. A retrosternal route
should, however, be avoided if the liver is large.
Pyloromyotomy or pyloroplasty as a drainage
procedure is usually performed if the vagus
nerves have been damaged while removing the
esophagus (Sharma and Gupta 2020). It is per-
formed as a Heineke-Mikulicz type with a
single-­layer anastomosis. Finally, the colocolic
anastomosis is performed to establish bowel
59  Esophageal Replacement 785

In cases with proximal anastomotic strictures,

dilatation may be done; if that is unsuccessful,
surgical revision of the colo-esophageal anasto-
mosis is planned. Postoperative redundancy and
graft dilatation leading to stasis of the food and
vomiting, if any, is corrected surgically.

59.6.4 Complications of Colon

Interposition

The complications of colon interposition are

given in Table 59.2.

Table 59.2  Complications of colon interposition (Sharma

and Gupta 2017)
Immediate
   Graft necrosis
   Anastomotic leak [20–50%]
   Mediastinitis, mediastinal abscess
Fig. 59.10  In the abdomen, the gastrocolic anastomo-   Respiratory complications—atelectasis,
sis is performed in two layers, usually near the cardia or pneumothorax, pneumonia, empyema
to the anterior wall of the stomach proximal to the    Mortality 1–10%
antrum. Reproduced with permission from Springer, Early
Gupta DK, Sharma S.  Colonic interposition for    Anastomotic leak
Esophageal replacement. 2nd Edition Pediatric Surgery    Gastric outlet obstruction
(Springer Surgery Atlas Series) Prem Puri & Hollwarth    Colonic anastomotic leak, peritonitis
M.  Germany Springer 2019. Chapter 9: 59–65    Intussusception
https://doi.org/10.1007/978-3-662-56282-6_9    Intestinal obstruction due to adhesions
   Sepsis
continuity. Care should also be taken to close the Late
window in the mesentery after the colonic resec-    Anastomotic stricture 20–30% cases
tion. Gastrostomy or a feeding jejunostomy is   Obstructive symptoms due to kinking of lower end
of colon
optional and is considered only if a prolonged
  Redundancy of colon or upper esophageal
ileus is anticipated or the anastomosis has been pouching 5–10%
precarious. The abdomen is closed in layers with    Colonic segment dilatation
a mediastinal drain.    Halitosis with fetid smell
  Acid reflux. Cologastric reflux and peptic ulcer
disease
   Graft-related complications—colitis
59.6.3 Postoperative Care
  Iron-deficiency anemia and transient
malabsorption
The drains are removed after 4–5 days, once the   Pulmonary symptoms due to chronic acid reflux
output has stopped. Patients can be kept on IV and aspiration
fluids or total parenteral nutrition or can be fed by    Intestinal obstruction due to adhesions
the gastrostomy or jejunostomy for 7–10 days. A    Dumping syndrome
   Late graft stenosis
contrast study is performed, and if there is no
  Respiratory problems due to pressure of
leakage, feeding is started. Any leaking usually esophagus
stops within a week or so. A gastrostomy, if per-    Delayed fistulization
formed, is removed about 3 months after surgery.    Mortality—due to respiratory failure, sepsis
786
59.6.5 Treatment of Complications
Most proximal anastomosis strictures respond to
dilatation. However, 30–50% may require surgi-
cal revision, especially those following corrosive
injuries. Lower-end strictures, although less com-
mon, usually require surgical revision.
Mild redundancy of the colon may not neces-
sitate revision of the colonic conduit if there is no
dysphagia. Redundancy may lead to pulmonary
compression, food stasis, recurrent regurgitation,
and aspiration (Fig. 59.11). Surgical revision may
be required if the patient is symptomatic with
vomiting has poor weight gain and dysphagia, to
improve quality of life and to prevent complica-
tions, such as aspiration. Successful transhiatal
mobilization of the intrathoracic colon and sta-
pled tapering coloplasty has been described for
redundancy.
Cologastric reflux and peptic ulcer disease are
uncommon. A cologastric anastomosis on the
Fig. 59.11  Redundant colon was tapered surgically by
an abdominal incision extending it to thorax
S. Sharma and D. K. Gupta
posterior wall of the stomach may prevent the
reflux. The formation of an antireflux submuco-
sal gastric tunnel can eliminate gastrocolic reflux
without impeding normal passage of food into
the stomach. Redo surgery may be needed for
persistent fistula.
59.6.6 Long-Term Outcome
and Follow-Up
Colonic interposition is the safest procedure
among the esophageal replacements, with mini-
mal short-term and long-term complications
(Gupta and Shilpa 2008).
Most patients catch up in terms of growth.
Digestive symptoms with feeding difficulties are
common in the initial 5  years of follow-up.
Gastroesophageal reflux may be present in up to
60% cases. Nutritional complications may be
seen in up to 25% (Coopman et al. 2008). Lung
function may be abnormal in 50–60%, while sco-
liosis may be present in 35%. An endoscopy is
important to look for any changes in the colonic
mucosa in both upper and lower junctions
(Elshafei et al. 2012). A follow-up contrast study
may delineate the neoesophagus and should be
done in the presence of any symptoms
(Fig. 59.12). The pathologic changes in the tho-
racic colon may develop mild chronic nonspe-
cific inflammation of the colonic mucosa in about
one-quarter of cases (Elshafei et al. 2012).
Corrosive strictures are common in many
countries such as Egypt, Turkey, and Romania.
The largest experience in colonic interposition
of more than 850 cases has been reported by
Hamza et al. (Hamza et al. 2003). The reported
complications in their last 475 cases included
10% cervical leakage, 5% proximal strictures,
2% postoperative intestinal obstruction, 1%
mortality, and 0.6% late graft stenosis (Hamza
et al. 2003). Long-term follow-up has substanti-
ated the colonic interposition as a durable and
highly acceptable method for esophageal
replacement. Malignancy has not been reported
in children at the site of neo-anastomosis follow-
ing colon interposition.
59  Esophageal Replacement 787

a b

Fig. 59.12  Barium study showing replaced esophagus with colon. (a) Good flow. (b) Tortuous course with aspiration
suggestive of stricture at the upper end
788
59.7 Gastric Transposition
As the colon interposition was associated with
several long-term unacceptable complications,
alternate procedures were sought. Gastric trans-
position was first done in children about 40 years
ago, though the procedure had quite often been
performed in adults with esophageal carcinoma.
The gastric transposition for replacing the esoph-
agus in children was popularized by Prof. Lewis
Spitz (UK, 1981), after noting the unacceptably
high complication rate with the use of colon
interposition (Spitz et al. 2004). The authors have
been using this technique for the past 30 years in
India (Gupta and Sharma 2008). After the initial
experience in infants and children, its use was
even extended to newborns. Recently, this proce-
dure has gained popularity to replace the esopha-
gus in children. Many large series have reported
satisfactory early and long-term results in new-
borns, infants, and children, without much risk of
respiratory compromise due to the presence of
the bulky stomach in the chest (Ludman and
Spitz 2003; Spitz and Coran 2019).
The colon and jejunum have a blood supply
that is very delicate in the newborn period and,
thus, have not been considered very safe for
replacing the esophagus by using colon or the
jejunum. The stomach is comparatively small in
size to fashion a gastric tube in newborns. Thus,
the stomach, with its reliable blood supply, is
considered the best option available in newborns
requiring esophageal replacement. Also, the size
of the stomach in newborns and infants less than
6 months of age is quite small and unfit for pre-
paring the gastric tube. Moreover, the gastric wall
is also quite vascular and muscular to withstand
mediastinal infection if any. Thus, gastric trans-
position is the only option to replace the esopha-
gus in the neonatal period following postoperative
major leaks and also with the long or wide gap
esophageal atresia and tracheoesophageal fistula
to avoid diversion and offer definitive procedure
even in the newborn stage to avoid repeated visits
to the hospital (Sharma and Gupta 2017; Soccorso
and Parikh 2016).
S. Sharma and D. K. Gupta
59.7.1 Advantages of Gastric
Transposition
1. Technically easier operation.
2. Stomach has a thick muscular wall with reli-
able vascular supply.
3. Adequate length is possible for anastomosis
without tension. If needed, duodenum can be
kocherized.
4. A single anastomosis is required in the neck.
5. There is no suture line in the chest, except the
closure sites of the gastrostomy and the
esophageal hiatus.
6. Least short-term morbidity and mortality.
A drainage procedure, preferably pyloromy-
otomy (or pyloroplasty), is to be added as bilat-
eral vagotomy is invariably done while mobilizing
the stomach for pull-up.
Precautions with the use of the stomach include
taking care of the bulk of stomach in the medias-
tinum/thorax, which can cause respiratory dis-
tress. Adequate space needs to be created to house
the stomach at the thoracic inlet and prevent the
decrease in venous return from head and neck
region. Aspiration pneumonitis can occur due to
reflux. Gastric emptying may be delayed despite
pyloromyotomy. There may also be rapid gastric
emptying leading to dumping. The loss of gastric
reservoir function affects the growth and develop-
ment of the patient. Bile gastritis may result in a
change in gastric mucosa in the long term.
59.7.2 Operative Procedure
Bowel preparation is recommended, to ensure an
empty colon in the event that the stomach is
found bulky and not suitable for the transposition
procedure.
A midline supraumbilical incision or a left
subcostal incision is made (Fig. 59.13). The gas-
troesophageal junction is mobilized. The esopha-
geal stump in pure esophageal atresia is delivered
in the abdomen. Adequate dissection is done in
the neck at the thoracic inlet.
59  Esophageal Replacement
Fig. 59.13  A midline upper abdominal incision or a left
subcostal incision is made. Spitz L, Coran A.  Gastric
Transposition for Oesophageal Replacement 2nd Edition
Pediatric Surgery (Springer Surgery Atlas Series) Prem
Puri & Hollwarth M.  Germany Springer 2019. Chapter
10: 67–71 https://doi.org/10.1007/978-3-662-56282-6_9
The stomach is exposed, gastrostomy, if
present, is taken down, and the defect in the
stomach is closed in two layers (Fig.  59.14).
The greater and lesser curvatures of the stomach
are mobilized, preserving the integrity of the
right gastroepiploic and right gastric arcades.
The mobilization of the stomach continues
­proximally by dividing the short gastric vessels
between the fundus of the stomach and the
spleen, using diathermy or preferably by liga-
tures, to avoid any vascular spasm. The left gas-
tric artery and vein are ligated securely close to
789
Fig. 59.14  The stomach is exposed, the gastrostomy
site is taken down, and the defect in the stomach is
closed. Spitz L, Coran A.  Gastric Transposition for
Oesophageal Replacement 2nd Edition Pediatric Surgery
(Springer Surgery Atlas Series) Prem Puri & Hollwarth
M.  Germany Springer 2019. Chapter 10: 67–71
https://doi.org/10.1007/978-3-662-56282-6_9
their origin. Kocherization of the second part of
the duodenum is performed to gain length and
bring the duodenum and the pylorus in a straight
line. The stomach maintains good blood supply
on intact right gastric artery and vein and the
right gastroepiploic vessels. After mobilization,
the proximal 1/3 of the stomach may appear
atonic and even dusky, but it never becomes isch-
emic due to the presence of vascular anastomosis
developing at subserosal level.
The stump of the distal esophagus (in the case
of long-gap atresia) is mobilized from the poste-
rior mediastinum by dividing the phrenoesopha-
geal membrane and dissecting out the esophagus
(Fig.  59.15). The anterior and posterior vagal
nerves are divided. The esophageal stump is
­transected from its junction with the stomach and
the defect is closed in two layers.
The site of the gastrostomy is also closed in
two layers. The fundus is the highest point. Two
stay sutures are applied to the fundus of the stom-
790
Fig. 59.15  The stump of the distal esophagus is mobilized
from the posterior mediastinum. Spitz L, Coran A. Gastric
Transposition for Oesophageal Replacement 2nd Edition
Pediatric Surgery (Springer Surgery Atlas Series) Prem
Puri & Hollwarth M. Germany Springer 2019. Chapter 10:
67–71 https://doi.org/10.1007/978-3-662-56282-6_9
ach with a right and left side marked, for taking it
up in the neck for the anastomosis, taking due
care to avoid any rotation.
The sutured gastrostomy site and the closed-­
off gastroesophageal junction are shown. Two
sutures of different materials are placed in the
fundus. The orientation of these sutures is used to
ensure that rotation of the stomach does not occur
while it is being pulled up into the neck.
Alternatively, a 32-Fr stiff chest tube can be sewn
to the fundus with two sutures. It is used to iden-
tify the orientation of the stomach and to draw
the stomach fundus up into the neck.
Via the cervical incision, the full thickness of
the esophagus is mobilized.
A pyloroplasty or pyloromyotomy is per-
formed (Fig.  59.16). The recurrent laryngeal
nerves must be preserved during the mobilization
procedure (Fig. 59.17). The length of the stom-
ach is assessed (Fig. 59.18).
The appropriate route is now chosen. No tho-
racotomy is performed. The mediastinal route is
quite feasible and rather preferred in all neonates.
If the creation of the tunnel is found difficult due
S. Sharma and D. K. Gupta
Fig. 59.16  A pyloroplasty or pyloromyotomy is per-
formed. Spitz L, Coran A.  Gastric Transposition for
Oesophageal Replacement 2nd Edition Pediatric Surgery
(Springer Surgery Atlas Series) Prem Puri & Hollwarth
M. Germany Springer 2019. Chapter 10: 67–71 https://doi.
org/10.1007/978-3-662-56282-6_9
to previous thoracotomy adhesions, the attempt is
abandoned and the substernal route is favored. In
that case, the hiatus will be closed with a couple
of figure-8 sutures. The plane of dissection for
the mediastinal tunnel is directly anterior to the
prevertebral fascia. From above, the dissection
proceeds immediately posterior to the trachea
and caudally into the posterior mediastinum.
From below, through a widened hiatus, dissection
is carried out under vision in the prevertebral
space behind the heart. The tunnel is completed
from above and below by gentle digital dissec-
tion in the posterior mediastinum (Fig. 59.19). If
any problem is encountered in creating the poste-
rior mediastinum tunnel by blunt finger dissec-
tion, it is advisable to perform a lateral transpleural
thoracotomy and complete the dissection under
direct view. This approach is also essential to
remove a post corrosive densely adherent scarred
esophagus or a tumor of the esophagus.
The substernal route can be created easily by a
blunt finger dissection starting just behind the
xiphisternum, the stomach is taken up to the
neck. The anastomosis between the end of the
59  Esophageal Replacement 791

Fig. 59.17  The full thickness esophagus is mobilized

preserving the recurrent laryngeal nerves. Spitz L,
Coran A.  Gastric Transposition for Oesophageal
Replacement 2nd Edition Pediatric Surgery (Springer
Fig. 59.18  The stomach is pulled up through the hiatus in
Surgery Atlas Series) Prem Puri & Hollwarth
the diaphragm and the posterior mediastinal tunnel until the
M. Germany Springer 2019. Chapter 10: 67–71 https://
fundus appears at the cervical incision. Spitz L, Coran
doi.org/10.1007/978-3-662-56282-6_9
A. Gastric Transposition for Oesophageal Replacement 2nd
Edition Pediatric Surgery (Springer Surgery Atlas Series)
cervical esophagus and the top of the fundus of Prem Puri & Hollwarth M. Germany Springer 2019. Chapter
the stomach is made using a single layer of 5/0 or 10: 67–71 https://doi.org/10.1007/978-3-662-56282-6_9
6/0 sutures, taking the full thickness of the walls
of the esophagus and the stomach (Fig.  59.20). wound. The stomach is decompressed with a
Before completing the anterior wall of the anas- nasogastric tube, the tip of which is carefully
tomosis, a size 10F–12F nasogastric tube is placed in the mid-thoracic region and is left open
passed, with the tip in the intrathoracic part of the to drain the contents freely. A feeding jejunos-
stomach. The wounds are closed, with a soft rub- tomy proves useful for nutritional support in the
ber drain at the cervical incision. Pyloromyotomy postoperative period.
or pyloroplasty is always performed to provide
drainage procedure as bilateral vagotomy has
invariably been done. The pyloric antrum is fixed 59.7.3 Complications
to the diaphragmatic edges at the hiatus. The
stomach is also hitched to the prevertebral fascia In more than 200 cases by Lewis Spitz and over
in the neck to prevent any tension on the anasto- 190 cases by Arnold Coran, the mortality of this
mosis. A soft glove drain is used to drain the neck
792
Fig. 59.19  The plane of dissection for the mediastinal
tunnel is directly anterior to the prevertebral fascia.
Spitz L, Coran A. Gastric Transposition for Oesophageal
Replacement 2nd Edition Pediatric Surgery (Springer
Fig. 59.20  The anastomosis between the end of the cer-
vical esophagus and the top of the fundus of the stomach
is fashioned using a single layer. Spitz L, Coran A. Gastric
Transposition for Oesophageal Replacement 2nd Edition
procedure was about 5% (Spitz 1992). The morbid-
ity has been significant, including these problems:
• Anastomotic leak in the neck (12–20%)
• Anastomotic stricture (20–30%)
• Swallowing problems (30%)
• Delayed gastric emptying (5–9%)
• Complications with the jejunal feeding tube
(4%)
• Dumping syndrome (1–3%)
Most of the children prefer to take small, fre-
quent meals, although a normal eating pattern is
generally established in older children. Many of
S. Sharma and D. K. Gupta
Surgery Atlas Series) Prem Puri & Hollwarth
M. Germany Springer 2019. Chapter 10: 67–71 https://
doi.org/10.1007/978-3-662-56282-6_9
Pediatric Surgery (Springer Surgery Atlas Series) Prem
Puri & Hollwarth M.  Germany Springer 2019. Chapter
10: 67–71 https://doi.org/10.1007/978-3-662-56282-6_9
the patients (particularly children who are born
with esophageal atresia) grow at a slower rate
than normal and are in the lower half of the
growth charts for both weight and height.
Anastomotic leaks in the neck are not uncom-
mon but most of them are minor and resolve
without any treatment within a few days. Major
leaks result in stricture formation and thus need
regular follow-up, including the endoscopy at
regular intervals. Most of these respond to dila-
tation. Few may, however, need a surgical cor-
rection. Anastomotic strictures are more
common in patients with caustic esophageal
injury. Gastric transposition has relatively few
59  Esophageal Replacement 793

complications compared to other procedures

(Table 59.3, Fig. 59.21).
Gastric transposition for esophageal substitu-
tion has been associated with about 5% mortal-
ity, even in the best hands. Most deaths occur in
the early postoperative period due to ongoing
sepsis and associated congenital heart disease

Table 59.3  Complications following gastric transposition

Immediate
   Leak
   Chest infection
   Lung collapse
   Acute gastric dilatation
   Cardiac compression
   Bleeding
   Pneumothorax
   Aspiration pneumonitis
   Recurrent laryngeal nerve palsy
   Sepsis
   Mortality
Early
   Anastomotic leak in the neck [15–35%]
   Neck or abdominal wound infection
   Recurrent aspiration
   Delay in feeding
   Jejunostomy complications, leak, perforation
   Pulmonary compression
   Dumping
   Vomiting
Fig. 59.21  Barium studies in a patient with gastric trans-
   Adhesive intestinal obstruction
position, depicting an anastomotic stricture in the neck
   Gastroesophageal reflux
   Duodenogastric reflux
   Bronchogastric fistula besides aspiration pneumonitis (Spitz 1992).
   Mortality The most common indication for redo surgery is
Late following the postoperative stricture in the neck,
   Stricture in the neck mechanical obstruction of the stomach at the
   Intestinal obstruction due to adhesions diaphragm level, feeding jejunostomy related
Dyspnea
complications, and the intestinal obstruction due
   Uncoordinated swallowing
   Reflux gastritis
to adhesions.
   Graft loss—rare Symptoms mimicking obstruction have been
   Dumping syndrome reported due to stasis of food in the thoracic
   Functional obstruction due to vagotomy stomach (functional obstruction due to vagot-
   Anemia omy) without any evidence of mechanical
   Loss of reservoir function and poor growth obstruction. This needs careful assessment for
   Pulmonary function compromise
the need of jejunostomy to provide nutrition. It
   Problems during swallowing
may also lead to aspiration pneumonia during
   Malabsorption
   Para-neoesophageal hernia sleep.
  Mediastinal abscess secondary to esophageal
perforation as a result of dilatation procedure for
stricture
794
59.7.4 Long-Term Outcome
and Follow-Up
In larger series, more than 90% had a good to
excellent outcome in terms of the absence of
swallowing difficulties or other gastrointestinal
symptoms (Spitz et  al. 2004). The authors also
have the similar experience with a 10-year or
more follow-up of patients who had undergone
gastric transposition. The majority of the patients
learned the habit of eating and swallowing in due
course of time (Jain et al. 2012). There has been
no deterioration in the function of the stomach in
those patients followed up for more than 10 years.
There is no peristalsis in the thoracic stomach.
The transposed stomach responds well to the
food bolus by gravity. The manometric studies
performed after gastric pull-up documented mass
contractions without any propulsive or peristaltic
waves (Gupta et al. 2004). Gastric emptying was
found delayed in more than half and hurried in
one-third of patients (Ravelli et al. 1994; Angotti
et al. 2017). The quality of life is good and gastric
transposition is compatible with normal life
(Ludman and Spitz 2003). The growth and devel-
opment of these patients has remained subnormal
followed up for 5–10  years in some series.
Anemia has been found in 70% of the cases dur-
ing long-term follow-up (Jain et al. 2012).
The authors have noted the decreased func-
tional capacity of the lungs, especially in those
patients who had undergone previous thoracoto-
mies (Sharma and Gupta 2019). This may be sec-
ondary to the restrictive lung disease, pulmonary
compression, or recurrent aspirations.
Surgical experience has been a great factor in
lowering the complication rates, with mortality
lowering to 2.5% in 236 gastric transpositions
reported by Spitz in 2014 with a satisfactory out-
come in over 90% of patients (Spitz 2014).
However, the leak rate of 12% and stricture rate
of 20% has remained the same over time.
59.8 Gastric Tube
The use of a gastric tube for esophageal replace-
ment was described in 1968 by Burrington and
Stephens (Burrington and Stephens 1968). A gas-
S. Sharma and D. K. Gupta
tric tube matches with the size of the esophageal
lumen. However, it is technically more demand-
ing and associated with serious immediate and
early postoperative complications. The gastric
tube may be formed as an isoperistaltic antral
based tube or as an antiperistaltic fundus-based
tube. As the fundus has more acid cells while the
antrum has gastrin-secreting cells, most surgeons
routinely use the reversed gastric tube as it would
have less acid in the neck and thus less chance of
a leak and stricture formation. In the case of the
isoperistaltic tube, the food is deposited directly
near the pylorus bypassing the body of the stom-
ach resulting in dumping-like symptoms (Ionescu
2008).
59.8.1 Operative Procedure
The site of the gastric tube to be formed is
marked with stay sutures in the stomach. The
proximal end of the antiperistaltic gastric tube
starts 2–5 cm proximal to pylorus, on the antral
segment.
To make a gastric tube in children, a red rub-
ber catheter of approximately #18–22 Fr is used.
A 2  cm incision is made on the antrum, not to
create narrowing of the passage; the tube is
inserted retrograde within the stomach and placed
close to the great curvature. The gastric tube is
then separated from the great curvature. Whether
a reversed or isoperistaltic tube is to be fashioned
is decided and the gastroepiploic vessels are
ligated and divided at the right or the left end of
the gastroepiploic arcade accordingly. The proxi-
mal 3–5 cm of the tube should be approximated
with interrupted sutures in double layer and
should be larger than the rest, and a little funnel is
shaped for a good cervical anastomosis. If sta-
plers are used to create the tube, the edges should
be oversewn with a second interrupted suture
layer manually, so as to make it water proof to
avoid blood loss and also risk of leak from the
suture line. Only the short gastric vessels are
divided. The gastrostomy tube is reposited, most
commonly through the same site or a new one.
Often, the gastrostomy could be comfortably
included in the long suture line. A feeding jeju-
nostomy may be added for early feeding.
59  Esophageal Replacement
After the gastric tube is fashioned, the route
for the replacement, transhiatal, or retrosternal is
developed. The long suture line should be kept in
the anterior position. About 2–5 cm of the length
of the neotube should be left in the abdomen to
prevent gastroneoesophageal reflux. A drain may
be left in the posterior mediastinum. The cervical
anastomosis is the most important part of the
operation. The proximal tube is trimmed for
1–2 cm and examined for a good vascular supply.
The cervical anastomoses are made end to end, in
one layer, with interrupted 5/0 absorbable suture.
The operation ends by repositioning the gastros-
tomy and closure of the abdomen.
59.8.2 Complications of Gastric Tube
The complications of gastric tube are outlined in
Table 59.4.
A leak in the neck is the most common com-
plication, mostly closing spontaneously in due
course of time. The second most common com-
plication is a stricture that mostly resolves with
dilatations. The severe ones may require resec-
tion and redo anastomosis.
59.8.3 Long-Term Outcome
and Follow-Up
The long-term results following gastric tube are
quite satisfactory (Ein 1998). However, compli-
cations may arise later. Lee et al. found that gas-
Table 59.4  Complications of gastric tube
Immediate
   Anastomotic leak (50–60%)
   Mediastinitis due to leak
   Graft necrosis
Early
   Anastomotic leak in the neck
  Respiratory symptoms usually due to aspiration
secondary to a stricture in the neck
Late
   Stricture of the tube, in the neck (20–30%)
   Gastric tube ulceration and hemorrhage
795
tric tube replacement had more long-term
complications (86%) compared to delayed pri-
mary anastomosis for esophageal atresia (30%)
(Lee et al. 2014). Acid secretion from the gastric
tube may result in Barrett’s esophagus in the neck
above the anastomosis indicating the need for
lifelong endoscopic follow-up for these patients
(Borgnon et al. 2004). The quality of life and the
general life status of the patients who have under-
gone a gastric tube esophagoplasty or the colon
tube esophagoplasty has been found to be the
same (Gavrilescu et al. 2013).
59.9 Jejunal Graft
A jejunal graft for esophageal replacement was
described in 1947 (Azizkhan 2014). The proce-
dure is technically demanding for esophageal
replacement and may take up to 24 h. The experi-
ence is limited with a risk of life-threatening vas-
cular compromise. Thus, the procedure has not
become very popular. Moreover, it involves dis-
carding some length of jejunum to make a suffi-
cient length of vascular pedicle due to the arcades.
The jejunum also has a short mesentery and may
involve the need of microvascular anastomosis of
the jejunal vessels with the neck vessels, with a
chance of failure, graft loss, and leaks in the neck.
Some experts believe that the disadvantage of
longer operation time is easily compensated by
less chances of postoperative dilatation of the
graft and better quality of life of these patients.
The function of the jejunal segment is considered
much better than the other substitutes of the
esophagus thus making the jejunum the most
appropriate substitute of esophagus in infants and
children (Hashizume and Dessanti 2008).
The most disastrous complication is total
necrosis of the interposed jejunum due to vascular
insufficiency, i.e., arterial thrombosis or venous
occlusion. Hence, it is not advisable to perform
jejunum interposition in neonates with friable and
delicate mesenteric arteries and veins. The proce-
dure is considered safe and feasible at the age of
6 months with a minimal weight of 6 kg.
796
59.9.1 Operative Procedure
The first jejunal artery with accompanying vein is
preserved. The second and third jejunal arteries
are usually divided and ligated near their origins,
and the fourth artery is preserved and is used as
the part of vascular pedicle of the jejunal loop.
With this, one can get adequate length of the vas-
culature pedicle. The distal end of divided jeju-
num is then placed on the anterior chest wall to
ensure that this part can be brought to the level of
the place designated for the upper anastomosis
(Sharma and Gupta 2019). The arcade is then
ligated and divided just distal to the point of the
confluence with the fourth jejunal artery. A lon-
ger length of the jejunal loop is usually redundant
so the part is resected. The continuity of the jeju-
num is restored. The jejunal loop so prepared is
brought to the upper abdomen through a small
hole in the mesocolon. Regarding the route to the
thoracic cavity, if the length of the lower
­esophagus is long enough to permit the lower
anastomosis to be done in the right thoracic cav-
ity, the loop should be brought anterior to the
stomach and through a hole in the diaphragm.
However, if the length of the lower esophagus is
not adequate enough and the anastomosis needs
to be done in the abdomen, the jejunal loop
should be brought to the right thorax through a
retrogastric route and through the esophageal
hiatus. Maximum precision is necessary to avoid
a kinking and torsion of the vascular pedicle.
The end-to-side anastomosis of the upper
esophagus and the jejunal loop is preferred to
avoid undue tension to the vascular pedicle. The
anastomosis of the interposed jejunum and the
lower esophagus is accomplished by end-to-end
fashion. A thoracic drain is inserted to the right
thorax and a Penrose or suction drain is placed
near the esophageal hiatus. A feeding jejunostomy
distal to the anastomosis may be considered.
59.9.2 Complications
If there is a leak, parenteral nutrition or enteral
feeding through a gastrostomy should be initi-
ated. Usually, the leak heals spontaneously after
S. Sharma and D. K. Gupta
Table 59.5  Complications of jejunal graft
Immediate
   Graft necrosis
   Perforation of the graft
   Anastomotic leak
Early
   Anastomotic leak
   Cervical fistulas
   Necrosis of the tip of graft
Late
   Cervical stricture
   Peptic strictures
   Cervical dysphagia
   Esophageal diverticulum
   Redundancy of the jejunum
   Delayed swallowing
2 weeks. The complications of a jejunal graft are
outlined in Table 59.5.
59.9.3 Long-Term Outcome
and Follow-Up
Although there is little experience with the use of
jejunum, the results are encouraging. A meta-­
analysis of adult experience with jejunal grafts
reported up to 10% mortality, 36% anastomotic
leak rate, and 5–11% graft loss (Gaur and
Blackmon 2014). However, in another study with
32 cases, the results were found to be excellent,
with no failures of the jejunum to reach the neck,
no loss of graft, and no deaths (Ring et al. 1982).
All patients could eat a regular diet at normal
speeds. Transient stagnation of swallowed barium
at the lower esophagus has been observed in 7/12
cases in another series (Saeki et al. 1988). In com-
parison with using colon, the long-term results of
jejunal interposition seem to be extremely good,
possibly due to less chances of anastomotic stric-
ture formation due to good vasculature, better
peristalsis of jejunum, less chances of diverticu-
lum formation, less chances of adenoma forma-
tion, and the less tendency of dilatation even after
several years. In addition, after jejunal interposi-
tion, children have fewer problems, and they show
normal growth as the influence of resection of
redundant part of the jejunum on growth is nil or
minimum. Jejunal interposition remained an
59  Esophageal Replacement
option in a patient after the failed colon interposi-
tion (Ho et  al. 2014). Gallo et  al. compared the
long-term outcomes of the gastric pull-up and the
jejunal interposition for long gap esophageal atre-
sia and found functional obstruction in 46% after
jejunal interposition versus none after gastric
pull-up (p = 0.02) (Gallo et al. 2015).
59.10 Recent Advances
Colon patch esophagoplasty with preservation of
the injured esophagus is an alternative for long
caustic esophageal strictures.
A new procedure of isoperistaltic isolated gas-
tric tube interposition with favorable short- and
long-term functional results has been reported
(Gounot et al. 2006).
The pedicled ileum graft has also been suc-
cessfully used in children, including in the redo
cases with good results (Bax and Van Renterghem
2005; Ivanov et al. 2012).
Minimally invasive procedures have also been
performed for esophageal replacement. Esopha-
gectomy has been successfully performed thora-
coscopically. Hence, the risk for malignancy in
the redundant esophagus, especially following
corrosive injury, can be reduced. Laparoscopic
gastric transposition by minimally invasive tech-
niques avoids the trauma of open access (Ng
et  al. 2014). However, laparoscopy may not be
feasible in the presence of dense adhesions.
Robotic surgery, however, may help in future to
perform the in vivo anastomosis in the abdomen
for colonic interposition in suitable cases.
Regenerative medicine and tissue engineering
is the new ray of hope to create a biologically
functional esophagus. However, it is likely to take
at least two more decades to become a reality.
59.11 Conclusion
Patient’s own esophagus is still considered the
best; thus, all attempts should be made to save,
preserve, and rescue it. However, if the quality of
life is poor due to an esophageal stricture or other
associated diseases, esophageal replacement is
797
strongly indicated. Being technically demanding,
the procedure should be performed by experienced
surgeons and a surgical team and in an institutional
setup so that if the operative and postoperative
complications develop, the infrastructure facilities
can be utilized immediately to diagnose and man-
age them at the earliest time. Outcome following
esophageal replacement has significantly improved
over the years. One should prefer to learn a par-
ticular technique, master it, and record the long-
term results instead of trying all four techniques
and becoming a master of none.
The colon is considered the most preferred
and safest organ for replacement. The stomach is
a vascular and muscular organ with lower risk of
ischemia. A gastric tube is a demanding tech-
nique. The jejunum and ileum are good alterna-
tives for redo cases but expertise is needed. As
the mortality has reduced, more cases of compli-
cations are being tackled with redo surgeries.
Endoscopy has been added recently to the arma-
mentarium for follow-up.
References
Angotti R, Molinaro F, Noviello C et  al (2017) Gastric
transposition as a valid surgical option for esophageal
replacement in pediatric patients: experience from
three Italian medical centers. Gastroenterol Rep (Oxf)
5(1):47–51
Azizkhan R (2014) Chapter 25: Esophageal replacement.
In: Ziegler MM, Azizkhan RG, Allmen DV, Weber
TR (eds) Operative pediatric surgery. Macgraw-Hill,
China
Bax NM, Van Renterghem KM (2005) Ileal pedicle graft-
ing for esophageal replacement in children. Pediatr
Surg Int 21:369–372
Borgnon J, Tounian P, Auber F et  al (2004) Esophageal
replacement in children by an isoperistaltic gas-
tric tube: a 12-year experience. Pediatr Surg Int
20:829–833
Burrington JD, Stephens CA (1968) Esophageal replace-
ment with a gastric tube in infants and children. J
Pediatr Surg 3:24–52
Chen H, Lu JJ, Zhou J, Zhou X, Luo X, Liu Q, Tam J
(2009) Anterior versus posterior routes of reconstruc-
tion after esophagectomy: a comparative anatomic
study. Ann Thorac Surg 87:400–404
Coopman S, Michaud L, Halna-Tamine M, Bonnevalle
M, Bourgois B, Turck D, Gottrand F (2008) Long-
term outcome of colon interposition after esopha-
gectomy in children. J Pediatr Gastroenterol Nutr
47:458–462
798
Ein SH (1998) Gastric tubes in children with caustic
esophageal injury: a 32-year review. J Pediatr Surg
33:1363–1365
Elshafei H, Elshafei E, ElDebeiky M, Hegazy N, Zaki
A, Abdel Hay S (2012) Colonic conduit for esopha-
geal replacement: long-term endoscopic and histo-
pathologic changes in colonic mucosa. J Pediatr Surg
47:1658–1661
Gallo G, Zwaveling S, Van der Zee DC, Bax KN, de
Langen ZJ, Hulscher JB (2015) A two-center com-
parative study of gastric pull-up and jejunal interpo-
sition for long gap esophageal atresia. J Pediatr Surg
50:535–539
Gaur P, Blackmon SH (2014) Jejunal graft conduits after
esophagectomy. J Thorac Dis 6(Suppl 3):S333–S340
Gavrilescu S, Hanganu E, Sarbu I, Aprodu SG (2013)
Quality of life of patients with esophageal replace-
ment for congenital and acquired esophageal anoma-
lies. Rev Med Chir Soc Med Nat Iasi 117:334–336
Gounot E, Borgnon J, Huet F, Sapin E (2006) Isolated
isoperistaltic gastric tube interposition for esoph-
ageal replacement in children. J Pediatr Surg
2006(41):592–595
Gupta DK, Sharma S (2008) Chapter 34: Gastric transpo-
sition. In: Gupta DK (ed) Pediatric surgery diagnosis
and management. Jaypee, pp 409–421
Gupta DK, Sharma S (2011) Primary gastric pull up in
pure esophageal atresia: technique, feasibility and
outcome. A prospective observational study. Pediatric
Surg Int 27:583–585
Gupta DK, Sharma S (2019) Chapter 9: Colonic interposi-
tion for esophageal replacement. In: Puri P, Hollwarth
M (eds) Pediatric surgery, Springer Surgery Atlas
Series, 2nd edn. Springer, Germany, pp 59–65. https://
doi.org/10.1007/978-­3-­662-­56282-­6_9
Gupta DK, Shilpa S (2008) Chapter 32: Esophageal
replacement with colon. In: Gupta DK (ed) Pediatric
surgery diagnosis and management. Jaypee,
pp 382–390
Gupta DK, Charles AR, Srinivas M (2004) Manometric
evaluation of the intrathoracic stomach after gastric
transposition in children. Pediatr Surg Int 20:415–418
Gupta DK, Sharma S, Arora MK, Agarwal G, Gupta M,
Grover VP (2007) Esophageal replacement in the neo-
natal period in infants with esophageal atresia and tra-
cheoesophageal fistula. J Pediatr Surg 42:1471–1477
Hamza AF, Abdelhay S, Sherif H et  al (2003) Caustic
esophageal strictures in children: 30 years' experience.
J Pediatr Surg 38:828–833
Hashizume K, Dessanti A (2008) Chapter 35: Jejunal
interposition as a substitute of esophagus. In: Gupta
DK (ed) Pediatric surgery diagnosis and management.
Jaypee, pp 422–425
Ho AC, Yeo MS, Ciudad P, Chen HC (2014) 2-stage free
and pedicle jejunum for esophageal replacement after
failed colon interposition for caustic injury in a 5-year-­
old child. J Plast Reconstr Aesthet Surg 67:417–419
Ionescu GO (2008) Chapter 33: Gastric tube esophageal
replacement. In: Gupta DK (ed) Pediatric surgery
diagnosis and management. Jaypee, pp 391–408
S. Sharma and D. K. Gupta
Ivanov AP, Nabokov VV, Miroshnikov BI, Galkina NV
(2012) Repeated plasty of the esophagus in children
with the application of transplant from the ileum.
Vestn Khir Im I I Grek 171:54–60
Jain V, Sharma S, Kumar R, Kabra SK, Bhatia V, Gupta
DK (2012) Transposed intrathoracic stomach: func-
tional evaluation. Afr J Paediatr Surg 9:210–216
Lee HQ, Hawley A, Doak J, Nightingale MG, Hutson
JM (2014) Long-gap oesophageal atresia: compari-
son of delayed primary anastomosis and oesopha-
geal replacement with gastric tube. J Pediatr Surg
49:1762–1766
Ludman L, Spitz L (2003) Quality of life after gastric
transposition for oesophageal atresia. J Pediatr Surg
38:53–57. discussion 53–7
Ng J, Loukogeorgakis SP, Pierro A, Kiely EM, De Coppi
P, Cross K, Curry J (2014) Comparison of minimally
invasive and open gastric transposition in children. J
Laparoendosc Adv Surg Tech A 24:742–749
Ravelli AM, Spitz L, Milla PJ (1994) Gastric empty-
ing in children with gastric transposition. J Pediatr
Gastroenterol Nutr 19:403–409
Ring WS, Varco RL, L'Heureux PR, Foker JE (1982)
Esophageal replacement with jejunum in children:
an 18 to 33 year follow-up. J Thorac Cardiovasc Surg
83:918–927
Saeki M, Tsuchida Y, Ogata T, Nakano M, Akiyama H
(1988) Long-term results of jejunal replacement of the
esophagus. J Pediatr Surg 23:483–489
Sandblom P (1948) The treatment of congenital atresia of
the esophagus from a techncal popint of view. Acta
Chir Scand 97:25–34
Sharma S, Gupta DK (2017) Surgical techniques for
esophageal replacement in children. Pediatr Surg Int
33:527–550
Sharma S, Gupta DK (2019) Esophageal replacement. In:
Puri P (ed) Encyclopedia of pediatric surgery, vol 2.
Springer
Sharma S, Gupta DK (2020) Chapter 36: Colonic inter-
position for esophageal replacement in children. In:
Redkar R (ed) IAPS textbook of pediatric surgery.
Jaypee, New Delhi, pp 352–358
Sherman CD, Waterston DJ (1957) Esophageal recon-
struction in children using colon. Arch Dis Child
32:11. http://adc.bmj.com/
Soccorso G, Parikh DH (2016) Esophageal replacement
in children: challenges and long-term outcomes. J
Indian Assoc Pediatr Surg 21:98–105
Spitz L (1992) Gastric transposition for esophageal sub-
stitution in children. J Pediatr Surg 27:252–259
Spitz L (2014) Esophageal replacement: overcoming the
need. J Pediatr Surg 49:849–852
Spitz L, Coran A (2019) Chapter 10: Gastric transposition
for oesophageal replacement. In: Puri P, Hollwarth M
(eds) Pediatric surgery, Springer Surgery Atlas Series,
2nd edn. Springer, Germany, pp  67–71. https://doi.
org/10.1007/978-­3-­662-­56282-­6_9
Spitz L, Kiely E, Pierro A (2004) Gastric transposition
in children  – a 21-year experience. J Pediatr Surg
39:276–281. discussion 276–81
 Infantile Hypertrophic Pyloric
Stenosis
Takashi Doi and Takao Fujimoto
60.1 Introduction Including
Definition and Incidence
Infantile hypertrophic pyloric stenosis (IHPS) is
a common surgical condition encountered in
early infancy, occurring in 2–4 per 1000 live
births. It occurs as an isolated condition or
together with other congenital anomalies. IHPS
is caused by hypertrophy of the pyloric sphincter
muscle, causing narrowing of the pyloric canal
and elongation, resulting in gastric outlet obstruc-
tion. Although the etiology of IHPS is not clearly
understood, it is known that boys are affected
four times more than girls Galea & Said (2018).
The causes of hypertrophic circular muscle of
the pylorus are still obscure, and various hypoth-
eses have been advocated.
Since Ramstedt introduced the extramucosal
pyloromyotomy in 1912, a variety of approaches
to the pylorus have been described.
T. Doi (*)
Department of Pediatric Surgery, Kansai Medical
University, Osaka, Japan
e-mail: doitak@hirakata.kmu.ac.jp
T. Fujimoto
Fujimoto Children’s Clinic, Tokyo, Japan
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_60
60
60.2 Etiology
There is evidence of a genetic predisposition to
the development of this condition. IHPS has been
associated with a number of inherited syndromes
(e.g., Smith-Lemli-Opitz and Cornelia de Lange
syndromes) and with a variety of chromosome
abnormalities. Nonsyndromic IHPS shows famil-
ial aggregation. Siblings of patients with IHPS
are 15 times more likely to suffer from the condi-
tion than children who have no family history of
IHPS.  The classic studies of Carter and Evans
defined the disease as a paradigm for multifacto-
rial, sex-modified threshold model of inheritance.
The cause of hypertrophic circular muscle of the
pylorus in IHPS is thought to result from the fail-
ure of relaxation of the sphincter smooth muscle.
IHPS may be a developmental abnormality in
which the pyloric muscle hypertrophies after
birth. There are sound data showing that the
dimensions of the pyloric muscle at birth are the
same in normal infants as in those destined to
develop IHPS. On the contrary, there have been a
number of reports that have diagnosed this condi-
tion at birth, even in utero. Various hypotheses
have been advocated including abnormal pepti-
dergic innervations, abnormalities of extracellu-
lar matrix proteins, abnormalities of smooth
muscle cells, and abnormalities of intestinal hor-
mones. The last decades have shed some light
with immunohistochemical studies on the enteric
nervous system to clarify the pathogenesis of
799
800
IHPS.  A decrease in the immunoreactivity of
neurotransmitters, such as substance P, soma-
tostatin, neuronal structural proteins of S-100,
GFAP, and D7, has been reported. In addition,
recent studies implicate the cytokine epidermal
growth factors (EGF), glial-derived growth fac-
tors (GDGF), transforming growth factor alpha
(TGF-α), platelet-derived growth factor subtype
BB (PDGF-BB), platelet-derived endothelial cell
growth factor (PDEGF), and insulin-like growth
factor-1 (IGF-1) in the pathogenesis of IHPS.
Recently, it has been proposed that disease
susceptibility may be associated with an altered
expression of neuronal nitric oxide synthesis,
which is the major mediator of smooth muscle
relaxation. In particular, NOS1, the gene encod-
ing neuronal nitric oxide synthase, has been
implicated in the disease pathogenesis by expres-
sion studies, animal models, and genetic analysis
of small IHPS data sets Peeters et al. (2012).
On the other hand, the changing rates of IHPS,
which have been reported in several areas and
countries, have highlighted the importance of
environmental factors such as bottle feeding,
younger maternal age, maternal smoking, and
erythromycin administration in the first 2 weeks
of life Galea & Said (2018).
60.3 Pathology
The characteristic gross pathological feature in
IHPS consists of thickening of the antropyloric
portion of the stomach (“olive-like mass”) and
crowding of redundant and edematous mucosa
within the lumen. Abnormally circumferentially
thickened antropyloric muscle (thickness,
4–6  mm; length, 16–20  mm) separates the nor-
mally distendable portion of the antrum from the
duodenal cap. It stops abruptly at both ends. The
rigid antropyloric canal is unable to accommo-
date the redundant mucosa, which protrudes into
the gastric antrum. These anatomical abnormali-
ties cause obstruction to the passage of gastric
contents. Histologically, IHPS is characterized
by thickened, hypertrophied, and edematous
mucosa and its relationship to the underlying
hypertrophied musculature, primarily involving
T. Doi and T. Fujimoto
the circular muscle. Various results of immuno-
histochemical studies in relation to the enteric
nervous system in IHPS were described in the
etiology section.
60.4 Diagnosis
The diagnosis of IHPS is usually based on clini-
cal history, physical examination, and imaging
studies, such as ultrasonography and barium
meal study.
60.4.1 Clinical Features
The usual onset of symptoms occurs between 2
and 8  weeks of age with peak occurrence at
3–5 weeks of age. It has rarely been reported in
premature infants, especially extremely low birth
weight infants, and those premature infants with
IHPS present the signs and symptoms 2–4 weeks
later compared to normal term infants. The clini-
cal features vary with the length of symptoms.
Initially, the vomiting may not be frequent and
forceful, but over several days, it progresses to
every feed and becomes forceful nonbilious vom-
iting described as “projectile.” The emesis con-
sists of gastric contents, which may become
blood tinged with protracted vomiting and likely
related to gastritis, with “coffee-ground” appear-
ance (17–18% of cases). Infants with IHPS do
not appear ill or febrile in the early stages. A sig-
nificant delay in diagnosis leads to severe dehy-
dration and weight loss due to inadequate fluid
and calorie intake. Severe starvation can exacer-
bate diminished glucuronyl transferase activity
and jaundice associated with indirect hyperbiliru-
binemia, as seen in 2–5% of infants with
IHPS. Associated anomalies are seen in 6–20%
of patients Peters et al. (2014); Puri et al. (2011).
60.4.2 Physical Examinations
It should be possible to diagnose IHPS on clinical
features alone in 80–90% of infants. The impor-
tant diagnostic features are visible gastric peri-
60  Infantile Hypertrophic Pyloric Stenosis
staltic waves in the left upper abdomen and a
palpable enlarged pylorus (olive-like mass).
Physical examination requires a calm and coop-
erative infant with a relaxed abdomen. It is easy
to observe the gastric peristaltic waves after test
feeding in a warm environment. However, to
obtain successful palpation of the hypertrophic
pylorus, it is very difficult if the stomach dis-
tends. Aspiration, using the nasogastric tube,
facilitates the successful palpation of an enlarged
pylorus. After the edge of the liver has been iden-
tified with the fingertip, apply gentle pressure
deep to the liver and progress caudally to reveal
an enlarged pylorus. In most cases, an enlarged
pylorus is located just above the umbilicus at the
lateral border of the rectal muscle below the liver
edge Peters et al. (2014); Puri et al. (2011).
60.4.3 Diagnostic Imaging
Ultrasonographic examination has become the
most common imaging study for the diagnosis of
the IHPS and is easily carried out without any
sedation. The examination should be carried out
Longitudinal Scanning
Fig. 60.1  Ultrasonographic findings of IHPS. Longitudinal ultrasonography shows variable degree of hypertrophied
muscle (arrow) and intervening mucosa that protrudes into the fluid-filled antrum (arrow head)
801
with a high-frequency linear transducer (between
6 and 10  MHz). It is preferably carried out by
placing the patient in a right oblique position,
permitting the fluid to gravitate to the antrum for
adequate evaluation. In patients with IHPS, lon-
gitudinal ultrasonography shows a variable
degree of hypertrophied muscle and the interven-
ing mucosa that protrudes into the fluid-filled
antrum. Cross-sectional studies show circumfer-
ential muscular thickening surrounding the cen-
tral channel filled with mucosa (Fig. 60.1). The
diagnosis of IHPS by ultrasonography is made by
identifying these findings, and the most com-
monly used criteria are pyloric muscle thickness
of 4 mm or more and pyloric channel length of
16  mm or more. Barium meal study can be
employed in equivocal cases of IHPS. The char-
acteristic radiological feature of IHPS is a nar-
rowed elongated pyloric canal giving a “string”
or “double track” sign caused by compressed
invaginated folds of mucosa in the pyloric canal
(Fig.  60.2). However, barium meal study pro-
vides indirect information about the antropyloric
canal status. The failure of the relaxation of the
antropyloric lesion, known as pylorospasm, dem-
Cross section
802 T. Doi and T. Fujimoto

Table 60.1  Differential diagnosis of IHPS

Surgical conditions
Gastric volvulus
Antral web
Duplication cyst
Preampullar duodenal
stenosis
Ectopic pancreas within the
pyloric muscle
Medical conditions
Gastroenteritis
Increased intracranial
pressure
Metabolic disease
Pylorospasm
Gastroesophageal reflux

than IHPS. Common medical causes of nonbilious

Fig. 60.2  Barium meal study of IHPS. Narrowed, elon-
gated pyloric canal giving a “string” or “double track”
sign caused by compressed invaginated folds of mucosa in
the pyloric canal (arrow)
vomiting are gastroenteritis, increased intracranial
pressure, and metabolic disorders. Associated
anomalies including esophageal atresia, intestinal
malrotation, Hirschsprung’s disease, anorectal
malformations, and urogenital anomalies are found
in 6–20% Pandya & Heiss (2012).

onstrates the same findings as those of IHPS. The
emptying speed of the barium meal to the distal
bowel is important to differentiate these two con-
ditions Peters et al. (2014); Puri et al. (2011).
60.5 Differential Diagnosis
Several conditions must be considered if the patient
demonstrates nonbilious vomiting. Table  60.1
gives the list of common differential diagnosis.
Patients with bilious vomiting do not have IHPS
and are not directed to an initial ultrasonographic
examination. Pylorospasm and gastroesophageal
reflux (GER) give similar clinical findings, and it
may be difficult to differentiate them from IHPS
without further evaluation. However, both condi-
tions are more easily excluded with ultrasonogra-
phy than barium meal study because of the ability
of the former to detect and measure the antropylo-
ric muscle thickness. Herniation of gastric fundus
and regurgitation of gastric contents in patients
with GER can also be identified by ultrasonogra-
phy. Other surgical causes of nonbilious vomiting
include gastric volvulus, antral web, preampullar
duodenal stenosis, duplication cyst of the antropy-
loric lesion, and ectopic pancreatic tissue within an
antropyloric muscle, which are all far less common
60.6 Management
60.6.1 Preoperative Management
Persistent vomiting in these patients results in
chloride and potassium depletion with metabolic
alkalosis. The estimation of serum electrolyte
level, urea nitrogen level, hematocrit, and blood
gases must be carried out to determine the state
of dehydration and acid-base abnormalities.
Fluid resuscitation should be based on the degree
of dehydration and the extent of electrolyte
abnormalities. Nowadays, many babies with
IHPS do not show any clinical evidence of dehy-
dration and electrolyte abnormalities on admis-
sion. If the infant develops mild dehydration and
hypochloremic alkalosis, maintenance fluid with
5% dextrose in 0.45% normal saline containing
20–40 mEq/l potassium chloride can be adminis-
trated. The operation for IHPS is not an emer-
gency and should never be undertaken until
serum electrolytes level and acid-base balance
have returned to normal. There has been much
debate regarding the preoperative placement of
the nasogastric tube.
Most infants with IHPS do not have complete
gastric outlet obstruction and can tolerate their gas-
tric secretion. A nasogastric tube removes additional
fluid and hydrochloric acid from the stomach.
60  Infantile Hypertrophic Pyloric Stenosis
60.6.2 Operation
The Ramstedt’s pyloromyotomy for IHPS is uni-
versally accepted. A 2–3 cm transverse incision at
the right upper quadrant provides excellent expo-
sure of the pylorus. Another incision that is com-
monly used is a supraumbilical skin-fold incision
Kapoor et al. (2019). The hypertrophied antropy-
loric lesion is delivered by gentle traction through
the surgical wound (Fig. 60.3a). Then, a longitu-
dinal serosal incision is made on the anterosupe-
rior aspect of the pylorus, beginning approximately
1–2 mm proximal to the duodenum and extended
into the nonhypertrophied antrum. The blunt end
of the scalpel handle is used to initially disrupt the
a b
Fig. 60.3  Operative findings of hypertrophied antropylo-
ric lesion. (a) Hypertrophic antropyloric lesion is deliv-
ered through the surgical wound. (b) After spreading the
803
muscle fiber. The hypertrophied circular muscle is
then further disrupted down to the mucosa using
the Benson spreader (Fig.  60.3b). When the
pyloric muscle is adequately split, the mucosa can
be seen to be bulging (Fig. 60.3c).
Recent studies have shown that, by using lapa-
roscopic approach, complications, operative times,
and postoperative recovery times can be reduced
when compared to the open approach, though it
was initially associated with incomplete myoto-
mies and higher complication rates. However, there
is not much cosmetic advantage when compared to
the umbilical approach. There is therefore still no
strong evidence to support a recommendation of
one procedure over the other Huang et al. (2020).
pyloric muscle, loose prolapsing of intact mucosa is an
evidence of satisfactory myotomy. (c) Prolapsing mucosa
after pylorus myotomy
804
60.6.3 Postoperative Feeding
There has been much debate in relation to the
timing of initiation of the enteral feeds following
pyloromyotomy. The ultimate goal of initiating
early feeding is to achieve full feeds more quickly
and therefore safely decrease total hospital stay.
Starting feeds shortly after recovery (earlier than
6  h postoperatively) is an accepted method for
feeding and allows most infants to advance to full
feeding without any complications and risks.
60.6.4 Nonoperative Treatment
There is no strong evidence-based support for the
treatment of IHPS with atropine. Comparative but
nonrandomized studies indicate that atropine is
less effective than pyloromyotomy in the treat-
ment of IHPS. Although the incidence of compli-
cations is comparable between the two groups,
pyloromyotomy significantly shortens the length
of hospital stay. Therefore, it is recommended that
atropine is used only in selected ISPS, such as
those with a history of major abdominal surgical
procedure or those at high risk for general anes-
thesia or surgical procedures Lauriti et al. (2018).
60.7 Complications
Regardless of operative approach, perioperative
complications of surgery for IHPS including
wound-related complications, mucosal perfora-
tion, and incomplete pyloromyotomy rarely occur.
A recent meta-analysis revealed a 4% higher rate
of incomplete pyloromyotomy with the laparo-
scopic technique compared to open surgery.
Conversely, an overzealous pyloromyotomy leads
to full-thickness division of the pylorus with per-
foration Kelay & Hall (2018); Oomen et al. (2012).
T. Doi and T. Fujimoto
60.8 Conclusion
Epidemiological studies suggest that the preva-
lence of IHPS has decreased in the USA and
Europe in recent years. The strong family aggre-
gation of IHPS and co-occurrence with multiple
genetic syndromes suggest a strong genetic basis
for IHPS.  Improvements in genome-scan tech-
niques and increasing knowledge of the genetic
basis of diseases should help to search for IHPS
susceptibility genes Zhu et al. (2017).
Although recent studies have shown that the
laparoscopic approach can reduce complications,
operative times, postoperative recovery times,
and tolerance of early feedings compared to the
open approach, there is not much cosmetic
advantage when compared to the open umbilical
approach Sathya et al. (2017).
References
Galea R, Said E (2018) Infantile hypertrophic pyloric
stenosis: an epidemiological review. Neonatal Netw
37(4):197–204
Huang WH, Zhang QL, Chen L, Cui X, Wang YJ, Zhou
CM (2020) The safety and effectiveness of laparo-
scopic versus open surgery for congenital hyper-
trophic pyloric stenosis in infants. Med Sci Monit
26:e921555
Kapoor R, Kancherla V, Cao Y, Oleson J, Suhl J, Canfield
MA, Druschel CM, Kirby RS, Meyer RE, Romitti PA
(2019) Prevalence and descriptive epidemiology of
infantile hypertrophic pyloric stenosis in the United
States: a multistate, population-based retrospective
study, 1999-2010. Birth Defects Res 111(3):159–169
Kelay A, Hall NJ (2018) Perioperative complications of
surgery for hypertrophic pyloric stenosis. Eur J Pediatr
Surg 28(2):171–175
Lauriti G, Cascini V, Chiesa PL, Pierro A, Zani A (2018)
Atropine treatment for hypertrophic pyloric stenosis:
a systematic review and meta-analysis. Eur J Pediatr
Surg 28(5):393–399
Oomen MW, Hoekstra LT, Bakx R, Ubbink DT, Heij HA
(2012) Open versus laparoscopic pyloromyotomy for
hypertrophic pyloric stenosis: a systematic review and
60  Infantile Hypertrophic Pyloric Stenosis
meta-analysis focusing on major complications. Surg
Endosc 26(8):2104–2110
Pandya S, Heiss K (2012) Pyloric stenosis in pediatric
surgery: an evidence-based review. Surg Clin North
Am 92(3):527–539
Peeters B, Benninga MA, Hennekam RC (2012) Infantile
hypertrophic pyloric stenosis--genetics and syn-
dromes. Nat Rev Gastroenterol Hepatol 9(11):646–660
Peters B, Oomen MW, Bakx R, Benninga MA (2014)
Advances in infantile hypertrophic pyloric stenosis.
Expert Rev Gastroenterol Hepatol 8(5):533–541
805
Puri P, Kutasy B, Lakshmanadass G (2011) Hypertrophic
pyloric stenosis. In: Puri P (ed) Newborn surgery.
Hodder Arnold, London, pp 433–442
Sathya C, Wayne C, Gotsch A, Vincent J, Sullivan KJ,
Nasr A (2017) Laparoscopic versus open pyloromyot-
omy in infants: a systematic review and meta-analysis.
Pediatr Surg Int 33(3):325–333
Zhu J, Zhu T, Lin Z, Qu Y, Mu D (2017) Perinatal risk fac-
tors for infantile hypertrophic pyloric stenosis: a meta-­
analysis. J Pediatr Surg 52(9):1389–1397
 Gastrostomy and Jejunostomy
Julia Brendel and Michael W. L. Gauderer
61.1 Introduction
Enterostomies are not only applied for major
congenital anomalies of the abdominal wall and
gastrointestinal tract, but even more frequently,
for a variety of nonsurgical conditions. Typically,
these are children with central nervous lesions
and swallowing difficulties or patients requiring
feeding supplementation for cancer treatment,
trauma, or chronic malnutrition (Gauderer 2012;
Abdelhadi et  al. 2016; Zimmer and Gauderer
2020). Enterostomies are usually implemented
for long-term enteral feeding purposes and can
be used for drug administration and, in the case
of a gastrostomy, also gastric venting. For short
and intermediate access methods, please see
Chap. 10.
61.2 Gastrostomy
Gastrostomy is the preferred access for long-term
enteral feeding, both continuous and bolus, with-
out interference between oral feedings and physi-
J. Brendel (*)
Department of Pediatric Surgery, Hannover Medical
School, Hannover, Germany
e-mail: brendel.julia@mh-hannover.de
M. W. L. Gauderer
Surgery and Pediatrics, University of South Carolina,
Greenville, SC, USA
University of South Carolina, School of Medicine
Greenville, Greenville, SC, USA
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_61
61
ological activities. It also works for gastric
venting and drug application. Although consid-
ered physiologic and well tolerated, gastrosto-
mies have potential early and late complications
leading to considerable morbidity and mortality
(Baker et al. 2015; Landisch et al. 2016; Chatwin
et  al. 2013; Liu et  al. 2013; Sutherland et  al.
2017; Thaker and Misra 2019).
61.2.1 Indications
61.2.1.1 Esophageal Abnormalities
Gastrostomies are indicated for patients with
esophageal atresia without fistula and staged pro-
cedures or in the case of associated anomalies
that may interfere with feeding. Esophageal stric-
tures, such as those secondary to the ingestion of
alkaline substances, are also often bypassed by
this enterostomy (Gauderer and Zimmer 2017).
61.2.1.2 Duodenal Obstruction
Although rarely performed, gastrostomies may
be employed if prolonged gastric decompression
is anticipated after the repair of a duodenal
obstruction. A fine silicone rubber catheter can be
placed alongside a gastrostomy catheter across
the anastomosis and into the proximal jejunum
during the initial procedure, potentially decreas-
ing or eliminating the need for parenteral nutri-
tion (Zimmer and Gauderer 2020).
807
808
61.2.1.3 Short Bowel Syndrome
The loss of 50% of small bowel (e.g., due to atre-
sia or necrotizing enterocolitis) leads to a pro-
found alteration in gastrointestinal physiology.
Prolonged drainage might be necessary due to
gastric hypersecretion and adaptive changes of
the remaining intestine. Both processes can be
fairly lengthy, making a gastrostomy a suitable
access method (Gauderer and Zimmer 2017).
61.2.1.4 Other Pathologies
Any (nonsurgical) condition leading to failure to
thrive or an inability to swallow, requiring feed-
ing supplementation or non-palatable medica-
tions (e.g., malignancies, trauma, chronic
malabsorption syndromes), is a suitable indica-
tion for a gastrostomy (Zimmer and Gauderer
2020). Anti-reflux procedures may be added for
neurologically impaired children, who frequently
have foregut dysmotility and gastroesophageal
reflux, in addition to swallowing difficulties.
Gastrostomies can also facilitate the manage-
ment of any surgical condition in which a pro-
longed ileus or partial luminal occlusion (e.g.,
complicated meconium ileus, small bowel
Hirschsprung’s disease) is anticipated or in whom
a complex feeding regimen is likely (e.g., those
with intestinal lymphangiectasia). Although
decompressive gastrostomies are not routinely
indicated for major wall defects, they can be
helpful in patients with severe gastroschisis and
associated atresia, particularly those requiring
long-term continuous feeding.
61.2.2 Choice of Technique
A wide variety of gastrostomy techniques are
available. There are three basic types (Gauderer
and Stellato 1986; Gauderer 2006):
(a) The formation of a channel from the anterior
gastric wall to the skin surface around a cath-
eter. The catheter is placed in the stomach
and exits vertically to the gastric wall (Stamm
gastrostomy).
J. Brendel and M. W. L. Gauderer
(b) The formation of a tube or conduit from a
full-thickness gastric wall to the skin surface
(Witzel gastrostomy).
(c) Percutaneous techniques, in which the intro-
duced catheter holds the gastric and abdominal
wall in apposition, with or without the aid of
special fasteners (Stamm-type gastrostomy).
With certain modifications, each of these
methods can be performed, employing minimally
invasive approaches. Table 61.1 demonstrates the
most commonly used gastrostomies and their
characteristics.
61.2.2.1 Open Technique
The open “Stamm” technique is the most widely
applied gastrostomy via laparotomy, suitable for
children of all sizes, either as an isolated inter-
vention or along with other intra-abdominal pro-
cedures (Gauderer 2006).
General anesthesia is preferred for gastros-
tomy placement because abdominal wall relax-
ation is required. The child is placed on the table
with a small roll behind the back. If the underly-
ing conditions allows it, a nasogastric tube is
inserted for decompression and identification of
the stomach. A small transverse incision is made
over the left upper rectus abdominis muscle
(Fig. 61.1). An incision too high would bring the
catheter too close to the costal margin, while an
incision too low bears the risk of injury to the
colon and small bowel. Alternatively, a short ver-
tical incision can be used, although this approach
is less desirable because the linea alba is the thin-
nest area of the abdominal wall. Fascial layers are
incised transversely, and the rectus muscle is
retracted or transected. If the identification of the
stomach is difficult, downward traction of the
flimsy greater omentum usually allows the visu-
alization of the transverse colon and stomach.
The correct placement of the gastrostomy on the
anterior gastric wall is crucial. It should not be
too close to the greater curvature to avoid the so-­
called gastric pacemaker and to minimize the
potential for gastrocolic fistula. On the anterior
gastric wall, a position midway between the pylo-
61  Gastrostomy and Jejunostomy 809

Table 61.1  A comparison of the most commonly used gastrostomies. Adapted from Puri and Höllwarth, Pediatric
Surgery: Diagnosis and Management, Springer 2009, page 371 (Gauderer 2009)
Percutaneous
Serosa-­ Percutaneous imaging-guided Laparoscopic and
lined Gastric endoscopic “radiological” laparoscopically
channels tubes techniques techniques assisted techniques
Catheter/stoma device Yes No Yes Yes Yes
continuously in use
Laparotomy Yes Yes No No No
Laparoscopically Possible Yes Yes N/A Yes
feasible
Need for gastric No No Yes No No
endoscopy
Need for abdominal Yes Yes No No Yes and insufflation
relaxation during
operation
Procedure time Short Moderate Very short Short Short
Postoperative ileus Yes Yes No No Some
Potential for bleeding Yes Yes Remote Remote Small
Potential for wound Yes Yes No No No
dehiscence/hernia
Potential for early Yes No Rare Yes Small
dislodgement of
catheter
Potential for gastric Possible Possible Yes Yes Possible
separation
Potential for infection Yes Yes Yes Yes Yes
Potential for Low No Yes Low Low
gastrocolic fistula
Incidence of external Moderate Significant Low Low Low
leakage
Suitable for passage of Yes No No No Possible
dilators for esophageal
stricture

rus and the esophagus close to the smaller curva-
ture should be chosen. The site should neither be
too high, because this would interfere with a fun-
doplication, should one be needed in the future,
nor too low, because stomas at the level of the
antrum are prone to pyloric obstruction by the
catheter (Gauderer and Stellato 1986).
The anterior gastric wall is lifted with two guy
sutures at the site of the stoma, ensuring that the
posterior wall is not included. A concentric purse
string suture (synthetic, absorbable material) is
placed (Fig. 61.2). The gastrostomy, at the center
of the purse string, is made sharply through the
serosa and muscular wall of the stomach.
A small hemostat is introduced to confirm
access into the gastric lumen. A mushroom-type
catheter (Pezzer), with sizes 12–14 French gauge,
is preferable for neonates. The mushroom head
of the catheter is stretched with a short stylet to
allow atraumatic introduction into the stomach
(Fig. 61.2). The purse string is tied to invert the
seromuscular gastric wall around the tube
(Figs. 61.3 and 61.4). Other suitable catheters are
the Malecot or the “T tube,” but both have the
disadvantage of becoming more easily dislodged.
However, a short “T tube” is useful if the stomach
is very small. Long balloon-type catheters, which
may rupture, also have a greater propensity for
distal migration into the small bowel. Skin-level
devices (buttons or balloon-type) may be inserted
during the operation, instead of the traditional
long tubes. The exit site for the catheter should be
through the midportion of the rectus muscle,
about 1–2  cm above or below the laparotomy
incision. The exit of the catheter through the pri-
mary abdominal incision bears the risk of com-
810 J. Brendel and M. W. L. Gauderer

Fig. 61.2  Gastrostomy site on the anterior gastric wall.

The traction guy suture and the purse string suture are
depicted. Introduction of a Pezzer catheter. From Gauderer
MWL. Gastrostomy. In: Spitz L, Coran AG, eds. Operative
pediatric surgery. 6th ed. London: Hodder Arnold; 2006.
Fig. 61.1  Gastrostomy incision and catheter exit site. An p. 330–355
alternative is a short vertical midline incision. From
Gauderer MWL. Gastrostomy. In: Spitz L, Coran AG, eds.
Operative pediatric surgery. 6th ed. London: Hodder
Arnold; 2006. p. 330–355

plex wound complications (Zimmer and Gauderer

2020). Once the exit site is chosen, the anterior
gastric wall is secured to the posterior aspect of
the anterior abdominal wall with four equidistant
sutures or with a continuous suture of double-­
ended synthetic monofilament thread (Wilson
and Oliva-Hemker 2001). The catheter position is
tested by injecting and aspirating saline. Gentle
traction on the catheter assures that its intragas-
tric position is maintained.
The posterior rectus sheath is closed with a
running suture of absorbable, synthetic material.
The anterior rectus sheath is approximated with
interrupted sutures of the same material. The sub-
cutaneous layer is closed with synthetic, absorb-
able sutures. The skin can be approximated with Fig. 61.3  The purse string suture is tied. Partial place-
either interrupted or continuous subcuticular ment of the continuous monofilament suture, used to
sutures. Adhesive strips cover the incision. The anchor the stomach to the anterior abdominal wall. The
catheter is firmly secured with two sutures of catheter is brought out through the counter-incision. From
Gauderer MWL. Gastrostomy. In: Spitz L, Coran AG, eds.
synthetic monofilament thread. These sutures are Operative pediatric surgery. 6th ed. London: Hodder
removed after 1  week, and a small crossbar is Arnold; 2006. p. 330–355
61  Gastrostomy and Jejunostomy
Fig. 61.4  The continuous monofilament suture placement
is continued anteriorly and then tied, providing a 360° fixa-
tion of the stomach to the anterior abdominal wall with a
watertight seal. From Gauderer MWL.  Gastrostomy. In:
Spitz L, Coran AG, eds. Operative pediatric surgery. 6th
ed. London: Hodder Arnold; 2006. p. 330–355
placed loosely to prevent distal catheter migra-
tion. The conversion of a long tube to a “button”
can be performed after a firm adherence between
the gastric and abdominal walls is established
(Gauderer and Zimmer 2017).
61.2.2.2 Percutaneous Endoscopic
Gastrostomy (PEG)
The first gastrostomy without laparotomy was
the percutaneous endoscopic gastrostomy (PEG),
initially developed for high-risk pediatric patients
and eventually became known as the “pull” tech-
nique (Gauderer et  al. 1980). The procedure is
suitable for children of all sizes and ages, usually
for the purpose of long-term enteral feeding
(Heuschkel et  al. 2015). Although there is no
need for abdominal wall relaxation, general
endotracheal anesthesia is recommended to pro-
tect the airway from compression during endos-
copy. The procedure is usually very short with no
postoperative ileus and only minimal interfer-
ence with subsequent interventions on the stom-
811
ach, such as a fundoplication. The main
disadvantage of this and other pure endoscopic
techniques is that the virtual space between the
stomach and the abdominal wall cannot be visu-
alized. Typically, long tubes are initially
employed; however, a primary insertion of a
skin-level device is also possible (Novotny et al.
2009).
The (“pull”) PEG technique incorporates,
according to Gauderer (2006), these basic
elements:
(a) Gastroscopic insufflation brings the stomach
into apposition to the abdominal wall.
(b) With the stomach apposed to the abdominal
wall, a cannula is introduced percutaneously
into the gastric lumen under direct endo-
scopic guidance (Fig. 61.5).
(c) The cannula serves as access to introduce a
guide wire, which is then withdrawn out of
the patient’s mouth with the gastroscope. A
tract is thus established.
(d) A PEG catheter with a tapered end is attached
to the oral end of the guide-wire and pulled
in a retrograde fashion until it’s final posi-
tion, keeping the stomach firmly apposed to
the abdominal wall.
There are multiple variations of the original
PEG technique and several types of catheters
(Gauderer and Stellato 1986; Gauderer et  al.
1980; Novotny et  al. 2009). For neonates and
infants, a 16 French gauge (or smaller), a com-
mercially available silicone rubber pediatric PEG
catheter, is recommended. Larger, stiffer cathe-
ters, or those with a stiff, non-collapsible inner
retainer, can easily tear the infant’s esophagus.
Technically, PEGs are employed as follows
(Gauderer 2006):
The administration of a single shot of IV
broad-spectrum antibiotics and supine position of
the patient. The abdomen is cleansed and ster-
ilely draped. Gastroscopy is performed with the
smallest pediatric gastroscope available. The
scope is inserted and advanced slowly into the
stomach, at which point the light is seen through
the left upper quadrant abdominal wall. With the
812
Fig. 61.5 Percutaneous endoscopic gastrostomy (PEG).
Insufflation of air through the endoscope to approximate the
stomach to the abdominal wall and displace the colon cau-
dally. Digital pressure is applied to the proposed gastrostomy
site, which usually corresponds to the area where transillumi-
nation is brightest. Transillumination and clear visualization
of an anterior gastric wall indentation are key points. Drawn
J. Brendel and M. W. L. Gauderer
of long-lasting local anesthetic into a syringe and injection of
the proposed PEG. The needle is advanced further, and con-
tinuous aspiration pressure is applied to the plunger. Air aspi-
ration should only occur when the tip of the needle enters the
gastric lumen. From Gauderer MWL. Gastrostomy. In: Spitz
L, Coran AG, eds. Operative pediatric surgery. 6th ed.
London: Hodder Arnold; 2006. p. 330–355
61  Gastrostomy and Jejunostomy
gastroscope in place, insufflation distends the
stomach, apposes it against the anterior abdomi-
nal wall, and displaces the colon downward.
When the room lights are dimmed, the gastric
contour is clearly visible, particularly in small
children.
The preferred gastrostomy site is over the mid-
portion of the left rectus muscle. Digital pressure
is exerted at this site (Fig. 61.5), which is seen by
the endoscopist as a “polypoid lesion” or “mound”
Fig. 61.6  A small incision is made, and a Kelly-type
hemostat is applied to maintain the intragastric indenta-
tion. The endoscopist places the polypectomy snare
around the “mound”; the cannula is introduced between
the slightly spread prongs of the hemostat and then thrust
813
on the anterior gastric wall (gastric transillumina-
tion and endoscopically visualized digital inden-
tation of the stomach are the most important
factors in safe PEG placement). The endoscopist
then places an endoscopic polypectomy snare
around this invagination of the anterior gastric
wall. Digital pressure is released and a small skin
incision is made. A hemostat with slightly opened
prongs is placed in the incision, recreating and
maintaining the intragastric “mound” (Fig. 61.6).
through the abdominal and gastric walls into the open
snare. The snare is partially closed but not tightened
around the cannula. From Gauderer MWL. Gastrostomy.
In: Spitz L, Coran AG, eds. Operative pediatric surgery.
6th ed. London: Hodder Arnold; 2006. p. 330–355
814
Fig. 61.7  The needle is removed and the guide wire
inserted. The polypectomy snare (alternatively alligator or
biopsy forceps) grasps the guide and exits it through the
Through this incision and through the prongs of
the hemostat, a 16 gauge, smoothly tapered, an IV
cannula, and a needle are thrust through the
abdominal and gastric walls under endoscopic
visualization. This should be performed quickly
to avoid displacing the stomach from the abdomi-
nal wall. The snare, if initially positioned prop-
erly, will be around the advancing cannula. If not,
it can be maneuvered to encircle the cannula. A
long, monofilament synthetic suture or a plastic-
covered steel guide wire is then advanced through
the cannula and grasped by the snare (Fig. 61.7).
If there is difficulty with the snare, a biopsy or
alligator-type forceps may help. As the gastro-
scope and snare are withdrawn, the suture is
brought out of the patient’s mouth (Fig. 61.7). The
previously selected PEG catheter is then con-
nected to the suture outside the patient’s mouth,
and both the suture and catheter are coated with a
water-­soluble lubricant. Traction on the abdomi-
nal portion of the suture or guide wire pulls the
J. Brendel and M. W. L. Gauderer
mouth. From Gauderer MWL. Gastrostomy. In: Spitz L,
Coran AG, eds. Operative pediatric surgery. 6th ed.
London: Hodder Arnold; 2006. p. 330–355
catheter in a retrograde fashion, through the
mouth, esophagus, and stomach, and across the
abdominal wall (Fig.  61.8). The gastroscope is
reintroduced to verify the catheter position under
direct vision.
Traction on the catheter is continued until the
inner catheter retainer or “dome” loosely touches
the gastric mucosa (Fig. 61.8). Markings on the
commercially available catheters, or markings
added to tubes without marks, are helpful in indi-
cating the final position of the tube. The external
crossbar is then placed (Fig.  61.8). Excessive
pressure by the external crossbar on the abdomi-
nal wall will produce pressure necrosis and even-
tual catheter extrusion and should be avoided.
The tapered catheter end is cut off and a connec-
tor is attached. Tape is used for temporary cathe-
ter immobilization. The catheter may be
converted to a skin-level device by using the
external port valve at any time. To replace the
PEG catheter with a “button” or balloon-type
61  Gastrostomy and Jejunostomy
Fig. 61.8 The appropriately sized PEG catheter is
attached to the oral end of the guide wire and pulled in a
retrograde manner through the infant’s esophagus and
stomach and then across the gastric and abdominal walls.
The inset shows the position of the catheter at the end of
the procedure. From Gauderer MWL.  Gastrostomy. In:
Spitz L, Coran AG, eds. Operative pediatric surgery. 6th
ed. London: Hodder Arnold; 2006. p. 330–355
skin-level device, firm adhesions between the
stomach and abdominal wall should be estab-
lished. This may take 1–3  months or longer
(Gauderer and Zimmer 2017).
Although the PEG catheter can be used imme-
diately after insertion, some centers wait for 24 h,
or even longer, before starting enteral feedings.
However, several studies have shown that early
feedings (within 3–6 h) are not associated with a
higher complication rate (Jensen et  al. 2018;
Corkins et  al. 2010; Wiernicka et  al. 2019;
Watanabe and Kotani 2020).
PEG catheters should routinely be replaced
every 2 years, which is usually performed endo-
scopically as the internal flange is mostly not col-
lapsible (Volpe and Malakounides 2018).
However, some devices are designed to collapse
and may be pulled through the track externally,
but this bears the risk of track disruption with the
force of pulling the internal bolster through the
track causing serious injury and complications
(Heuschkel et al. 2015).
815
Various other methods of gastrostomy with-
out laparotomy have been introduced over the
years. The percutaneous endoscopic “push”
technique is often applied, but not as popular a
technique as the PEG among endoscopists due to
its technical difficulties and associated compli-
cations (Heuschkel et al. 2015).
The anterior gastric wall is fixed to the wall
under endoscopic control by either suturing or
T-fasteners. The puncture site is identified at the
center of the gastropexy under endoscopic guid-
ance followed by the Seldinger method of guide
wire introduction. After progressive tract dilata-
tions, a long tube or skin-level gastrostomy
device can be inserted. The opportunity for a pri-
mary skin-level device is a major advantage of
the push technique over a traditionally placed
PEG tube because it avoids a second general
anesthetic for the removal of the tube and replace-
ment with a low-level device (Heuschkel et  al.
2015). Although new devices are now available
with appropriate sizes for children (14 and 16
French), the published experience in children
remains limited (Heuschkel et al. 2015).
61.2.2.3 Minimally Invasive
Gastrostomy Techniques
Laparoscopic-aided techniques are expansions of
the methods mentioned above and can also be
employed in infants. The insertion of a gastros-
tomy device can be performed either entirely
with laparoscopic assistance (LAP), or laparo-
scopic control can be employed to assist with the
PEG approach (LA-PEG) (Thaker and Misra
2019; Sandberg et  al. 2018; Baker et  al. 2013,
2015; Livingston et al. 2015; Vasseur Maurer and
Reinberg 2015).
Several comparative studies have been pub-
lished regarding the outcome, efficiency, and
complications of different open, minimally inva-
sive, and endoscopic techniques, favoring laparo-
scopic procedures (Thaker and Misra 2019;
Petrosyan et al. 2016; Landisch et al. 2016; Liu
et al. 2013; Baker et al. 2015; Merli et al. 2016;
Sulkowski et al. 2016).
Over the years, numerous variations for lapa-
roscopically controlled or aided gastrostomy
have been introduced, including various altera-
816
tions using one or two trocar techniques, microin-
vasive strategies, or different suturing/stitching
methods (Baker et al. 2013; Mason et al. 2018;
Lotti et al. 2020; Kaya et al. 2018; Fujiogi et al.
2018; Sayadi Shahraki et al. 2020; Villalona et al.
2011; Antonoff et al. 2009; Backman et al. 2010).
Laparoscopically Assisted Gastrostomies
Direct visualization by a laparoscope adds safety
and expands technical variations for constructing
a gastrostomy. In addition to videoscopically
controlled PEG, the two most common methods
are adaptations of the Stamm technique and mod-
ifications of the “push” PEG (Vasseur Maurer
and Reinberg 2015; Livingston et al. 2015).
Minimally invasive gastrostomies are applied
under general anesthesia. A single shot of IV
broad-spectrum antibiotics is administered, and
the patient is placed on the table in a supine posi-
tion. The trocar site must be sufficiently enlarged
to place a purse string suture through the exposed
segment of the anterior gastric wall. However,
this may predispose the site to leakage. The
­stomach is temporarily anchored to the abdomi-
nal wall via T-fasteners and U stitches. A naso-
gastric tube is inserted and the most suitable site
for the gastrostomy selected in the left upper
quadrant and marked. A trocar is placed at the
umbilicus, pneumoperitoneum is established,
and the laparoscope is introduced. A needle is
pushed through the previously marked abdomi-
nal wall site and the appropriate relation between
the anterior gastric wall and the stoma site estab-
lished. A 5  mm trocar is inserted after a small
skin incision. A grasper is introduced, and the
gastrotomy site on the anterior gastric wall is
lifted toward the parietal peritoneum (Fig. 61.9).
A U stitch is passed through the abdominal wall,
through the anterior gastric wall, and back out
through the abdominal wall. A second U stitch is
passed parallel to the first one, 1–2  cm apart
(Figs.  61.9, 61.10, and 61.11). The sutures are
lifted, maintaining the stomach in contact with
the abdominal wall (Fig. 61.11). The grasper and
trocar are removed, and the stomach is insuf-
flated with air through the nasogastric tube. A
needle is then inserted through the trocar site into
J. Brendel and M. W. L. Gauderer
the gastric lumen, between the two U stitches. A
Seldinger-type guide wire is passed through the
needle into the stomach (Fig. 61.10). The tract is
dilated over the guide wire to the size required to
insert either a Foley-type catheter or a balloon-
type skin-level device. These are placed over the
same guide wire. A thin metallic dilatator may be
helpful for the stiffening of the catheter shaft.
The previously placed U-stitches are tied over the
wings of the “button” (Fig. 61.12). If a long tube
is placed, a pair of bolsters is employed. Care
must be taken to avoid excessive tension.
A recent study found that the primary place-
ment of low-profile gastrostomy buttons seems
superior regarding tube-related complications,
such as tube dislodgement or leakage, compared
with that of long gastrostomy tubes (Cortez et al.
2020).
In experienced centers, elective laparoscopic
gastrostomy tube placement in children requires a
median hospital stay length of 2  days (Williams
et al. 2019). It has also been shown that the initiation
of feeds within 4 h of laparoscopic gastrostomy and
discharge within 24 h are feasible and safe (Williams
et al. 2019; Hendrickson et al. 2018).
Laparoscopic-Assisted Percutaneous
Endoscopic Approach
For the optimal visualization of the structures,
laparoscopy has been added to the conventional
PEG in various techniques (Livingston et  al.
2015; Akay et  al. 2010; Hassan and Pimpalwar
2011; Smitherman and Pimpalwar 2009).
The gastroscope is placed within the stomach,
and a laparoscopic port is introduced trough the
umbilicus. After the achievement of pneumoperi-
toneum, a camera is introduced into the perito-
neal cavity, the ideal gastrostomy site is chosen,
and a skin incision is made for the abdominal
wall exit site. Under direct visualization by both
the gastroscope and laparoscope, T-fasteners are
placed from the gastrostomy site on the abdomi-
nal wall into the insufflated stomach. A needle is
placed between the sutures from the abdominal
incision into the stomach. A guide wire is then
introduced. Under direct visualization, an appro-
priately sized dilator peel-away sheath is passed
61  Gastrostomy and Jejunostomy
Fig. 61.9  Laparoscopically assisted gastrostomy. After
the establishment of pneumoperitoneum, a trocar is
inserted at the gastrostomy site, a grasper is introduced,
and the appropriate portion of the anterior gastric wall is
over this guide wire into the stomach. The guide
wire and dilators are removed. An appropriately
sized balloon-type button is introduced through
the peel-away sheath. The latter is removed and
the balloon inflated with saline (usually 3–5 ml).
The anchoring sutures are lifted outside and tied
subcutaneously. An inspection with both gastro-
scope and laparoscope ensures a proper gastros-
tomy device placement (Hassan and Pimpalwar
2011; Livingston et al. 2015).
817
lifted. Placement of two sutures in the depicted manner.
From Gauderer MWL.  Gastrostomy. In: Spitz L, Coran
AG, eds. Operative pediatric surgery. 6th ed. London:
Hodder Arnold; 2006. p. 330–355
As a variation, besides the umbilical port, two
transabdominally incisions can be made for lapa-
roscopic working instruments, and only three
sutures may be placed around a proposed gastros-
tomy site (Smitherman and Pimpalwar 2009).
Although multiple variations exist, the princi-
ples are essentially the same: secure approxima-
tion of the stomach to the abdominal wall,
protection of adjacent organs, and safe and
tension-­free gastrostomy device placement.
818
Fig. 61.10  Insufflation of the stomach with air through
the nasogastric tube. Insertion of a needle through the tro-
car site into the gastric lumen, between the two U stitches.
A Seldinger-type guide wire is passed through the needle
into the stomach. The tract is dilated over the guide wire
to the size required to insert either a Foley-type catheter or
J. Brendel and M. W. L. Gauderer
a balloon-type skin-level device or another low-profile
access device. These are placed over the same guide wire.
From Gauderer MWL.  Gastrostomy. In: Spitz L, Coran
AG, eds. Operative pediatric surgery. 6th ed. London:
Hodder Arnold; 2006. p. 330–355
61  Gastrostomy and Jejunostomy 819

61.2.3 Access Devices

There are several types of commonly used gas-

trostomy tubes. Table 61.2 shows a comparison of
these devices. Long catheters are characterized by
the mode of retention, meaning semirigid intra-
gastric portion and inflatable intragastric balloon.
Skin-level devices, the so-called buttons, are
modifications in which the intragastric retention
mechanism is similar to those of the long tubes.
The diameter of each device is chosen accord-
ing to the size of the child and the purpose of the
gastrostomy. For infants and small children,
12–16 French tubes are well suited, while for
older children, sizes 18–20 French are used.
Long tubes, such as Foley or balloon cathe-
Fig. 61.11  Laparoscopic gastrostomy procedure, intra-­ ters, are easy to insert, but not only bear the risk
abdominal view. U sutures are placed. The sutures are of dislodgement due to balloon deflation but are
lifted, maintaining the stomach in contact with the abdomi- also prone to distal migration with possible intes-
nal wall. Removal of grasper and trocar. Insufflation of the
tinal obstruction. Malecot-type catheters have a
stomach. Insertion of a needle through the trocar site into
the gastric lumen, between the two U stitches. A Seldinger- relatively soft-winged tip, which makes them
type guide wire is passed through the needle into the stom- easier to insert but are more prone to dislodge-
ach. The tract is dilated over the guide wire to the size ment. Pezzer and PEG-type devices are a little
required to insert either a Foley-type catheter or a balloon-
harder to insert but less prone to dislodgement.
type skin-level device. From Gauderer MWL. Gastrostomy.
In: Spitz L, Coran AG, eds. Operative pediatric surgery. 6th
ed. London: Hodder Arnold; 2006. p. 330–355

Fig. 61.12  Placement of a balloon-type “button.” The In: Spitz L, Coran AG, eds. Operative pediatric surgery.
previously placed U stitches are tied over the “wings” of 6th ed. London: Hodder Arnold; 2006. p. 330–355
the skin-level device. From Gauderer MWL. Gastrostomy.
820 J. Brendel and M. W. L. Gauderer

Table 61.2  A comparison of the most commonly used gastrostomy devices. From Puri and Höllwarth, Pediatric
Surgery: Diagnosis and Management, Springer 2009, page 374 (Gauderer 2009)
PEG-type, Pezzer, Malecot, Foley (balloon Skin-level (“button”
t-tube type) type)
Suitable for initial insertion Yes Yes Yes
Suitable for decompression Yes Yes Yes
Tendency for accidental dislodgement or Moderate (with special Moderate Very low (except
external migration adaptor) balloon type)
Tendency for internal (distal) migration Moderate High None
Tendency for peristomal leakage Moderate Moderate Low
(particularly large tubes)
Balloon deflation No Yes Yes, with balloon type
Reinsertion Easy to moderately Easy Easy to moderately
difficult difficult
Long-term (particularly ambulatory Adequate Adequate Best suited
patients)
Overall complication rates Significant Significant Low

Skin-level devices are best suited for long-­ 61.2.4 Complications

term care. The three most commonly employed
skin-level devices are shown in Fig.  61.13. The
original button has the lowest external profile, but
its insertion may be painful as the mushroom
head often does not collapse completely. It may
be difficult to judge the appropriate button length
of a combined PEG catheter button prior to inser-
tion, and once the button is deployed, it cannot be
changed. Modern balloon-type skin-level devices
are usually easy to change but often have a
slightly taller external profile and a shorter lifes-
pan due to balloon deflation. The third type is a
device with a changeable external port valve,
meaning that an originally placed long gastros-
tomy tube can be converted to a skin-level device
by simply cutting the tube to the desired length
above the skin and placing the valve in the shaft.
Because the tract is not disturbed, there is no risk
of gastric separation. The main disadvantage is
late valve failure. If this occurs, the external port
valve can be changed without the removal of the
initially placed catheter. Because the valve can be
applied at any level on the shaft of the tube, this
skin-level device becomes specific for each
patient (Gauderer and Zimmer 2017).
and Management
Although frequently considered a “simple” pro-
cedure, gastrostomies have considerable poten-
tial for early and late morbidity, particularly
among neonates (Gauderer and Zimmer 2017;
Farrelly and Stitelman 2016).
61.2.4.1 Complications Related
to Operative Technique
–– The separation of stomach from the abdomi-
nal wall most frequently occurs shortly after
an early gastrostomy tube replacement before
a firm adhesion between the gastric and
abdominal walls has occurred (but can also
occur at any time thereafter). When attempt-
ing to replace a dislodged catheter, the stom-
ach is pushed away from the abdominal wall,
leading to a partial or complete separation of
the stoma, which can lead to severe peritonitis
and death (Gauderer and Stellato 1986). A
firm anchoring of the stomach to the anterior
abdominal wall can avoid separation. In the
case of an early removal or dislodgement of
the tube, the tract can be gently probed and a
61  Gastrostomy and Jejunostomy 821

Fig. 61.13  The three main gastrostomy device groups. and management. first ed. Berlin, London: Springer;
From Gauderer MWL. Gastrostomy and Jejunostomy. In: 2009. p. 369–381
Puri P, Höllwarth ME, eds. Pediatric surgery: Diagnosis
822
thin Foley catheter inserted. A contrast study
should be performed to assure an intragastric
position of the tube and the absence of intra-
peritoneal leakage. If the position of the cath-
eter remains uncertain, immediate exploration
is necessary (Gauderer and Zimmer 2017).
–– Wound separation, dehiscence, and ventral
hernia usually have a considerable morbid-
ity and mortality. A leakage from an enlarged
incision may be life-threatening (Gauderer
and Stellato 1986). The risk can be mini-
mized by using appropriate, small incisions
and by bringing the tube out through a
counter-incision.
–– Infections can occur with any type of gastros-
tomy (Zimmer and Gauderer 2020; Goldin
et  al. 2016). Although usually limited to the
skin and subcutaneous tissue, the whole
abdominal wall may get affected. Infections
can usually be avoided through the use of pro-
phylactic antibiotic administration and a skin
incision only slightly larger than the diameter
of the tube.
–– Injury to the posterior stomach wall and other
organs. The posterior gastric wall can be dam-
aged or perforated, not only during the initial
procedure but also later during catheter change
(Gauderer and Stellato 1986). To test the
tube’s position and function after insertion,
gastric remains should be withdrawn, and air
or saline should be injected. During open sur-
gery, damage to the liver and spleen can occur
through the improper use of retractors or other
instruments. The distended colon may be mis-
taken for the stomach, particularly in patients
with intra-abdominal adhesions in whom the
mobility of intestinal loops is limited (Zimmer
and Gauderer 2020).
–– Hemorrhage is usually related to inadequate
hemostasis at the time of catheter insertion.
Gentle traction on the catheter can control the
bleeding, but reoperation may become
necessary.
–– Gastrocolic fistula is usually encountered with
the percutaneous endoscopic techniques
(Gauderer and Zimmer 2017; Gauderer and
Stellato 1986) and can be avoided by proper
J. Brendel and M. W. L. Gauderer
gastric insufflation with downward colonic
displacement, transillumination, and the
indentation on the anterior abdominal wall.
Otherwise, overinflation can distort the local
anatomy, including the position of the colon.
Additionally, air-filled small bowel loops will
displace the colon cranially and move it
between the stomach and the abdominal wall
(Zimmer and Gauderer 2020).
61.2.4.2 Complications Related
to Stoma Care
–– Skin irritation and candidiasis are usually
related to leakage and compounded by occlu-
sive dressings. Irritation is best prevented by
avoiding any occlusive devices, including tape
or gauze pads (Gauderer and Stellato 1986).
The gastrostomy site should be kept open and
dry at all times. Ointments and other solu-
tions, except for the treatment of candidiasis,
should be avoided. Long tube catheters can be
immobilized with a small external crossbar
(Gauderer and Zimmer 2017).
–– Tube plugging. Catheters must be flushed with
water after each feeding to prevent blockage.
–– Administration of improper feedings. The
careful and slow administration of the appro-
priate nutrient prevents metabolic abnormali-
ties as well as excessive reflux and diarrhea.
–– Delay and trauma in the reintroduction of a
dislodged catheter. Accidental dislodgement
of gastrostomy catheters is quite common.
Unless the tract is already well-matured and
epithelium-lined, the catheter must be replaced
immediately before the tract closes, which can
happen within a couple of hours. Dislodgement
of the jejunal tube into the stomach may occur
in gastrostomy skin-level devices with jejunal
extension, usually requiring radiologic (fluo-
roscopy)- or endoscopic-guided repositioning.
Improper catheter reintroduction can damage
other organs, especially if long stylets or other
traumatic instruments are used to elongate a
mushroom-type tip. Gentle insertion and aim-
ing toward the gastric cardia or fundus is the
method least likely to cause injury (Zimmer
and Gauderer 2020).
61  Gastrostomy and Jejunostomy
61.2.4.3 Complications Related
to Catheters
–– Granulation tissue is by far the most frequent
problem associated with gastrostomies. In
mild cases, usually a few applications of silver
nitrate are curative. However, if neglected,
granulation tissue will predispose to leakage,
bleeding, and chronic discharge. With exces-
sive growth, excision and cauterization
become necessary. Granulation tissue forma-
tion will cease once epithelialization of the
tract has occurred.
–– Leakage is usually caused by the enlargement
of the stoma trough the pivoting motion of the
gastrostomy tube. Catheters brought through
the incision site or the thinner midline are
more prone to this problem. Severe widening
of the stoma can result in skin excoriation, dis-
lodgment of the tube, metabolic imbalance,
and even death (Gauderer and Stellato 1986).
Smaller and softer catheters help to avoid piv-
oting motion, but in extreme cases, reopera-
tion or stoma relocation becomes necessary.
–– Internal migration is particularly common
with long, balloon-type catheters.
–– External migration is caused by an overzeal-
ous approximation of external immobilizing
devices (bumper), which leads to embedding
of the inner crossbar of the PEG catheter,
mushroom tip, or balloon in the gastric and
abdominal wall. This phenomenon, which can
occur in any gastrostomy type and device, is
also called “buried bumper syndrome” (BBS)
(Gauderer and Stellato 1986; Zimmer and
Gauderer 2020; Abdelhadi et  al. 2016), pre-
senting with lack of motion, leakage, limited
flow, or the abscess formation. The catheter
should be removed and replaced either endo-
scopically or surgically. To avoid BBS, it’s
recommended to rotate the tube regularly,
starting 1–2 weeks after insertion (Volpe and
Malakounides 2018). BBS can be avoided by
giving the catheter enough “play,” i.e., a little
to-and-fro motion.
–– The perforation of esophagus and small bowel
is caused by an accidental inflation of a
balloon-­type catheter in one of these organs
(Gauderer and Zimmer 2017).
823
61.2.5 Gastrostomy Closure
and Persistent
Gastrocutaneous Fistula
Except for endoscopically placed gastrostomies,
the catheter can be simply removed when no lon-
ger needed. If the gastrostomy has been in place
for less than 6–12 months, the tract usually closes
spontaneously. Otherwise, operative closure
under general anesthesia may become necessary,
dissecting the stoma tract down to the deepest
possible extraperitoneal level and closing the
tract in layers using absorbable sutures.
61.3 Jejunostomies
Jejunostomies are useful enterostomas avoiding
gastroesophageal reflux, but they are considered
less physiologic because bolus feeding is not
possible, and they need more careful placement
compared to gastrostomies (Gauderer 2012;
Abdelhadi et  al. 2016; McCann et  al. 2019;
Broekaert et al. 2019).
61.3.1 Indications
Jejunostomies are usually implemented in neo-
nates and small children with complex medical
problems, who are in need of long-term enteral
access and especially postpyloric feedings
(Broekaert et al. 2019). Administration of medi-
cation is a less commonly used. Jejunostomies
are used to overcome problems related to gastro-
esophageal reflux and aspiration in patients with
an upper tracheoesophageal cleft, microgastria,
and in older children with gastric emptying prob-
lems and gastric paresis after severe head trauma,
gastric outlet obstruction. or gastroesophageal
reflux disease in some severe mentally reduced
patients (Vermilyea and Goh 2016; Broekaert
et al. 2019).
A feeding gastrojejunostomy is a special feed-
ing route that extends from the abdominal wall,
through the stomach and into the proximal small
bowel with the added capability of gastric vent-
ing (Simons and Bulat 2021).
824 J. Brendel and M. W. L. Gauderer

61.3.2 Choice of Technique
Various techniques for the insertion of jejunal
feeding tubes have been proposed, such as fine-­
needle catheter jejunostomy, endoscopically, and
radiologically or laparoscopically controlled
placement or percutaneously inserted jejunal
insertion (Gauderer 2012; Pang et al. 2017; Singh
et al. 2018; Fascetti-Leon et al. 2018; Broekaert
et al. 2019). Options of jejunal access for select
feeding and decompressing feeding are illus-
trated in Fig. 61.14.
–– Needle catheter jejunostomy are for short-­
term use via direct jejunal access and can be
inserted as an adjunct during other intra-­
abdominal interventions. However, these cath-
eters clog easily, are nearly impossible to
change, and are associated with several seri-
ous complications.
–– Catheter placement directly into the jejunum is
for long-term usage. The traditional technique
is the formation of a Stamm-type channel. The
placement by small left upper laparotomy
allows the placement of the stoma in the proxi-
mal jejunum. In neonates and small infants, a
classical Stamm gastrostomy is usually not
possible, but the chosen part of the first free
jejunum part after the ligament of Treitz is
fixed to the anterior abdominal wall with three
or four fine sutures around the feeding tube.
One must carefully avoid any kinking of the
jejunum with some additional sutures and to
choose the placement of the stoma in a way
that intestinal loops cannot glide left from the
stoma upward, leading to an ileus or volvulus.
It is important to introduce the feeding tube at
least 15–20 cm downward avoiding any retro-
pulsation of the tube due to the to-and-fro peri-
stalsis of the small bowel. In neonates and
small children, the fixation of the feeding tube
is crucial because if it gets lost, usually a new
tube can only be introduced by a laparotomy.
In older children or adolescents, a typical
Stamm-type stoma can be created without los-
ing too much from the jejunal lumen.
–– Catheter placement in a partially excluded
loop (Roux-en-Y feeding jejunostomy) is also
for long-term use of jejunal access with the
advantages of reduced leakage rates and the
possibility of safe catheter change. However,
this approach is more complex and increases
the possibility of early and late bowel-related
complications (Gauderer 2012).

a b c

d e f

Fig. 61.14  Options of jejunal access for select feeding and
decompressing feeding. (a) Tunneled catheter. (b) Needle
catheter. (c) T-tube. (d) Button. (e) Proximal decompres-
sion and distal feeding across an anastomosis. (f) Tempo-
rary decompression feeding using catheters when primary
anastomosis is unsafe and intestinal exteriorization is not
possible. (g) Roux-en-Y feeding jejunostomy. Gauderer
MWL. Gastrostomy. In: Spitz L, Coran AG, eds. Operative
pediatric surgery. 6th ed. London: Hodder Arnold; 2006.
p. 330–355
61  Gastrostomy and Jejunostomy
61.3.3 Devices
A non-balloon button for older children and a
T-tube for infants are recommended as direct
jejunal access. Balloon catheters should be
avoided because they might occlude the bowel
lumen. Any of the gastric access devices will
work in the Roux-en-Y setup.
The so-called gastrojejunal (GJ) buttons have
the same design as gastrostomy buttons but also
have a long distal jejunal tube component. They
are low-profile/skin-level, have a one-way valve
to prevent leakage when not in use, and are kept
in place by a retention balloon (Abdelhadi et al.
2016; Vermilyea and Goh 2016). The gastric part
of the GJ tube comes in different French diame-
ters to match the patient’s gastrostomy stoma
length. The jejunal length is determined by the
patient’s size (Vermilyea and Goh 2016). The
gastric access is usually used for the administra-
tion of medications, while the jejunal extension is
used for enteral nutrition (Abdelhadi et al. 2016).
61.3.4 Postoperative Care
and Complications
Feeding should be started at a much slower rate
compared to gastrostomies to reduce the risk for
ileus and diarrhea and should be administered
continuously by a pump (Broekaert et al. 2019).
Feeding can be safely started a few hours after
the insertion, but some centers wait for the first
postoperative day (Broekaert et  al. 2019).
Following a period of adaptation, the regimen
can be modified to allow for feeding pauses.
Some patients may even tolerate a limited amount
of bolus feeds. There is no valid data on the
required frequency of jejunal feeding tube
changes (Simons and Bulat 2021). GJ tubes,
however, require frequent exchanges, usually
every 3 months (Vermilyea and Goh 2016).
Overall, jejunostomies have significantly
more complications than gastrostomies. Adhesive
bowel obstruction, dumping syndrome-­associated
diarrhea, intestinal volvulus around the stoma,
and internal hernia are specifically jejunostomy-­
related complications (Abdelhadi et  al. 2016;
825
Farrelly and Stitelman 2016; Zimmer and
Gauderer 2020; Singh et al. 2018; Fascetti-Leon
et al. 2018; Broekaert et al. 2019; McCann et al.
2019). GJ buttons that are too big in size may
cause pyloric obstruction in smaller children
(Vermilyea and Goh 2016).
Morbidity is related to tube dislodgement,
persistent reflux, and intestinal perforation
(Campwala et  al. 2015; Farrelly and Stitelman
2016; Demehri et al. 2016; McCann et al. 2019).
Due to the length of the jejunostomy tube, it
is unlikely for the tube to be completely removed
from the abdominal wall by tugging, but this
may occur (Simons and Bulat 2021). When a
mature tract already exists (a minimum of
4 weeks after initial tube placement), an incom-
pletely dislodged tube can be taped in place until
the tube can be repositioned or replaced radio-
logically or endoscopically (Simons and Bulat
2021). If a tube completely falls out of an already
mature tract, a place holder (e.g., Foley catheter)
should be inserted within 24  hrs to prevent the
closure of the stoma (Simons and Bulat 2021).
Because tube change/position control always
needs to be checked with fluoroscopy imaging or
conventional X-ray, some patients experience a
considerable exposure to radiation over time
(Fascetti-Leon et al. 2018).
In the case of an immature tract, a catheter
should not be reinserted due to the risk of tube
penetration, leakage, and infection into the intra-­
abdominal cavity (Simons and Bulat 2021). If the
tract is too large for the catheter, leakage may
occur, leading to luminal fluid and feed loss and
later tissue breakdown, increasing the risk of
infection (Simons and Bulat 2021).
61.4 Conclusion
Although both gastrostomy and jejunostomy are
basic surgical procedures, the medical team, par-
ents, and caregivers must carefully consider the
advantages and disadvantages of each method,
keeping the patient’s comorbidities in mind as
these procedures mean major intervention in the
children’s life. A problematic stoma can compli-
cate the management of even a simple, temporary
826
condition. Patients with enterostomies benefit
from an interdisciplinary panel consisting of a
pediatrician, pediatric surgeon, nutritionist, and
primary nurse, not only for indication but also for
a regular follow-up to assure optimal functional-
ity and minimal morbidity.
References
Abdelhadi RA, Rahe K, Lyman B (2016) Pediatric enteral
access device management. Nutr Clin Pract 31(6):748–
761. https://doi.org/10.1177/0884533616670640
Akay B, Capizzani TR, Lee AM, Drongowski RA, Geiger
JD, Hirschl RB, Mychaliska GB (2010) Gastrostomy
tube placement in infants and children: is there a pre-
ferred technique? J Pediatr Surg 45(6):1147–1152.
https://doi.org/10.1016/j.jpedsurg.2010.02.079
Antonoff MB, Hess DJ, Saltzman DA, Acton RD (2009)
Modified approach to laparoscopic gastrostomy
tube placement minimizes complications. Pediatr
Surg Int 25(4):349–353. https://doi.org/10.1007/
s00383-­009-­2340-­z
Backman T, Sjövie H, Kullendorff C-M, Arnbjörnsson
E (2010) Continuous double U-stitch gastrostomy in
children. Eur J Pediatr Surg 20(1):14–17. https://doi.
org/10.1055/s-­0029-­1238316
Baker L, Emil S, Baird R (2013) A comparison of tech-
niques for laparoscopic gastrostomy placement in
children. J Surg Res 184(1):392–396. https://doi.
org/10.1016/j.jss.2013.05.067
Baker L, Beres AL, Baird R (2015) A systematic review
and meta-analysis of gastrostomy insertion techniques
in children. J Pediatr Surg 50(5):718–725. https://doi.
org/10.1016/j.jpedsurg.2015.02.021
Broekaert IJ, Falconer J, Bronsky J, Gottrand F,
Dall’Oglio L, Goto E et  al (2019) The use of jeju-
nal tube feeding in children: a position paper by the
Gastroenterology and Nutrition Committees of the
European Society for Paediatric Gastroenterology,
Hepatology, and Nutrition 2019. J Pediatr
Gastroenterol Nutr 69(2):239–258. https://doi.
org/10.1097/MPG.0000000000002379
Campwala I, Perrone E, Yanni G, Shah M, Gollin G
(2015) Complications of gastrojejunal feeding tubes
in children. J Surg Res 199(1):67–71. https://doi.
org/10.1016/j.jss.2015.06.058
Chatwin M, Bush A, Macrae DJ, Clarke SA, Simonds AK
(2013) Risk management protocol for gastrostomy
and jejunostomy insertion in ventilator dependent
infants. Neuromuscul Disord 23(4):289–297. https://
doi.org/10.1016/j.nmd.2013.01.006
Corkins MR, Fitzgerald JF, Gupta SK (2010) Feeding after
percutaneous endoscopic gastrostomy in children: early
feeding trial. J Pediatr Gastroenterol Nutr 50(6):625–
627. https://doi.org/10.1097/MPG.0b013e3181bab33d
J. Brendel and M. W. L. Gauderer
Cortez AR, Warren PW, Goddard GR, Jenkins TM, Sauser
JA, Gerrein BT, Rymeski BA (2020) Primary place-
ment of a low-profile gastrostomy button is safe and
associated with improved outcomes in children. J
Surg Res 249:156–162. https://doi.org/10.1016/j.
jss.2019.11.027
Demehri FR, Simha S, Herrman E, Jarboe MD, Geiger
JD, Teitelbaum DH, Gadepalli SK (2016) Analysis
of risk factors contributing to morbidity from gas-
trojejunostomy feeding tubes in children. J Pediatr
Surg 51(6):1005–1009. https://doi.org/10.1016/j.
jpedsurg.2016.02.072
Farrelly JS, Stitelman DH (2016) Complications in
pediatric enteral and vascular access. Semin Pediatr
Surg 25(6):371–379. https://doi.org/10.1053/j.
sempedsurg.2016.10.006
Fascetti-Leon F, El Agami H, Gobbi D, Clarke S, Haddad
M, Choudhry M (2018) Feeding jejunostomy: is it
a safe route in pediatric patients? Single institution
experience. Eur J Pediatr Surg 28(3):293–296. https://
doi.org/10.1055/s-­0037-­1603526
Fujiogi M, Tanaka Y, Amano H, Deie K, Suzuki K,
Kawashima H et al (2018) A new laparoscopic triangle
fixation technique for gastrostomy: a safe and effec-
tive procedure for reduction of the wound infection
rate. Nagoya J Med Sci 80(4):497–503. https://doi.
org/10.18999/nagjms.80.4.497
Gauderer MWL (2006) Gastrostomy. In: Spitz L, Coran
AG (eds) Operative pediatric surgery, 6th edn. Hodder
Arnold, London, pp 330–355
Gauderer MWL (2009) Gastrostomy and jejunostomy.
In: Puri P, Höllwarth ME (eds) Pediatric surgery:
diagnosis and management, 1st edn. Springer, Berlin,
pp 369–381
Gauderer MWL (2012) Stomas of the small and large
intestine. In: Coran AG, Adzick NS, Krummel TM,
Laberge J-M, Shamberger RC, Caldamone AA (eds)
Pediatric surgery. Elsevier Saunders
Gauderer MWL, Stellato TA (1986) Gastrostomies:
evolution, techniques, indications, and complica-
tions. Curr Probl Surg 23(9):661–719. https://doi.
org/10.1016/0011-­3840(86)90020-­1
Gauderer MWL, Zimmer J (2017) Access for enteral
nutrition. In: Puri P (ed) Newborn surgery, 4th edn.
CRC Press, Boca Raton
Gauderer MWL, Ponsky JL, Izant RJ (1980) Gastrostomy
without laparotomy: a percutaneous endoscopic
technique. J Pediatr Surg 15(6):872–875. https://doi.
org/10.1016/S0022-­3468(80)80296-­X
Goldin AB, Heiss KF, Hall M, Rothstein DH, Minneci PC,
Blakely ML et al (2016) Emergency department visits
and readmissions among children after gastrostomy
tube placement. J Pediatr 174:139–145.e2. https://doi.
org/10.1016/j.jpeds.2016.03.032
Hassan SF, Pimpalwar AP (2011) Modified laparoendo-
scopic gastrostomy tube (LEGT) placement. Pediatr
Surg Int 27(11):1249–1254. https://doi.org/10.1007/
s00383-­011-­2977-­2
61  Gastrostomy and Jejunostomy
Hendrickson RJ, Poola AS, Sujka JA, Weaver KL,
Rentea RM, St Peter SD, Oyetunji TA (2018) Feeding
advancement and simultaneous transition to dis-
charge (FASTDischarge) after laparoscopic gastros-
tomy. J Pediatr Surg 53(11):2326–2330. https://doi.
org/10.1016/j.jpedsurg.2018.04.035
Heuschkel RB, Gottrand F, Devarajan K, Poole H, Callan
J, Dias JA et al (2015) ESPGHAN position paper on
management of percutaneous endoscopic gastrostomy
in children and adolescents. J Pediatr Gastroenterol
Nutr 60(1):131–141. https://doi.org/10.1097/
MPG.0000000000000501
Jensen AR, Renaud E, Drucker NA, Staszak J, Senay A,
Umesh V et  al (2018) Why wait: early enteral feed-
ing after pediatric gastrostomy tube placement. J
Pediatr Surg 53(4):656–660. https://doi.org/10.1016/j.
jpedsurg.2017.06.015
Kaya M, Sancar S, Ozcakir E (2018) A new method for
laparoscopic stamm gastrostomy. J Laparoendosc Adv
Surg Tech A 28(1):111–115. https://doi.org/10.1089/
lap.2017.0447
Landisch RM, Colwell RC, Densmore JC (2016)
Infant gastrostomy outcomes: the cost of complica-
tions. J Pediatr Surg 51(12):1976–1982. https://doi.
org/10.1016/j.jpedsurg.2016.09.025
Liu R, Jiwane A, Varjavandi A, Kennedy A, Henry G,
Dilley A et  al (2013) Comparison of percutaneous
endoscopic, laparoscopic and open gastrostomy inser-
tion in children. Pediatr Surg Int 29(6):613–621.
https://doi.org/10.1007/s00383-­013-­3313-­9
Livingston MH, Pepe D, Jones S, Bütter A, Merritt NH
(2015) Laparoscopic-assisted percutaneous endo-
scopic gastrostomy: insertion of a skin-level device
using a tear-away sheath. Can J Surg 58(4):264–268.
https://doi.org/10.1503/cjs.014814
Lotti M, Carrara G, Lovece A, Giulii CM (2020)
Laparoscopic tubularized continent gastros-
tomy: an alternative to tube gastrostomies. Updat
Surg 72(3):901–905. https://doi.org/10.1007/
s13304-­020-­00795-­6
Mason CA, Skarda DE, Bucher BT (2018) Outcomes after
laparoscopic gastrostomy suture techniques in chil-
dren. J Surg Res 232:26–32. https://doi.org/10.1016/j.
jss.2018.05.022
McCann C, Cullis PS, McCabe AJ, Munro FD (2019)
Major complications of jejunal feeding in children. J
Pediatr Surg 54(2):258–262. https://doi.org/10.1016/j.
jpedsurg.2018.10.078
Merli L, Marco EA d, Fedele C, Mason EJ, Taddei A,
Paradiso FV et  al (2016) Gastrostomy placement in
children: percutaneous endoscopic gastrostomy or
laparoscopic gastrostomy? Surg Laparosc Endosc
Percutan Tech 26(5):381–384. https://doi.org/10.1097/
SLE.0000000000000310
Novotny NM, Vegeler RC, Breckler FD, Rescorla
FJ (2009) Percutaneous endoscopic gastrostomy
buttons in children: superior to tubes. J Pediatr
Surg 44(6):1193–1196. https://doi.org/10.1016/j.
jpedsurg.2009.02.024
827
Pang T, Sesia SB, Holland-Cunz S, Mayr J (2017)
Replaceable jejunal feeding tubes in severely ill chil-
dren. Gastroenterol Res Pract 2017:2090795. https://
doi.org/10.1155/2017/2090795
Petrosyan M, Khalafallah AM, Franklin AL, Doan T,
Kane TD (2016) Laparoscopic gastrostomy is supe-
rior to percutaneous endoscopic gastrostomy tube
placement in children less than 5 years of age. J
Laparoendosc Adv Surg Tech A 26(7):570–573.
https://doi.org/10.1089/lap.2016.0099
Sandberg F, Viktorsdóttir MB, Salö M, Stenström P,
Arnbjörnsson E (2018) Comparison of major com-
plications in children after laparoscopy-assisted
gastrostomy and percutaneous endoscopic gas-
trostomy placement: a meta-analysis. Pediatr Surg
Int 34(12):1321–1327. https://doi.org/10.1007/
s00383-­018-­4358-­6
Sayadi Shahraki M, Berjis N, Bighamian A, Mahmoudieh
M, Shahabi Shahmiri S, Sheikhbahaei E (2020)
Minimally invasive technique for gastrostomy tube
insertion: a novel laparoscopic approach. Asian J
Endosc Surg 13(4):610–613. https://doi.org/10.1111/
ases.12780
Simons M, Bulat R (2021) Chapter 29  – Enteric tube
placement for gastroparesis: gastrostomy, gas-
trojejunostomy and jejunostomy. In: RW MC,
Parkman HP (eds) Gastroparesis. Academic
Press, pp  397–411. https://doi.org/10.1016/
B978-­0-­12-­818586-­5.00029-­6
Singh RR, Eaton S, Roebuck DJ, Barnacle AM,
Chippington S, Cross KMK et al (2018) Surgical jeju-
nostomy and radiological gastro-jejunostomy tube
feeding in children: risks, benefits and nutritional out-
comes. Pediatr Surg Int 34(9):951–956. https://doi.
org/10.1007/s00383-­018-­4303-­8
Smitherman S, Pimpalwar A (2009) Laparoendoscopic
gastrostomy tube placement: our all-in-one technique.
J Laparoendosc Adv Surg Tech A 19(1):119–123.
https://doi.org/10.1089/lap.2007.0210
Sulkowski JP, Roo AC d, Nielsen J, Ambeba E, Cooper
JN, Hogan MJ et  al (2016) A comparison of pediat-
ric gastrostomy tube placement techniques. Pediatr
Surg Int 32(3):269–275. https://doi.org/10.1007/
s00383-­015-­3847-­0
Sutherland C, Carr B, Biddle KZ, Jarboe M, Gadepalli SK
(2017) Pediatric gastrostomy tubes and techniques:
making safer and cleaner choices. J Surg Res 220:88–
93. https://doi.org/10.1016/j.jss.2017.06.075
Thaker S, Misra MV (2019) A comparison of laparoscopic
gastrostomy tubes and percutaneous ­ endoscopic
gastrostomy in infants: results from a single institu-
tion. J Pediatr Surg 54(11):2453–2456. https://doi.
org/10.1016/j.jpedsurg.2019.06.024
Vasseur Maurer S, Reinberg O (2015) Laparoscopic tech-
nique to perform a true Stamm gastrostomy in chil-
dren. J Pediatr Surg 50(10):1797–1800. https://doi.
org/10.1016/j.jpedsurg.2015.06.010
Vermilyea S, Goh VL (2016) Enteral feedings in
children: sorting out tubes, buttons, and formu-
828
las. Nutr Clin Pract 31(1):59–67. https://doi.
org/10.1177/0884533615604806
Villalona GA, Mckee MA, Diefenbach KA (2011)
Modified laparoscopic gastrostomy technique reduces
gastrostomy tract dehiscence. J Laparoendosc Adv
Surg Tech A 21(4):355–359. https://doi.org/10.1089/
lap.2010.0201
Volpe A, Malakounides G (2018) Feeding tubes in chil-
dren. Curr Opin Pediatr 30(5):665–670. https://doi.
org/10.1097/MOP.0000000000000666
Watanabe J, Kotani K (2020) Early versus delayed feed-
ing after percutaneous endoscopic gastrostomy place-
ment in children: a meta-analysis. Children 7:124.
https://doi.org/10.3390/children7090124
Wiernicka A, Matuszczyk M, Szlagatys-Sidorkiewicz
A, Landowski P, Toporowska-Kowalska E, Gębora-­
Kowalska B et  al (2019) Tolerability and safety of
J. Brendel and M. W. L. Gauderer
early enteral nutrition in children after percutaneous
endoscopic gastrostomy placement: a multicentre ran-
domised controlled trial. Clin Nutr 38(4):1544–1548.
https://doi.org/10.1016/j.clnu.2018.08.018
Williams K, Baumann L, Abdullah F, Hendrickson RJ,
Oyetunji TA (2019) Elective laparoscopic gastros-
tomy in children: potential for an enhanced recovery
protocol. Pediatr Surg Int 35(6):643–647. https://doi.
org/10.1007/s00383-­019-­04472-­7
Wilson L, Oliva-Hemker M (2001) Percutaneous endo-
scopic gastrostomy in small medically complex
infants. Endoscopy 33(5):433–436. https://doi.
org/10.1055/s-­2001-­14268
Zimmer J, Gauderer MWL (2020) Access for enteral
nutrition. In: Puri P (ed) Encyclopedia of pedi-
atric surgery. Springer, Berlin. https://doi.
org/10.1007/978-­3-­642-­38482-­0_19-­1
 Duodenal Obstruction
62
Yechiel Sweed and Arcady Vachyan

62.1 Introduction

Congenital DO is one of the most common

anomalies in newborns and accounts for half of
all cases of neonatal intestinal obstruction
(Haeusler et al. 2002). DO occurs in 1 per 5000–
10,000 live births and affects boys more com-
monly than girls (Best et al. 2012; Rattan et al.
2016). DO is classified as an intrinsic lesion,
extrinsic lesion, or both, and it may be either
complete or partial (Dalla Vecchia et  al. 1998;
Escobar et al. 2004). Intrinsic DO may be caused
by atresia and stenosis in the diaphragm, a perfo- Fig. 62.1  DO caused by an annular pancreas, associated
rated diaphragm, or a “wind-sock web.” The with duodenal stenosis, in a postmortem of a 14-week-old
fetus with a diagnosis of Down syndrome. ST stomach,
“wind-sock web” is a duodenal membrane,
DU duodenum, AP annular pancreas. (The picture added
which is ballooned distally as a result of peristal- with the courtesy of Prof. Bronstein Moshe)
sis from above (Norton et  al. 1992). Extrinsic
DO may be caused by annular pancreas, malro-
tation, or preduodenal portal vein (Choudhry in patients with annular pancreas (Grosfeld et al.
et  al. 2009). Although the annular pancreas 1993; Papandreou et al. 2004). Similarly, predu-
forms a constricting ring around the second part odenal portal vein has also seldom been reported
of the duodenum (Fig. 62.1), it is not believed to to be the cause of DO and is often associated
be the cause of DO (Escobar et  al. 2004), and with other causes of intestinal obstruction such
there is usually an associated atresia or stenosis as malrotation or duodenal atresia (Srivastava
et al. 2016; Singal et al. 2009).
Y. Sweed (*)
Duodenal atresias are classified into three
Department of Pediatric Surgery, Galilee Medical types (Fig.  62.2) (Skandalakis and Gray 1994).
Center, Naharia, Israel Type 1 defect, the most common (65%), is char-
Bar Ilan University, Safed, Israel acterized by a mucosal web with normal muscu-
A. Vachyan
lar wall (Fig. 62.2a), type 2 is characterized by a
Pediatric Advanced Laparoscopic Surgical Unit, Ruth short fibrous cord connecting the two atretic ends
Rappaport Children’s Hospital, Rambam Medical of the duodenum (18%) (Fig. 62.2b), and type 3
Center, Haifa, Israel
e-mail: a_vachyan@rambam.health.gov.il
© Springer Nature Switzerland AG 2023 829
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_62
830
Fig. 62.2  Types of duodenal atresia (by Skandalakis and
Gray). (a) Type 1 (a mucosal membrane with intact mus-
cle wall), (b) Type 2 (fibrous cord connecting the two
a b
e f
Fig. 62.3 Various types of duodenal obstruction. (a)
Blind ends separated by a gap, (b) two ends in apposition,
(c) ends joined by a fibrous cord, (d) duodenal stenosis,
is characterized by a complete separation of the
atretic ends with a mesenteric defects (18%)
(Fig. 62.2c). Duodenal stenosis is approximately
half as prevalent as atresia (Dalla Vecchia et al.
1998).
The obstruction of the duodenum usually
occurs distal to the ampulla of Vater. Pre-­
ampullary obstruction is much less common,
Y. Sweed and A. Vachyan
atretic ends of the duodenum), (c) Type 3 (a complete
separation of the atretic ends with a mesenteric defect)
c d
g h
(e) complete duodenal membrane, (f) perforated dia-
phragm, (g) “wind-sock” web, and (h) annular pancreas
occurring in about 20% of cases (Knechtle and
Filston 1990). Occasionally, there may be a bifid
termination of the bile duct, with one limb of the
duct system opening into the duodenum above
the atresia and one below (Komuro et al. 2011).
Figure 62.3 shows the wide spectrum of the
various types of DO.  The proximal and distal
segments of the duodenum may be separated by
62  Duodenal Obstruction
a gap (Fig. 62.3a), be in apposition (Fig. 62.3b)
or be joined by a fibrous cord (Fig.  62.3c).
Other types include duodenal stenosis
(Fig. 62.3d), complete diaphragm (Fig. 62.3e),
perforated diaphragm (Fig.  62.3f), a “wind-
sock web” (Fig.  62.3g), and an annular pan-
creas (Fig. 62.3h).
62.2 Etiopathogenesis
and Pathophysiology
The underlying cause of duodenal atresia remains
unknown although its pathophysiology has been
well described. The frequent association of duo-
denal atresia or stenosis with other neonatal mal-
formations suggests that both anomalies are due
to a developmental error in the early period of
gestation. Duodenal atresia, web, and stenosis
usually occur in the second part of the duode-
num, close to the area of intense embryological
activity involved in the development of the bili-
ary and pancreatic structures. Causative theories
of congenital DO include abnormalities of pan-
creatic development and a failure of recanaliza-
tion of the duodenal lumen from its solid cord
stage (Tandler 1900). The latter theory is cur-
rently the most popular. It has been demonstrated
that from the 5th–10th week of gestation, the
endodermal lining of the duodenum proliferates
rapidly and obliterates the lumen, and the duode-
num becomes a solid cord. Later, the duodenal
lumen is recanalized and is formed by vacuoles
that coalesce. Duodenal atresia results from the
failure of vacuolization and recanalization.
Duodenal stenosis and web result from incom-
plete recanalization of the duodenum. An annular
pancreas results when the anterior and posterior
anlage of the pancreas becomes fused to form a
ring of pancreatic tissue that surrounds the sec-
ond part of the duodenum (Figs. 62.1 and 62.3h).
Duodenal atresia differs from other atresias of
the small and large bowel, which are isolated
anomalies caused by mesenteric vascular acci-
dents during the later stages of bowel develop-
ment. This theory of vascular disturbance was
presented by the classic study of Louw and
Barnard (1955).
831
No predisposing maternal risk factors are
known. Although up to one third of patients with
duodenal obstruction have Down syndrome (tri-
somy 21) (Bethell et  al. 2020a, b), it is not an
independent risk factor for developing duodenal
atresia. In a large California population-based
registry of 2.5 million infants, the risk of duode-
nal atresia was found to be 265 times higher in
infants with Down syndrome, compared to those
without it, and the corresponding frequencies
were 46 and 0.12 per 1000 births (Torfs and
Christianson 1998). The association between
duodenal atresia and Down syndrome suggests
an underlying genetic etiology. In mice, the inter-
ruption of fibroblast growth factor 10 (FGF10)
gene signaling results in duodenal atresia in
30–50% of embryos, supporting a genetic etiol-
ogy (Teague et al. 2018).
Although DO is usually not regarded as a
familial condition, there have been several reports
of familial cases (Gross et  al. 1996; Markljung
et al. 2012) and a very rare group of hereditary
multiple intestinal atresias with fatal outcome
(Lambrecht and Kluth 1998). Markljung et  al.
reported on a new familial case of annular pan-
creas and found one microduplication on chro-
mosome 6q24 by array-based comparative
genomic hybridization (CGH) shared by the
affected mother and son (Markljung et al. 2012).
This microduplication may be a causative aberra-
tion or present a risk factor for the development
of annular pancreas and duodenal atresia.
62.3 Associated Malformations
There is a high incidence (approximately 50%)
of associated congenital anomalies in patients
with DO (Chen et  al. 2014), especially Down
syndrome, which occurs in about 30% of these
infants Bethell et al. (2020a, b). recently reported
on 102 patients with DO in the United Kingdom,
who were prospectively collected from 28 spe-
cialist neonatal surgical centers over a 12-month
period from March 2016. They found that associ-
ated congenital anomalies were present in 93%
(31/33) of infants with T21 and 50% (32/64) of
infants without a chromosomal anomaly. They
832
Table 62.1 The incidence of associated congenital
anomalies (%) (collected statistics) (N = 1759 patients)
Associated anomaly
Down syndrome
Annular pancreas
Congenital heart disease
Malrotation
Esophageal atresia and tracheoesophageal fistula
Genitourinary
Anorectal
Other bowel atresia
Others
also found that trisomy 21 was present in 33
babies (32%) and that cardiac anomalies were
more common in those with Down Syndrome,
compared to those without a chromosomal anom-
aly (91% vs 17%, p < 0.001).
Table 62.1 presents the overall prevalence and
distribution of associated anomalies of duodenal
atresia. The data are the collected statistics of
1759 patients with DO from a dozen large series
(Sweed and Yulevich 2020). The associated
anomalies in order of frequency are Down syn-
drome (28.2%), annular pancreas (23.1%), con-
genital heart disease (22.6%), malrotation (20%),
esophageal atresia (8.5%), genitourinary malfor-
mations (8%), anorectal anomalies (4.4%), and
other bowel atresia (3.5%).
Other rare anomalies include vertebral anom-
alies (Bailey et al. 1993), Cornelia de Lange syn-
drome (Bailey et  al. 1993), chromosomal
abnormalities (Bethell et al. 2020a, b), multiple
intestinal abnormalities (Morikawa et  al. 2009),
choledchal cyst (Iwai et al. 2009), immunodefi-
ciency (Moore et  al. 1996), tracheomalacia
(Kimble et al. 1997), and situs inversus (Nawaz
et al. 2005).
Among all the associated malformations, the
complex cardiac anomalies are the major cause
of morbidity and mortality in patients with duo-
denal atresia (Choudhry et al. 2009; Dalla Vecchia
et al. 1998; Escobar et al. 2004; Piper et al. 2008).
Dalla Vecchia et  al. attributed all the operative
mortality (4%) to associated complex congenital
heart anomalies in a group of 138 patients with
DO in a 25-year survey. Two other important fac-
tors affecting higher morbidity and mortality of
this patients are low birth weight and prematurity
Y. Sweed and A. Vachyan
(Piper et  al. 2008). The mortality rate is even
higher in neonates born with three or more anom-
% alies of the VACTERL association, with an over-
28.2 all survival rate of 40–77% (Iuchtman et al. 1992;
23.1 Muraji, 1984). The VACTERL association is an
22.6
acronym describing the occurrence of specific
19.7
8.5 congenital anomalies that occur in a greater than
8.0 random frequency without hereditary factors.
4.4 The anomalies include vertebral or vascular
3.5 defects (V), anal malformation (A), cardiac
10.9 anomalies (C), tracheoesophageal fistula with
esophageal atresia (TE), renal anomalies (R), and
radial limb anomalies (R). Duodenal atresia,
which is not included in the VACTERL associa-
tion, is seen in 6.3% of babies born with this
spectrum of anomalies.
62.4 Prenatal Diagnosis
There has been an increase in prenatal ultrasono-
graphic diagnosis of duodenal atresia over the
last three decades (Murshed et al. 1999; Gfroerer
et al. 2019). Maternal polyhydramnios has been
reported to be present in 30–75% of cases of duo-
denal atresia (Rattan et al. 2016; Cohen-Overbeek
et al. 2008) and is the most common ultrasono-
graphic finding in fetuses with intrinsic
DO.  Ultrasound is usually performed for sus-
pected fetal or maternal abnormalities when
polyhydramnios or a large-for-date pregnancy is
established. Although the majority of cases are
diagnosed during the seventh or eighth month of
gestation (Yin et al. 2020), sonographic detection
of duodenal atresia was reported as early as 12
gestational weeks, by Tsukerman et al. 1993.
The prenatal sonographic diagnosis relies on
the demonstration of the “double bubble” sign,
which is due to the simultaneous distention of the
stomach and the first part of the duodenum
(Fig. 62.4). In many cases, this sonographic sign
is observed in the second half of pregnancy prob-
ably due to the hydrostatic pressure needed to
dilate the duodenum and the stomach and also to
the degree of the DO.
The visualization of a fluid-filled “double bub-
ble” (Fig.  62.4) on prenatal ultrasound scan is
associated with DO secondary to intrinsic or
62  Duodenal Obstruction 833

Bethell et al. reported recently in a prospective

study of 102 patients that 59% of cases were
diagnosed antenatally, and associated anomalies
were seen in 69% of patients; however, the out-
come of infants with DO was not changed by pro-
viding an antenatal diagnosis. They concluded
that the antenatal diagnosis of DO influenced
parents positively in coping with the anomaly,
because it allowed them time to prepare for the
medical and surgical intervention required after
the birth of their infant (Bethell et al. 2020a, b).
Cohen-Overbeek et al. have also reported on 91
cases diagnosed with isolated or non-isolated
DO.  They found that the outcome of prenatally
and postnatally diagnosed DO is not essentially
different, despite the fact that more prematurity
and a lower birth weight were observed in the
Fig. 62.4  Ultrasonography (transverse view) of a 24-week former (Cohen-Overbeek et al. 2008). Early pre-
gestational age fetus showing the “double bubble” sign. ST natal diagnosis should lead to chromosomal anal-
stomach, P pylorus, DU duodenum, g.b gallbladder. (The ysis by amniocentesis for prenatal screening of
picture added with the courtesy of Prof. Bronstein Moshe)
trisomy 21 and other associated anomalies
(Grosfeld and Rescorla 1993; Keckler et  al.
extrinsic lesion or a combined lesions. This 2008). The prenatal diagnosis allows the mother
­sonographic finding is known to have a low false-­ the opportunity to receive counselling by pediat-
positive rate. Zimmer and Bronstein have ric surgeon and to consider delivery at or near a
reported that in a few cases, it may represent a tertiary facility that is able to care for infants with
transient finding in an otherwise healthy fetus gastrointestinal anomalies (Haeusler et al. 2002).
(Zimmer and Bronstein 1996). On the ultrasound
examination, it is also important to demonstrate
the continuity between the gastric and duodenal 62.5 Clinical Presentation
bubbles (Fig. 62.4) to exclude other causes, such and Diagnosis
as choledochal cyst, which lacks such communi-
cation (Casaccia et al. 2002) or duodenal duplica- The diagnosis of DO is usually based on prenatal
tion (Malone et al. 1997). In all cases of combined ultrasonography, early clinical symptoms, physi-
polyhydramnios and “double bubble” sign, a cal examination, and plan radiographic abdomi-
detailed evaluation for other associated anoma- nal films (Bethell et  al. 2020a, b; Piper et  al.
lies, especially cardiac anomalies, should be 2008). About half of the patients are premature
undertaken (Bethell et  al. 2020a, b; Pameijer and of low birth weight. Vomiting and intoler-
et al. 2000; Yin et al. 2020). Other anomalies that ance of attempted feedings are the most com-
can be diagnosed by prenatal ultrasonography of mon symptoms and are usually present on the
the fetus include annular pancreas, with the typi- first day of life. Since 80% of the obstructions
cal ultrasonographic feature of “Pliers” and the are located in the post ampullary region of the
duodenal diaphragm based on the appearance of duodenum, in the majority of cases, vomitus is
the “rat tail” sign (Yin et  al. 2020). Ultrasound bile-stained. In supra-ampullary atresia, it is non-
examination can also show a fetus with a com- bilious. Orogastric aspiration also yields signifi-
bined duodenal and esophageal atresia associated cant volumes of bile-stained gastric fluid. There
with VACTERL anomalies (Pameijer et al. 2000). is minimal or no abdominal distention because of
834
the high level of obstruction. Infants may pass
some meconium in the first 24  h of life, and
thereafter, constipation may develop. Dehydration
with weight loss and electrolyte imbalance
(hypokalemic/hypochloremic metabolic alkalo-
sis) soon follows if the diagnosis is made late and
if fluid and electrolyte losses have not been ade-
quately replaced (Kilbride et al. 2010). Kilbride
et  al. reported that one out of eight infants
(12.5%) with a delayed diagnosis of DO died as a
result of the metabolic and pulmonary distur-
bances related to the proximal bowel obstruction
and concluded that these disturbances required
an early intravenous fluid correction before
surgery.
Incomplete DO usually leads to the delayed
onset of symptoms. Infants with duodenal steno-
sis, or a duodenal membrane with a central aper-
ture, may escape detection of an abnormality
soon after birth and may proceed into childhood
or early into adulthood before a partial obstruc-
tion is noted (Mikaelsson et al. 1997). Mikaelsson
et al. reported on the late diagnosis and treatment
of 8 out of 16 patients with membranous duode-
nal stenosis and found that these patients were
diagnosed and operated on at the age of 1 month
to 4 years old.
a b
Fig. 62.5 (a) Abdominal radiograph showing grossly
distended stomach and duodenum with “double bubble”
sign, with no air beyond the duodenum. GB gastric bub-
ble, DB duodenal bubble. (b) Abdominal radiograph
showing the “double bubble” sign. In this case, duodenal
Y. Sweed and A. Vachyan
The diagnosis of DO is confirmed on X-ray
examination (Fig.  62.5a, b). An abdominal
radiograph will show a dilated stomach and
duodenum, giving the characteristic appearance
of a double bubble sign (the stomach and the
proximal duodenum are air-filled), with no gas
beyond the duodenum. In partial DO, a plain
film of the abdomen will show a double bubble
appearance, but there is usually some air in the
distal intestine (Fig.  62.6). Occasionally in
cases of duodenal atresia, air may be seen distal
to the site of obstruction due to the associated
bile duct bifurcation (Knechtle and Filston
1990). Radiographic findings in patients with
annular pancreas are usually indistinguishable
from duodenal atresia or stenosis.
In some cases of partial DO, plain films may
be normal, and the upper gastrointestinal tract
contrast radiography is indicated in these patients
to establish the cause of incomplete DO.  This
may show a stenotic segment of duodenum with
dilatation of the proximal segment or a sharp ter-
mination of the dilated segment, indicating a per-
forated diaphragm (Fig.  62.7). Occasionally, a
duodenal diaphragm may be stretched and bal-
looned distally, giving the wind-sock appearance
on the contrast study (Fig. 62.8).
c
bulb is more prominent than the gastric bulb. At operation,
the duodenal membrane was found and excised. GB gas-
tric bubble, DB duodenal bubble. (c) Duodenal atresia
evident on the upper gastrointestinal radiograph contrast
study. S stomach, D duodenum
62  Duodenal Obstruction 835

Fig. 62.6  Duodenal stenosis erect abdominal X-ray dem-

onstrating a “double bubble” sign with air beyond the
duodenum

The wide variety of additional congenital

anomalies with special emphasis on cardiac
­malformation, often severe (Piper et  al. 2008),
makes preoperative diagnosis imperative.
Anterior-posterior and lateral chest and
abdominal radiographs ascertaining visualization
of the entire spine should also be performed.
Soon after the X-ray, cardiac and renal ultra-
sound should be carried out routinely in all these
babies. A micturating cystourethrogram should
be performed in those babies with abnormal uro-
genital ultrasound or an associated anorectal
Fig. 62.7  An abdominal X-ray contrast study showing
marked distention of the duodenum, terminating abruptly
with narrow caliber distally. A perforated diaphragm was
found at operation
anomaly. Rectal biopsy should be taken in babies
with constipation, and a combination of Down
syndrome and duodenal atresia, to exclude
Hirschsprung’s disease.
836 Y. Sweed and A. Vachyan

62.6 Differential Diagnosis

All the various conditions of upper gastrointestinal

obstruction should be included in the differential
diagnosis of DO (Table 62.2). The most important
ones for the pediatric surgeons are the DO caused
by malrotation, resulting in extrinsic compression
related to Ladd’s bands across the duodenum and
volvulus of the midgut loop (Fig. 62.9).

Table 62.2  Differential diagnosis of duodenal obstruction

Upper bowel obstruction
Malrotation
Pyloric atresia
Prepyloric antral diaphragm
Gastric volvulus
Pyloric stenosis
Fig. 62.8  “Wind-sock” web. Dilated duodenum demon- Jejunal-Ileal atresia and stenosis
strated with a duodenal membrane ballooned distally, giv- Preduodenal portal vein
ing characteristic “wind-sock” appearance. The reflux of Duodenal duplication cyst
contrast medium into pancreatic and common bile duct is Choledochal cyst
seen Superior mesenteric artery syndrome

a b

Fig. 62.9  Duodenal obstruction caused by (a) congenital adhesion-Ladd’s bands. (b) Duodenal obstruction caused by
volvulus
62  Duodenal Obstruction
62.6.1 Malrotation
Malrotation may appear as acute and complete,
most often seen in infancy, or as chronic and partial
in older children. The clinical presentation of mal-
rotation in the neonatal period may be either (1)
recurrent episodes of subacute obstruction with
intermittent bilious vomiting or (2) strangulating
intestinal obstruction as a consequence of midgut
volvulus. The infant presents with bile-­ stained
vomiting that may contain altered blood, abdomi-
nal distention, and tenderness; the passage of dark
blood per rectum; and shock. As the strangulation
progresses to gangrene, perforation, and peritoni-
tis, edema and erythema of the anterior abdominal
wall becomes evident. An upper gastrointestinal
contrast study is the procedure of choice showing
the abnormal configuration of the duodenum
(“corkscrew” appearance), the duodenojejunal
junction to the right of the midline, and small bowel
located on the right side of the abdomen. The diag-
nostic investigation should be done urgently fol-
lowed by an emergent operative intervention.
62.6.2 Pyloric Atresia and Prepyloric
Antral Diaphragm
Pyloric atresia and prepyloric antral diaphragm
are rare congenital malformations causing gastric
outlet obstruction. They present shortly after
birth with persistent nonbilious vomiting. In
pyloric atresia, a plain X-ray of the abdomen will
show gas in the dilate stomach—single gas bub-
ble sign—and no gas below the pylorus.
62.6.3 Gastric Volvulus
Gastric volvulus is a rare congenital anomaly.
Plain abdominal and chest radiographs are essen-
tial and will show a distended stomach in an
abnormal position. Contrast studies clarify the
anatomy and the site of obstruction.
62.6.4 Pyloric Stenosis
The clinical presentation of infants with hypertro-
phic pyloric stenosis is nonbilious vomiting usu-
837
ally occurring at 2–8 weeks of age. The diagnosis
is based on the clinical history, palpable “pyloric
tumor”—an “olive”—palpable by physical exami-
nation and ultrasonographic scanning of the abdo-
men that reveals a typical hypoechoic ring with an
echogenic center of increased muscle thickness.
62.6.5 Jejunoileal Atresia
and Stenosis
This type of obstruction presents with abdominal
distention and bile-stained vomiting. Diagnosis
is confirmed on supine and erect abdominal
radiographs that reveal distended small bowel
loops and air-fluid levels.
62.6.6 Preduodenal Portal Vein
Preduodenal portal vein is a rare and asymptom-
atic anomaly and is often impossible to diagnose
prior to surgery. It often coexists with other
anomalies resulting in bowel obstruction (Singal
et al. 2009).
Duodenal duplication cyst, choledochal cyst,
and superior mesenteric artery syndrome are rare
causes of incomplete DO in infancy and are usu-
ally diagnosed in childhood or adolescence.
62.7 Preoperative Management
Although duodenal atresia is a relative emergency,
the patient should not be rushed to the operating
room until the infant’s hemodynamic and fluid
and electrolyte status is stable. If the clinical his-
tory and findings on physical examination indi-
cate that the baby is in no distress and the
radiographs are consistent with the usual presen-
tation of duodenal atresia, with no air beyond the
second bubble (excluding malrotation), the opera-
tion should be performed on an elective basis.
An orogastric tube decompresses the stomach,
and intravenous fluid resuscitation can be initi-
ated. Blood samples for electrolyte determination
should be obtained, and any derangements should
be corrected. Prolonged vomiting can result in
hypokalemic hypochloremic metabolic alkalosis.
838
The passage of the orogastric tube rules out
esophageal atresia, and a careful inspection of
anal defect variants of imperforated anus should
be obtained.
Care is taken to preserve body heat and avoid
hypoglycemia, since many of these newborn
patients are premature and small for their age
(Murshed et  al. 1999). Very low birth weight
infants or those with respiratory distress syn-
drome and associated severe anomalies, e.g.,
congenital heart disease, may occasionally need
special preparation, such as resuscitation and
ventilation.
62.8 Operation
Duodenoduodenostomy is the procedure of
choice for patients with duodenal atresia, steno-
sis, and annular pancreas (Dalla Vecchia et  al.
1998; Bethell et al. 2020a, b).
Duodenoduodenostomy can be performed in
either “diamond-shaped” (proximal transverse to
distal longitudinal) anastomosis as described by
(Kimura et al. 1977) (Fig. 62.10) or a side-to-side
a b
Fig. 62.10  Diamond-shaped duodenoduodenostomy. (a)
A transverse incision is made in the distal end of the prox-
imal dilated duodenum, and a longitudinal incision is
made in the smaller limb of the duodenum distal to the
Y. Sweed and A. Vachyan
fashion (Fig. 62.11). “Diamond-shaped” duode-
noduodenostomy has been reported to allow
­earlier feeding, earlier discharge, and good long-
term results (Kimura et al. 1990).
Bax et  al. (2001) and Rothenberg (2002)
reported on the first case and the first series,
respectively, on the laparoscopic management of
DO.  They indicated that the laparoscopic
approach has proven to be safe and effective and
represents an alternative to the open procedure.
They also emphasized that this minimally inva-
sive surgical technique should only be used if the
surgeon has appropriate instruments and suturing
and laparoscopic skills (Bax et  al. 2001; Kay
et al. 2009; Rothenberg 2002).
Recently, a systematic review and meta-­
analysis investigating the outcome of surgical
laparoscopic repair of DO, compared to open
repair, found that laparoscopic repair shows com-
parable safety and efficiency to open repair and
that there were no significant differences in terms
of time to feeds, time to full enteral feeding, hos-
pital stay, and the overall complication rate; how-
ever, laparoscopy was associated with longer
operative time (Mentessidou and Saxena 2017).
occlusion. (b) A single-layer anastomosis using inter-
rupted 5–0 Vicryl sutures with posterior knots tied inside
the posterior wall of the anastomosis and anterior knots
tied outside the anterior wall is performed
62  Duodenal Obstruction
a b
Fig. 62.11  Side-to-side duodenoduodenostomy. (a) An
upper transverse abdominal incision. (b) Parallel incisions
of about 1 cm are made in the proximal and distal duode-
62.9 Incision
The baby is placed supine on the table with a
small roll under his upper abdomen on a warming
blanket. Endotracheal anesthesia is used. The
abdominal skin is prepared by cleaning with pre-
warmed povidone-iodine.
A transverse supra-umbilical abdominal inci-
sion is made 2 cm above the umbilicus, starting
in the midline and extending laterally into the
right upper quadrant for about 5 cm. The abdomi-
nal muscles are divided transversely with cutting
diathermy, and the peritoneal cavity is opened in
the line of incision.
62.10 Exploration
and Identification
of Pathology
After exposing the peritoneal cavity, the surgeon
inspects the entire bowel for the presence of other
bowel anomalies. There may be an associated
annular pancreas, malrotation (in about one third
of the patients), or in rare cases, preduodenal por-
839
c
num. (c) The anastomosis is performed using single-layer
interrupted 5–0 Vicryl sutures
tal vein. If the colon is in normal position, malro-
tation is probably not a coexisting factor.
The stomach and first portion of the duode-
num are usually thickened and dilated. The liver
is carefully retracted superiorly. The ascending
colon and the hepatic flexure of the colon are
mobilized medially and downward to expose the
dilated duodenum (Sweed and Vachyan 2019).
The duodenum is then adequately mobilized
and freed from its retroperitoneal attachments—
Kocher maneuver. Great care must be exercised
not to dissect or manipulate either segment of the
duodenum medially, to avoid injury to the
ampulla of Vater or the common bile duct. The
tube in the stomach is then passed distally into
the dilated duodenum and helps to locate the
point of obstruction and determine if a “wind-­
sock” deformity is present (Fig. 62.3g).
The type of atresia, as well as any pancreatic
abnormality (e.g., annular pancreas), is noted. In
patients with an annular pancreas, the pancreatic
tissue should never be divided and should always
be bypassed. The duodenum distal to the site of
obstruction is small and decompressed. The
requirements for distal mobilization vary accord-
840 Y. Sweed and A. Vachyan

ing to the location of the atresia and to the gap anterior wall. Before the completion of the ante-
between the two segments (Fig. 62.2). If neces- rior part of the anastomosis, a 5 F silicon nasojeju-
sary, the ligament of Treitz is divided, and mobi- nal transanastomotic feeding tube may be passed
lization and displacement of the distal duodenum down into the upper jejunum for an early postop-
are performed behind the superior mesenteric erative enteral feeding (Hall et  al. 2011; Bethell
vessels, thus allowing a satisfactory anastomosis et al. 2020a, b), using the same insertion technique
to be performed without any tension. as was reported for patients who underwent surgi-
cal repair for esophageal atresia and tracheoesoph-
ageal fistula (Sweed et al. 1992). Others, however,
62.11 “Diamond-Shaped” do not use the nasojejunal tube because they sug-
Duodenoduodenostomy gest that it may delay the commencement of oral
feeding (Kimura et al. 1990). Hall et al. reported
After abdominal exploration, the duodenum is recently that a transanastomotic tube significantly
adequately mobilized. With two traction sutures, shortens time to full enteral feeds in infants with
the redundant wall of the proximal duodenum is congenital DO as well as significantly reducing
pulled downward to overlie the proximal portion the need for central venous access and parenteral
of the distal duodenal segment. A transverse inci- nutrition (Hall et al. 2011). Then, the right colon is
sion is made in the distal end of the proximal returned to its former position so that the mesoco-
duodenum, and a longitudinal incision is made in lon covers the anastomosis. The Ladd procedure
the smaller limb of the duodenum distal to the with “inversion appendectomy” is performed in
occlusion. These are made in such a position as to patients with malrotation (Grosfeld and Rescorla
allow good approximation of the openings with- 1993). In these patients, the cecum should be
out tension. The papilla of Vater is located by placed in the left lower quadrant to reduce the risk
observing bile flow. This is performed by the of midgut volvulus.
gentle compression of the gallbladder. The wound is closed in layers: the peritoneum
The orientation of the sutures in the “diamond-­ and posterior fascia and the anterior fascia by two
shaped” anastomosis and the overlapping between layers using continuous 4–0 Vicryl. The skin is
the proximal transverse incision and the distal closed with a running intracuticular suture using
longitudinal incision are shown in Fig. 62.10a, b. 5–0 Vicryl.
Additionally, an 8 French Foley catheter
should be passed proximally into the stomach
and distally into the jejunum and pulled back 62.12 Side-to-Side
with the balloon inflated, to ensure that no Duodenoduodenostomy
­additional web or a “wind-sock” deformity is
overlooked. The distal duodenum can be dis- The dilated proximal duodenum and the col-
tended to a larger size during this maneuver, lapsed distal duodenum are approximated using
facilitating the anastomosis. Before pulling back two stay sutures (5–0 Vicryl). Then, parallel
the catheter from the distal duodenum, the sur- incisions with a length of about 1 cm are made in
geon should inject 30–40  ml of warm saline the proximal and distal duodenum (Fig. 62.11).
through the catheter to rule out distal atresias of An 8 French Foley catheter should be inserted
the distal small bowel. The catheter is then both to the proximal dilated duodenum and to
removed. the distal collapsed duodenum in order to rule
The Duodenoduodenal anastomosis is per- out “wind-­sock” membrane and distal atresias
formed by a single-layer anastomosis using 5–0 or similar to that described in the “diamond-
6–0 Vicryl sutures with posterior knots tied inside shaped” duodenoduodenostomy.
the posterior wall of the anastomosis and inter- The posterior layer of anastomosis is com-
rupted sutures with anterior knots tied outside the pleted using interrupted 5–0 Vicryl sutures.
62  Duodenal Obstruction
At this stage, a transanastomotic 5 French
gauge silastic nasojejunal tube may be inserted
for an early enteral feeding.
The anastomosis is then completed using
interrupted 5–0 Vicryl sutures for the anterior
layer. The abdomen is closed in the same man-
ner as described in the “diamond-shaped” duo-
denoduodenostomy.
In premature infants, some surgeons prefer to
perform a gastrostomy and insert the transanasto-
motic silicon tube via the gastrostomy. The tip of
the tube should be well down in the jejunum to
decrease the chance of it becoming displaced.
a b
d e
Fig. 62.12  Operative technique for duodenal web. (a)
Longitudinal incision above the “transitional zone” of the
duodenum. (b) An excision of the web leaving the medial
third of the membrane intact. (c) The duodenum is closed
transversely. (d) Intraoperative photograph of a 2-week-­
old infant born with duodenal web with an aperture. The
papilla of Vater was identified at the proximal and medial
part of the duodenal membrane (∗). DU, opened duode-
841
62.13 Operative Technique
for Duodenal Web
A longitudinal incision is performed above the
“transitional zone” between the wide and narrow
segments of the duodenum (Fig. 62.12a), and the
duodenum is opened. The membrane usually is
located in the second part and occasionally in the
third portion of the duodenum. It can be complete
or have a hole. Anatomically, the ampulla of
Vater may open directly into the medial part of
the membrane, or posteriorly close to it; thus the
close relationship of the membrane to papilla of
c
f
num; M, membrane; ∗, papilla of Vater; SB, small bowel.
The white arrow points to the excisional line of the mem-
brane. (e) Intraoperative photograph of a 2-day-old infant
with Down syndrome and AV canal showing: an 8 French
Foley catheter which was introduced through an aperture
of a duodenal membrane. M, membrane; the white arrow
points to the aperture of the membrane. (f) An excision of
the membrane
842
Vater makes its identification mandatory, before
the excision of the web. The excision of the web
should proceed from the lateral duodenal wall,
leaving the medial third of the wall intact to avoid
damaging the sphincter of Oddi or the ampulla of
Vater and continue leaving a circumferential rim
of tissue of 1–2 mm (Fig. 62.12b, f). The resec-
tion line is then sewn using continuous sutures of
5–0 Vicryl, and the duodenum is closed trans-
versely in one layer using 5–0 Vicryl (Fig. 62.12c).
Because of the pitfalls in cases of the lax mem-
brane that may bulge downward distally into the
distended duodenum (the so-called wind-sock
phenomenon), and in order to avoid missing the
anomaly, before closure of the duodenum, the
distal patency of the distal duodenum must be
verified by inserting an 8 French Foley catheter
through duodenotomy (Fig. 62.12e).
The experience with fiber optic duodenoscopy
indicates the usefulness of the technique for both
the diagnosis and nonoperative management of
duodenal membrane (Bittencourt et  al. 2012).
However, based on reports describing the anoma-
lous entry of the pancreatobiliary channels
(Adams 1986), the delineation of the ducts an
endoscopic retrograde cholangiopancreatogra-
phy (ERCP) may be necessary prior to endo-
scopic intervention.
Bittencourt et  al. reported on three female
patients aged between 9 and 12 months born with
duodenal membrane who were treated success-
fully by two endoscopic sessions. The first and
second sessions of endoscopic treatment included
dilatation and resection of the membrane, respec-
tively, and were carried out without complica-
tions (Bittencourt et al. 2012).
Most surgeons, however, believe that the sur-
gical management of DO is preferable and is the
procedure of choice due to the potential risk of
inadvertent pancreatic or bile duct injuries that
may occur from endoscopy (Adams 1986).
62.14 Laparoscopic Management
of DO
The application of minimally invasive surgical
techniques (MIS) for the correction of congenital
anomalies has increased significantly over the
Y. Sweed and A. Vachyan
last 10 years. The ability to perform delicate dis-
section and intracorporeal anastomosis has
broadened the scope of entities that can be
approached including neonatal DO.  Although
most neonatal conditions presenting with bowel
obstruction present a difficult problem for lapa-
roscopy because of the dilated bowel and limited
abdominal cavity, this is not the case in duodenal
atresia. The entire small and large bowel is
decompressed, and there is an excellent exposure
of the proximal duodenum (Bax et al. 2001; Kay
et al. 2009; Chung et al. 2017; Mentessidou and
Saxena 2017).
For the laparoscopic approach (Fig.  62.13),
neonatal laparoscopic instruments (3  mm) and
trocars are used. The patient is positioned supine
at the end of the operating table. The operating
surgeon stands at the patient’s feet. The abdomen
is insufflated through a 5-mm umbilical port, for
a 30° laparoscope. The 3-mm and 5-mm ports are
placed in the right lower quadrant and left upper
quadrant, respectively. The left upper quadrant
port is placed for the introduction of suture.
The duodenum is then kocherized, the type of
obstruction becomes easily visible, and the
dilated proximal and decompressed distal seg-
ments are identified (Rothenberg 2002; Son and
Kien 2017). A proximal transverse and distal lon-
gitudinal duodenotomy is then made.
As with the open repair, stay sutures are placed
at each corner to facilitate the anastomosis. A
diamond-shaped anastomosis is performed with
either a separate running suture for the posterior
and then anterior wall or single interrupted
Scissors, needle
driver
Curved
Camera
Dissector
Fig. 62.13  Trocar placement and instrumentation for
laparoscopic repair of duodenal atresia
62  Duodenal Obstruction
a b
Fig. 62.14  Laparoscopic repair of a 2-day-old baby with
Down syndrome (birth weight of 2200 gr). (a) The longi-
tudinal incision of the narrow distal segment of the duode-
stitches of Vicryl. Intracorporeal knot tying is
used. An extra port can be placed in the right
upper quadrant to help retract the liver and set up
the anastomosis. Alternatively, the apical stitch
can be tied and brought out through the abdomi-
nal wall to assist with retraction and align the
enterotomies for the anastomosis. The distal
bowel is examined in all cases to ensure that there
are no obvious secondary atresias. Once the anas-
tomosis is completed (Fig. 62.14), the ports are
removed, and the sites are closed with absorbable
sutures.
The main benefits of laparoscopic approach
for the treatment of duodenal atresia are the
excellent visualization of the obstruction and the
ease of the anastomosis. However, the possible
disadvantage of this approach may be that evalu-
ation of the distal bowel for other atretic seg-
ments is more difficult to accomplish (Chung
et al. 2017) if not specifically evaluated, and it is
feasible that a malrotation can also be missed
(Hill et al. 2011). Thus, the entire bowel should
be inspected visually for distal obstructed
segments.
Hill et al. (2011) reported on their results com-
paring 22 patients with DO treated by laparos-
copy and 36 patients treated by traditional
laparotomy during a 9-year period (2001–2010).
They found no difference between the groups in
843
num. (b) The picture after the completion of the duodenal
anastomosis (PWS, proximal wide segment of the duode-
num; DNS, distal narrow segment of duodenum)
time to full feeding, postoperative length of stay,
and complication rate. They found that the opera-
tive time was slightly longer in the laparoscopic
group (median time 116 min vs 103 min); how-
ever, laparoscopic management appeared to
allow a shorter postoperative ventilator require-
ment. Six patients (26%) of the laparoscopic
group were converted to open exploration
because of unclear anatomy.
Chung et  al. (2017) recently reported on the
results of a systematic review of 4 retrospective
cohort studies, comprised of 180 patients, com-
paring the laparoscopic and the open surgery, and
they found no significant differences in terms of
operative time, ventilator dependence, time to
initial enteral feeding, time to full enteral feed-
ing, and hospital stay. They concluded that lapa-
roscopic surgery for the repair of DO is a safe and
effective practice; however, they emphasized that
attention should be paid to the possibility of dis-
tal pathology (Chung et al. 2017).
The experience with laparoscopic duodeno-
duodenostomy (Chung et al. 2017; Son and Kien
2017) demonstrates that it can be performed
safely and successfully in the neonate with excel-
lent short-term outcomes. Surgeons with experi-
ence in advanced laparoscopic techniques can
learn the laparoscopic duodenoduodenostomy
and have good results.
844 Y. Sweed and A. Vachyan

62.15 Postoperative Care 62.17 Outcome and Long-Term

Results
The baby is returned to an incubator (or radiant
heat cot) at the thermoneutral temperature for its The survival of babies with DO has gradually
size and maturity. An intravenous infusion of the improved over the last 40  years (Table  62.3)
dextrose/saline is continued in the postoperative (Bailey et al. 1993; Cohen-Overbeek et al. 2008;
period, and further fluid and electrolyte manage- Dalla Vecchia et  al. 1998; Hill et  al. 2011;
ment depends on clinical progress, loss by gas- Kilbride et al. 2010; Wesley and Mahour 1977).
troduodenal aspiration, and serum electrolyte All agree that the three main factors contributing
levels. Postoperatively, patients may have a pro- to the morbidity and mortality in patients with
longed period of bile-stained aspirate through the DO are the high incidence of associated anoma-
nasogastric tube, which is mainly due to the lies, especially cardiac anomalies, which are the
inability of the markedly dilated duodenum to main factor affecting the mortality rate. The other
produce effective peristalsis. Enteral feeding two factors are prematurity and low birth weight
through the transanastomotic feeding tube is gen- (Choudhry et  al. 2009; Escobar et  al. 2004;
erally started within 24–48 h postoperatively. Grosfeld and Rescorla 1993). In a review cover-
The commencement of oral feeding depends ing 45 years (1951–1995) of management of DO,
on the decrease of the gastric aspirate and may Murshed et  al. (1999) found that in the first
be delayed for several days and occasionally for 15  years, survival reached 51%; in the next
1–2  weeks or longer. Once the volume of the
gastric aspirate decreases, the feeding tube is
withdrawn, and the infant can be started on oral Table 62.3  Survival rates of patients with duodenal atre-
sia and stenosis. The gradual improvement in survival
feeding. reported over the past 40 years. Most deaths are related to
the associated complex cardiac malformations
No. of %
62.16 Management of Persistent Author/year patients survival
Megaduodenum by Wesley and Mahour (1977) 72 74
Duodenoplasty Akhtar and Guiney (1992) 49 94
Bailey et al. (1993) 138 93
Grosfeld and Rescorla 103 95
The deformity and dysfunction of the first part (1993)
of the duodenum—the megaduodenum—are the Dalla Vecchia et al. (1998) 138 86
causes of well-known morbidity (Spigland and Cohen-Overbeek et al. 91 91
Yazbeck 1990), and occasionally, these patients (2008)
require duodenoplasty (Dewan and Guiney Choudhry et al. (2009) 65 96
1990). Several techniques of duodenoplasty Kilbride et al. (2010) 51 98
Kay et al. (2009)a 17 100
have been described, and in all, it is of the
Hill et al. (2011)b 58 100
utmost importance to visualize and identify the Bethell et al. (2020a, b)c 103 98
ampulla of Vater within the duodenal lumen a
  Laparoscopic duodenoduodenostomy (years: 2004–
prior to the resection and tapering of the duode- 2008), N = 17 patients
num. Hutton and Thomas have reported success b
 Laparoscopic versus open repair of duodenal obstruction
by extensive tapering duodenoplasty (Hutton (years: 2001–2010), laparoscopic group, N = 22 patients;
open group, N = 36 patients
and Thomas 1988). Dewberry et al. emphasized c
 There were two (2%) deaths within 28 days of operative
recently on the merit of tapering duodenoplasty intervention, due to causes unrelated to DO, both of whom
at the primary operation of neonates with dilated had trisomy 21. The majority of repairs were undertaken
duodenum, to improve the immediate postoper- via laparotomy (97/102, 95%), with only five patients
(5%) having laparoscopic repair. This was a prospective
ative gastrointestinal function and the preven-
observational study that was collected from 28 pediatric
tion of further development of megaduodenum surgical centers in the United Kingdom over a 1-year
(Dewberry et al. 2019). period from 1 March 2016
62  Duodenal Obstruction
15 years, it was 80%; and in the last 15 years, it
reached 95%.
Dalla Vecchia et al. reported a relatively low
rate of postoperative complications in a series of
138 infants (Dalla Vecchia et al. 1998). The early
complication rate included anastomotic obstruc-
tion in 3%, congestive heart failure in 9%, pro-
longed adynamic ileus in 4%, pneumonia in 5%,
and wound infection in 3%.
Late complications included adhesive bowel
obstruction in 9%, megaduodenum and duodenal
dysmotility that required tapering duodenoplasty
in 4%, and gastroesophageal reflux requiring sur-
gery in 5%.
Long-term results of congenital DO were
reported by Kokkonen et al. (1998), who studied
41 patients aged 15–35  years. They found that
growth and development, including body weight,
were satisfactory. Although the great majority
was symptom-free, on barium meal examination,
all but two had abnormal findings, including
megaduodenum in nine cases. They concluded
that some gastrointestinal disturbances are
common, even in asymptomatic patients, and
­ 201  ±  47  minutes p  <  0.004), and the time to
careful follow-up is important.
Recently, Son and Kien (2017) compared the
results of laparoscopic versus open surgery in the
management of duodenal obstruction in 112 neo-
nates. They reported that the laparoscopic treat-
ment is associated with lower postoperative
morbidity, shorter recovery time and postopera-
tive hospital stay, and better postoperative cos-
mesis than open surgery.
Mentessidou and Saxena (2017) and Chung
et al. (2017) recently performed systematic reviews
and found that laparoscopic repair shows compa-
rable safety and efficacy to the open repair for DO.
62.18 Conclusion
Duodenal atresia is a common intestinal obstruc-
tion, often complicated by associated anomalies,
low birth weight, and prematurity. Currently,
duodenoduodenostomy is the treatment of choice
and can be performed via an open approach or by
laparoscopic repair. The latest retrospective stud-
ies comparing the surgical outcome of laparo-
845
scopic repair versus open repair found that
laparoscopy is a safe and effective technique, and
the results including operative time, length of
stay, time to full feeding, and complication rate
were similar in both groups (Chung et al. 2017;
Son and Kien 2017).
Zani et al. reported for those preferring lapa-
roscopic DO repair that laparoscopic duodenoje-
junostomy might be a feasible alternative in these
babies as it is easier to perform and has equal
clinical outcomes compared to duodenoduode-
nostomy (Zani et al. 2017).
Holler et  al. recently reported on a novel,
safe, and feasible technique, using a laparo-
scopic miniature stapler for the repair of duode-
nal atresia. They compared the outcomes with
the new technique, using a miniature stapler in
10 patients, to those of a historic cohort of lapa-
roscopic hand-­ sewn (21 patients) and open
repairs (13 patients). They found that the mean
operative time was significantly shorter in the
laparoscopic stapled group compared to laparo-
scopic hand-sewn (145  ±  37  minutes versus
initiation of feeds and the time to full feeds
were significantly shorter in the laparoscopic
stapled group compared with the open approach.
The overall complication rate was similar
between the groups. They also reported that
duodenojejunostomy was performed more fre-
quently in the laparoscopic stapled group com-
pared with the open procedure (Holler et  al.
2019).
Over the last two decades, advancements in
neonatal intensive care; parenteral nutrition;
management of associated anomalies; and
improvements in operative technique, including
video equipment, smaller instruments, and better
postoperative care, have improved the outlook
for patients born with duodenal atresia and steno-
sis. Mortality rate today has been reduced to
nearly 5% and is now mostly related to the asso-
ciated cardiac anomalies.
Cross-References 
• Esophageal Atresia
• Jejunoileal Atresia and Stenosis
• Malrotation
846 Y. Sweed and A. Vachyan

References Grosfeld JL, Rescorla FJ (1993) Duodenal atresia and ste-

nosis: reassessment of treatment and outcome based
on antenatal diagnosis, pathologic variants and long
Adams DB (1986) Management of the intraluminal duo-
term follow up. World J Surg 17:301–309
denal diverticulum: endoscopy or duodenotomy? Am
Gross E, Armon Y, Abudalu K et al (1996) Familial com-
J Surg 151(4):524–526
bined duodenal and jejunal atresia. J Pediatr Surg
Akhtar J, Guiney EJ (1992) Congenital DO.  Br J Surg
31(11):1573
79:133–135
Haeusler MC, Berghold A, Stoll C et al (2002) Prenatal
Bailey PV, Tracy TF Jr, Connors RH et al (1993) Congenital
ultrasonographic detection of gastrointestinal obstruc-
DO; a 32 year review. J Pediatr Surg 28:92–95
tion: results from 18 European congenital anomaly
Bax NM, Ure BM, van der Zee DC et  al (2001)
registries. Prenat Diagn 22(7):616–623
Laparoscopic duodenoduodenostomy for duodenal
Hall NJ, Drewett M, Wheeler RA et  al (2011) Trans-­
atresia. Surg Endosc 15(2):217
anastomotic tubes reduce the need for central venous
Best KE, Tennant PWG, Addor MC et  al (2012)
access and parenteral nutrition in infants with congen-
Epidemiology of small intestinal atresia in Europe: a
ital DO. Pediatr Surg Int 27(8):851–855
regiser-based study. Arch Dis Child Fetal Neonatal Ed
Hill S, Koontz CS, Langness SM et al (2011) Laparoscopic
97:F353–F358
versus open repair of congenital DO in infants. J
Bethell GS, Long AM, Knight M et al (2020a) Congenital
Laparoendosc Adv Surg Tech 21(10):961–963
obstruction in the UK: a population-based study. Arch
Holler AS, Muensterer OJ, Martynov I et al (2019) 2019;
Dis Child Fetal Neonatal Ed 105:F178–F183
J. Laparoendosc Adv Surg Tech A 29(10):1216–1222
Bethell GS, Long AM, Knight M et  al (2020b) The
Hutton KA, Thomas DF (1988) Tapering duodenoplasty.
impact of trisomy 21 on epidemiology, management
Pediatr Surg Int 3:132–134
and outcomes of congenital duodenal obstruction: a
Iuchtman M, Brereton R, Spitz L et al (1992) Morbidity
population-­based study. Pediatr Surg Int 36:477–483
and mortality in 46 patients with the VACTERL asso-
Bittencourt PFS, Malheiros RS, Ferreira AR et al (2012)
ciation. Isr J Med Sci 28:281–284
Endoscopic treatment of congenital duodenal mem-
Iwai A, Hamada Y, Takada K et  al (2009) Choledochal
brane. Gastrointest Endosc 76(6):1273–1275
cyst associated with duodenal atresia: case report and
Casaccia G, Bilancioni E, Nahom A et  al (2002) Cystic
review of the literature. Pediatr Surg Int 25:995–998
anomalies of biliary tree in the fetus: it is possible to
Kay S, Yoder S, Rothenberg S (2009) Laparoscopic duo-
make a more specific prenatal diagnosis? J Pediatr
denoduodenostomy in the neonate. J Pediatr Surg
Surg 37:1191–1194
44(5):906–908
Chen QJ, Gao ZG, Tou JF et al (2014) Congenital duo-
Kilbride H, Castor C, Andrews W (2010) Congenital
denal obstruction in neonates: a decade’s experience
DO: timing of diagnosis during the newborn period. J
from one center. World J Pediatr 10:238–244
Perinatol 30(3):197–200
Choudhry MS, Rahman N, Boyd P et al (2009) Duodenal
Kimble RM, Harding J, Kolbe A (1997) Additional con-
atresia: associated anomalies, prenatal diagnosis and
genital anomalies in babies with gut atresia of ste-
outcome. Pediatr Surg Int 25:727–730
nosis: when to investigate, and which investigation.
Chung PHY, Wong CWY, Ip DKM et al (2017) J Pediatr
Pediatr Surg Int 12:565–570
Surg 52(3):498–503
Kimura K, Mukohara N, Nishijima E et  al (1990)
Cohen-Overbeek TE, Grijseels EW, Niemeijer ND et  al
Diamond-shaped anastomosis for duodenal atresia: an
(2008) Isolated or non-isolated DO: perinatal outcome
experience with 44 patients over 15 years. J Pediatr
following prenatal or postnatal diagnosis. Ultrasound
Surg 25(9):977–979
Obstet Gynecol 32(6):784–792
Kimura K, Tsugawa C, Ogawa K et al (1977) Diamond-­
Dalla Vecchia LK, Grosfeld JL, West KW et  al (1998)
shaped anastomosis for congenital DO.  Arch Surg
Intestinal atresia and stenosis: a 25-year experience
112(10):1262–1263
with 277 cases. Arch Surg 133:490–496
Knechtle SJ, Filston HC (1990) Anomalous biliary ducts
Dewan LA, Guiney EJ (1990) Duodenoplasty in the
associated with duodenal atresia. J Pediatr Surg
management of duodenal atresia. Pediatr Surg Int
25(12):1266–1269
5:253–254
Kokkonen ML, Kalima T, Jaaskelainen J et  al (1998)
Dewberry L, Hilton SA, Vuille-dit-Bille RN et al (2019)
Duodenal atresia: late follow-up. J Pediatr Surg
Tapering duodenoplasty: a beneficial adjunct in the
23:216–220
treatment of congenital duodenal obstruction. J Int
Komuro H, Ono K, Hoshino N et  al (2011) Bile duct
Med Res 48(1):1–6
duplication as a cause of distal bowel gas in neonatal
Escobar MA, Ladd AP, Grosfeld JL et al (2004) Duodenal
DO. J Pediatr Surg 46(12):2301–2304
atresia and stenosis: long-term follow-up over 30
Lambrecht W, Kluth D (1998) Hereditary multiple atre-
years. J Pediatr Surg 39(6):867–871
sias of the gastrointestinal tract: report of a case and
Gfroerer S, Theilen TM, Fiegel HC et  al (2019)
review of the literature. J Pediatr Surg 33:794–797
Comparison of outcomes between complete and
Louw JH, Barnard CN (1955) Congenital intestinal atre-
incomplete congenital duodenal obstruction. World J
sia: observations on its origin. Lancet 2:1065–1067
Gastroenterol 25(28):3787–3797
62  Duodenal Obstruction
Malone FD, Crmbleholme TM, Nores JA et  al (1997)
Pitfalls of the ‘double bubble’ sign: a case of congeni-
tal duodenal duplication. Fetal Diagn Ther 12:298–300
Markljung E, Adamovic T, Ortqvist L et al (2012) A rare
microduplication in a familial case of annular pancreas
and duodenal stenosis. J Pediatr Surg 47:2039–2043
Mentessidou A, Saxena AK (2017) Laparoscopic repair
of duodenal atresia: systematic review and meta-­
analysis. World J Surg 41(8):2178–2184
Mikaelsson C, Arnbjörnsson E, Kullendorff CM (1997)
Membranous duodenal stenosis. Acta Paediatr
86(9):953–955
Moore SW, de Jongh G, Bouic P et  al (1996) Immune
deficiency in familial duodenal atresia. J Pediatr Surg
31:1733–1735
Morikawa N, Kuroda T, Honna T et  al (2009) A novel
association of duodenal atresia, malrotation, segmen-
tal dilatation of the colon, and anorectal malformation.
Pediatr Surg Int 25:1003–1005
Muraji T, Mahour GH (1984) Surgical problems in
patients with VATER-associated anomalies. J Pediatr
Surg 19:550–554
Murshed R, Nicholls G, Spitz L (1999) Intrinsic DO:
trends in management over 45 years (1951–1995)
with relevance to prenatal counselling. Br J Obstet
Gynaecol 106:1197–1199
Nawaz A, Matta H, Hamchou M et al (2005) Situs inver-
sus abdominus in association with congenital DO: a
report of two cases and review of the literature. Pediatr
Surg Int 21:589–592
Norton K, Tenreiro R, Rabinowitz JG (1992) Sonographic
demonstration of annular pancreas and a distal duode-
nal diaphragm in a newborn. Pediatr Radiol 22:66–67
Pameijer CR, Hubbard AM, Coleman B et  al (2000)
Combined pure esophageal atresia, duodenal atresia,
biliary atresia and pancreatic ductal atresia: prena-
tal diagnostic features and review of the literature. J
Pediatr Surg 35:745–747
Papandreou E, Baltogiannis N, Cigliano B et  al (2004)
Annular pancreas combined with distal stenosis.
A report of four cases and review of the literature.
Pediatr Med Chir 26(4):256–259
Piper HG, Alesbury J, Waterford SD et al (2008) Intestinal
atresia: factors affecting clinical outcomes. J Pediatr
Surg 43(7):1244–1248
Rattan KN, Singh J, Dalal P (2016) Neonatal duodenal
obstruction: a 15-year experience. J Neonatal Surg.
5(2):13
Rothenberg SS (2002) Laparoscopic duodenoduodenos-
tomy for DO in infants and children. J Pediatr Surg
37(7):1088–1089
Singal AK, Ramu C, Paul S et al (2009) Preduodenal por-
tal vein in association with midgut malrotation and
duodenal web-triple anomaly. J Pediatr Surg 44(2):5–7
847
Skandalakis JE, Gray SW (1994) Embryology for sur-
geons: the embryological basis for the treatment of
congenital anomalies. Williams & Wilkins, Baltimore
Son TN, Kien HH (2017) Laparoscopic versus open sur-
gery in management of congenital duodenal obstruc-
tion in neonates: a single-center experience with 112
cases. J Pediatr Surg 52(12):1949–1951
Spigland N, Yazbeck S (1990) Complications associated
with surgical treatment of congenital intrinsic DO.  J
Pediatr Surg 25(11):1127–1130
Srivastava P, Shaikh M, Mirza B, Jaiman R, Arshad M
(2016) Preduodenal portal vein associated with duo-
denal obstruction of other etiology: a case series. J
Neonatal Surg 5(4):54
Sweed Y, Bar-Maor JA, Shoshany G (1992) Insertion of a
soft silastic nasogastric tube at operation for esopha-
geal atresia: a new technical method. J Pediatr Surg
27(5):650–651
Sweed Y, Vachyan A (2019) Duodenal obstruction. In:
Puri P (ed) Pediatric surgery atlas series. Springer,
Berlin, pp 191–200
Sweed Y, Yulevich A (2020) Duodenal obstruction. In:
Puri P (ed) . Springer Nature, Pediatric Surgery
Germany, pp 875–895
Tandler J (1900) Zur Entwicklungsgeschichte des men-
schlichen duodenums in fruhen Embryonalstadiun.
Morphol Jahrb 29:187–216
Teague WJ, Jones MLM, Hawkey L et al (2018) FGF10
and the mystery of duodenal atresia in humans. Front
Genet 9:530
Torfs CP, Christianson RE (1998) Anomalies in down
syndrome individuals in a large population-based reg-
istry. Am J Med Genet 77:431–438
Tsukerman GL, Krapiva GA, Krillova IA (1993) First-­
trimester diagnosis of duodenal stenosis associated
with oesophageal atresia. Prenat Diagn 13:371–376
Wesley JR, Mahour GH (1977) Congenital intrinsic duo-
denal obstruction: a twenty-five year review. Surgery
82(5):716–720
Yin C, Tong L, Ma M (2020) The application of prena-
tal ultrasound in the diagnosis of congenital duodenal
obstruction. BMC Pregnancy Childbirth 20(1):387
Zani A, Yeh JB, King SK, Chiu PP, Wales PW (2017)
Duodeno-duodenostomy or duodeno-jejunostomy for
duodenal atresia: is one repair better than the other?
Pediatr Surg Int 33(2):245–248
Zimmer EZ, Bronstein M (1996) Early diagnosis of duo-
denal atresia and possible sonographic pitfalls. Prenat
Diagn 16:564–566
 Intestinal Malrotation
Mark D Stringer and Prabal R. Mishra
63.1 Introduction
This spectrum of congenital anomalies is due to
aberrant intestinal rotation and fixation occurring
in early gestation. The clinical effects of intesti-
nal malrotation may manifest at any age, but
most symptomatic  cases present in infancy.
Malrotation is an important cause of intestinal
obstruction in the newborn. Failure to diagnose
and treat the condition promptly can result in
fatal midgut volvulus. In 1936, the North
American surgeon, William Ladd, emphasized
the importance of surgical treatment (Ladd 1936),
and his operation (Ladd procedure) remains the
definitive treatment today.
63.2 Epidemiology
The frequency of malrotation is difficult to deter-
mine because of variations in the age of the study
population, methods of detection, and whether or
M. D Stringer (*)
Department of Paediatric Surgery, Wellington
Children’s Hospital and Department of Paediatrics
and Child Health, Wellington School of Medicine,
University of Otago, Wellington, New Zealand
P. R. Mishra
Departments of Paediatric Surgery & Child Health,
Wellington Hospital, Wellington, New Zealand
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_63
63
not asymptomatic individuals or minor degrees of
intestinal malrotation are included. Symptomatic
or autopsy findings of intestinal malrotation were
found in 1:3500 live births and fetal deaths from a
birth defects registry in Hawaii (Forrester and
Merz 2003). A study from a UK regional center
estimated the prevalence of intestinal malrotation
presenting with bilious vomiting in term neonates
as 1 in 9000 (Drewett et al. 2016). The prevalence
of symptomatic malrotation treated by Ladd pro-
cedure between 1 and 18  years of age in the
United States was estimated to be about 1  in
200,000 (Malek and Burd 2005). In contrast, 1 in
500 adults undergoing radiographic contrast stud-
ies for various gastrointestinal symptoms had
abnormal intestinal rotation (Kantor 1934). In
most large series, there is no major difference in
sex incidence (Malek and Burd 2005).
63.3 Embryology
The embryonic period in the developing human
extends to the end of the 8th week of gestation
(56 days), after which the fetal stage of develop-
ment begins. The midgut of the human embryo
receives a blood supply from the superior mesen-
teric artery and is destined to become that part of
the bowel extending from the second part of the
duodenum to two-thirds of the way along the
transverse colon. In the embryo, the midgut
849
850 M. D Stringer and P. R. Mishra

forms a loop, the apex of which is continuous
with the narrowing stalk of the yolk sac (vitel-
lointestinal duct). During the 6th week of gesta-
tion, the midgut loop and its dorsal mesentery
elongate rapidly and temporarily herniate into the
extraembryonic celom within the umbilical cord.
Umbilical herniation occurs because the develop-
ing liver and kidneys are relatively large and
there is not enough room in the embryo's abdo-
men for the rapidly growing midgut. The proxi-
mal (cranial) limb of the midgut loop grows
rapidly to form the small intestine, while the dis-
tal (caudal) limb changes little except for the
development of a cecal diverticulum.
During the stage of physiological herniation,
the midgut rotates 90 degrees counterclockwise
around the axis of the superior mesenteric artery
so that the proximal limb comes to lie on the right
side of the embryo, while the distal limb lies on
the left side (Fig. 63.1). This initial phase of gut

a b
Liver
Dorsal aorta
Stomach Spleen
Spleen

Gallbladder Dorsal
pancreatic bud
Umbilical cord

Cranial
limb

Yolk stalk a1 b1
Caudal
limb Inferior mesenteric artery

Midgut loop Superior mesenteric artery

c
Liver Dorsal mesogastrium

Ventral mesentery Spleen

Stomach
Gallbladder

Duodenum
Umbilical cord

Hindgut

Cecal c1
diverticulum

Fig. 63.1  Rotation of the midgut, as seen from the left.
(a) Around the beginning of the sixth week, showing the
midgut loop in the proximal part of the umbilical cord.
(a1) Transverse section through the midgut loop, illustrat-
ing the initial relationship of the limbs of the loop to the
artery. (b) Later stage showing the beginning of midgut
rotation. (b1) An illustration of the 90-degree counter-
clockwise rotation that carries the cranial limb of the mid-
gut to the right. (c) About 10 weeks, showing the intestines
returning to the abdomen. (c1) An illustration of a further
rotation of 90 degrees. (d) About 11  weeks, after the
return of intestines to the abdomen. (d1) An illustration of
a further 90-degree rotation of the gut, for a total of 270
degrees. (e) Later fetal period, showing the cecum rotating
to its normal position in the lower right quadrant of the
abdomen. (Reproduced by kind permission of Elsevier
from Moore KL and Persaud TVN. The developing
human. seventh edition. Saunders, 2003, p. 270)
63  Intestinal Malrotation 851

d e
Dorsal aorta
Liver

Omental bursa
Lesser
omentum

Cecum Spleen

Descending
Ascending colon
Transverse colon colon

Small
intestine

Sigmoid colon
d1 Rectum
Cecum and appendix

Fig. 63.1 (continued)

rotation is due to the asymmetric expression of

genes on the left side of the dorsal mesentery
under the control of Nodal, a secretory protein
belonging to the transforming growth factor beta
superfamily and encoded by the Nodal gene on
chromosome 10 (Martin and Shaw-Smith 2010).
The midgut returns to the abdominal cavity
between the 10th and 12th weeks of gestation.
The proximal limb (small bowel) reenters first
and passes behind the superior mesenteric artery.
As the large bowel returns, it undergoes a further
180-degree counterclockwise rotation leaving the
cecum on the right side of the abdomen. With
continuing fetal growth, the ascending colon
elongates, and the cecum takes up its normal
position in the right iliac fossa.
As each part of the gut achieves its definitive
position during fetal life, mesenteric attachments
are modified and fixed. Some mesenteric attach-
ments shorten and others such as those associated
with the duodenum and ascending colon become
retroperitoneal by fusion of peritoneal layers. In
the early embryo, the bowel mesentery is attached
to the median plane of the posterior abdominal
wall, but with gut rotation and growth, the fan-­
shaped mesentery of the small intestine eventu-
Fig. 63.2  The normal attachment of the small bowel
ally takes origin along a line extending from the mesentery. The normal broad base of mesenteric attach-
duodenojejunal flexure (ligament of Treitz) to the ment extends from the ligament of Treitz to the ileocecal
ileocecal junction (Fig. 63.2). This broad base of junction. Both the  ascending and descending colon are
the small bowel mesentery helps to prevent axial fixed retroperitoneally. (Reproduced with permission
from Filston HC and Kirks DR. Malrotation—the ubiqui-
rotation and volvulus of the gut. tous anomaly. J Pediatr Surg 1981;16:616, Fig. 63.2)
852
63.4 Pathogenesis
Intestinal malrotation occurs as a result of abnor-
malities of rotation and fixation of the intestine.
The etiology is poorly understood, but several
different mechanisms have either been suggested
or identified:
• Disruption of the critical role played by the
dorsal mesentery in mediating normal gut
rotation is probably a factor in some cases.
• Abnormal ciliary mechanisms account for
some individuals with intestinal malrotation
and heterotaxy and defective genes are likely
to be involved (Salehi Karlslätt et al. 2019).
• Specific genetic mutations have been identi-
fied in a small proportion of patients with
intestinal malrotation (Table  63.1). More are
likely to be discovered in the future, particu-
larly in cases of familial malrotation, which
may be either non-syndromic or syndromic
(Martin and Shaw-Smith 2010). Inherited syn-
dromes include a variety of other intestinal
disorders associated with intestinal malrota-
tion: multiple intestinal atresias (an autosomal
recessive condition that is often fatal); “apple
peel” (type IIIb) jejunal atresia; short gut; or
the megacystis, microcolon, and intestinal
hypoperistalsis syndrome. In addition to
genetic mutations, genomic deletions and
duplications (copy number variants) have
Table 63.1  Genetic mutations in the etiology of intesti-
nal malrotation (Martin and Shaw-Smith 2010)
Gene mutation Potential associations
Mutations in forkhead Severe developmental
box transcription factor pulmonary vascular
FOXF1 (which is malformation (fatal), urinary
critical to formation of tract malformations, overlap
the dorsal mesentery) with heterotaxy syndromes
Mutations in X-linked Intestinal pseudo-­
gene filamin A (FLNA) obstruction, pyloric stenosis,
cryptorchidism
Mutations in genes Biliary atresia, splenic
controlling left-right malformation, congenital
patterning (e.g., CFC1, heart disease, etc.
ZIC3, lefty A, FOXA2)
M. D Stringer and P. R. Mishra
been identified in syndromic malrotation
(Salehi Karlslätt et al. 2019). Non-syndromic
intestinal malrotation has been reported in
monozygotic twins (Bourgouin et al. 2017).
• Intestinal malrotation often occurs in associa-
tion with other major abdominal defects that
mechanically interfere with normal gut rota-
tion such as congenital diaphragmatic hernia,
gastroschisis, and exomphalos.
• The reported association between intestinal
dysmotility and intestinal malrotation in a
minority of patients raises the possibility that
primary intestinal dysmotility could be the
cause of the malrotation in some cases
(Devane et al. 1992).
63.5 Pathology
There is confusion about the terminology, par-
ticularly with regard to non-rotation.
63.5.1 Incomplete Rotation
In the “typical” or “classic” form of malrotation,
the cecum lies in the epigastrium, and the duode-
nojejunal flexure is sited inferiorly and to the
right of the midline (Fig. 63.3). Peritoneal bands
pass from the right upper quadrant lateral to the
duodenum to the cecum and ascending colon
(Ladd bands). Whether these bands by them-
selves commonly cause duodenal obstruction is
controversial; duodenal obstruction is most often
due to associated midgut volvulus. The most haz-
ardous element in incomplete malrotation of the
gut is the narrow pedicle base of the midgut
between the cecum and duodenojejunal flexure
which predisposes to axial midgut volvulus.
Lesser degrees of incomplete rotation (often
referred to as “atypical” malrotation) are rela-
tively common, but the key feature dictating
management in these cases is the width of the
base of the small bowel mesentery which deter-
mines the risk of midgut volvulus.
63  Intestinal Malrotation
Fig. 63.3  Classical intestinal malrotation with a high
central cecum, the duodenojejunal flexure to the right of
the midline, and Ladd peritoneal bands crossing the
duodenum
63.5.2 Non-rotation
In its purest form, this is a rare and less hazardous
form of intestinal malrotation. The failure of
counterclockwise rotation of the midgut loop
around the superior mesenteric artery leaves the
bowel mesentery taking origin along a line close
to the median plane of the posterior abdominal
wall. The duodenal C loop is not properly devel-
oped, the cecum is unfixed, and the small bowel
tends to lie on the right of the abdomen. The
wider base of the gut mesentery in such cases sub-
stantially mitigates the risk of midgut volvulus.
853
63.5.3 Other Rarer Abnormalities
of Rotation/Fixation
In reverse rotation, a rare variant, the midgut
rotates in a clockwise rather than counterclock-
wise direction around the axis of the superior
mesenteric artery. Thus, the third part of the duo-
denum is anterior to the superior mesenteric
artery which in turn lies anterior to the transverse
colon. The transverse colon may become
obstructed by the superior mesenteric artery, or
the ileocecal region can undergo volvulus.
Mesocolic hernias are rare sequelae of malro-
tation and malfixation. A right-sided mesocolic
hernia can develop as a result of non-rotation or
reverse rotation of the proximal midgut (duode-
num and jejunum) in association with the clock-
wise rotation of the distal (colic) limb. The cecum
becomes attached to the retroperitoneum in the
right upper quadrant, and the small intestine
becomes trapped within a hernia sac created by
the mesocolon of the cecum and ascending colon.
A left-sided mesocolic hernia occurs when the
duodenum and jejunum rotate normally behind
and to the left of the superior mesenteric artery
but then invaginate into a sac formed by the left
mesocolon. These internal hernias are a rare
cause of intestinal obstruction and  potential
intestinal strangulation.
Sometimes, rotation occurs normally, but fixa-
tion is defective, resulting in a mobile cecum.
This may predispose to intussusception or cecal
volvulus.
63.6 Associated Anomalies
Intestinal malrotation may occur in isolation but
is associated with a wide range of other congeni-
tal malformations. Non-rotation is seen in asso-
ciation with gastroschisis and in many cases of
congenital diaphragmatic hernia and exomphalos
854
Table 63.2  Congenital anomalies associated with intes-
tinal malrotation
Commonly associated
Abdominal wall defects (gastroschisis, exomphalos)
Congenital diaphragmatic hernia
Well-recognized but less frequent associations
Duodenal stenosis/atresia
Intestinal atresia
Biliary atresia splenic malformation syndrome
Meckel diverticulum
Hirschsprung disease
Anorectal malformations (>with severe types)
Congenital cardiac disease (especially with heterotaxy
syndromes)
Esophageal atresia
Prune belly syndrome
Heterotaxy syndromes
Rarer associations
Intussusception (Waugh’s syndrome)
Meconium ileus
Megacystis, microcolon, intestinal hypoperistalsis
syndrome
Chromosomal disorders: ring chromosome 4; 16q24.1
deletion
References: Stewart et  al. (1976); Filston and Kirks
(1981); Brereton et  al. (1986); Martin and Shaw-Smith
(2010); Pachl et al. (2015); Martinez-Leo et al. (2016)
(although the latter two conditions may also show
typical malrotation). Other less common associa-
tions are shown in Table  63.2. Some cases of
small bowel atresia are associated with intestinal
malrotation, while others are caused by intestinal
malrotation resulting in prenatal volvulus and
infarction of the intestine.
An intrinsic duodenal obstruction (duodenal
stenosis or atresia) is found in a small proportion
of newborns with intestinal malrotation, and this
must be excluded at subsequent surgery.
Ileocolic intussusception has been associated
with intestinal malrotation (usually the incom-
plete atypical variety), and the mobile cecum in
such cases may be a factor in the etiology of
intussusception (Brereton et al. 1986).
63.7 Clinical Features
The majority of symptomatic patients present dur-
ing infancy, many of these in the newborn period
(Stewart et  al. 1976; Filston and Kirks 1981;
M. D Stringer and P. R. Mishra
Aboagye et al. 2014). However, intestinal malrota-
tion has been increasingly diagnosed in older chil-
dren and adults in recent years (Nehra and
Goldstein 2011). The newborn infant with intesti-
nal malrotation is at greatest risk of life-­threatening
midgut volvulus, although this complication can
occur at any age. The most common presentation
of malrotation is bilious (green) vomiting. In most
cases, this is due to duodenal obstruction from the
midgut volvulus rather than compression by Ladd
bands or kinking of the duodenum. Bilious vomit-
ing in infancy is a sign of intestinal obstruction
until proven otherwise. If the acute midgut volvu-
lus is intermittent or incomplete, the infant is well
between attacks of vomiting, although may show
evidence of failure to thrive. Rarely, intestinal mal-
rotation presents with episodic abdominal disten-
sion and lethargy. Persistent chylous ascites, often
beginning prenatally, has also been reported as a
presenting feature of intestinal malrotation (Long
et  al. 2018). The vomiting is not always bilious,
and the absence of bile in the vomit does not
exclude malrotation and volvulus. Abdominal
examination may show no abnormalities. However,
if the acute midgut volvulus becomes established,
then a strangulating closed loop obstruction of the
midgut ensues. The infant becomes progressively
sicker with the development of abdominal disten-
sion and tenderness and the passage of blood-
stained stools. Late and ominous features are
abdominal wall erythema, shock, metabolic acido-
sis, sepsis, and multiorgan failure.
Other presentations of intestinal malrota-
tion that should be considered are as follows:
Prenatal: Midgut volvulus can be a cause of
intestinal atresia(s) that may be suspected
from dilated bowel and/or the “whirlpool”
sign (vide infra) during fetal ultrasound
assessment (Bartholmot et  al. 2019). The
compression of the duodenum by Ladd bands
in utero may also explain the association
between malrotation and some cases of duo-
denal atresia or stenosis.
Premature infant: Extremely premature infants
can present with severe abdominal distension
and feed intolerance with no bilious vomiting
or aspirates (Mishra and Stringer 2021). The
63  Intestinal Malrotation 855

clinical picture may be confused with necro-

tizing enterocolitis.
Child: Beyond infancy, the child with malrota-
tion may present with recurrent or chronic
abdominal pain and/or intermittent vomiting
which may or may not be bilious (Fig. 63.4)
(Powell et  al. 1989; Spigland et  al. 1990;
Nehra and Goldstein 2011; Dekonenko et al.
2019). There may be poor weight gain and
failure to thrive. Chronic, nonspecific symp-
toms may delay the diagnosis of malrotation
and cause the child to be mislabeled as having
a functional disorder (Powell et  al. 1989).
Malabsorption and diarrhea, perhaps with
occult blood loss, arising from chronic vascu-
lar and lymphatic congestion of the gut, are
occasional presenting features of chronic mid-
gut volvulus in this age group. Pancreatitis
and biliary obstruction are exceptionally rare
presentations (Kirby et al. 2000).
Adult: In recent years, intestinal malrotation has
been recognized with increasing frequency in
adults due to more widespread use of abdomi- Fig. 63.4  Upper gastrointestinal contrast study in a boy
nal CT scans and/or a greater awareness of this with recurrent abdominal pain and vomiting due to intes-
disorder. It has long been known that intestinal tinal malrotation. Note that the duodenojejunal flexure
and small bowel lie to the right of the midline
malrotation in adults may be asymptomatic
(Stewart et al. 1976). However, it may present
acutely with midgut volvulus (Fig.  63.5a) in children (Nehra and Goldstein 2011).
(Nehra and Goldstein 2011; Coe et al. 2015) or Symptoms are improved in most affected indi-
appendicitis in an abnormal location or more viduals by a Ladd procedure, but up to one-
insidiously with chronic or recurrent abdomi- third continue to suffer with chronic abdominal
nal pain (Fig. 63.5b), postprandial discomfort, pain (Husberg et al. 2016).
and intermittent nausea and/or vomiting. Incidental: In some patients, malrotation is dis-
Delayed diagnoses or misdiagnoses as irritable covered incidentally during radiographic stud-
bowel syndrome, functional disorders, and ies or at laparoscopy or laparotomy undertaken
peptic ulceration are common. The risk of mid- for other indications (Powell et  al. 1989;
gut volvulus appears to be lower in adults than Nehra and Goldstein 2011).
856
a b
Fig. 63.5 (a) Abdominal CT scan showing the whirlpool
sign of midgut volvulus presenting acutely in a 64-year-­
old man. (b) CT scan showing intestinal malrotation in a
63.8 Radiological Diagnosis
Stable infants with a history of bilious vomiting
(and older children in whom bilious vomiting is a
major symptom) should undergo an urgent upper
gastrointestinal contrast study under fluoroscopy.
If a midgut volvulus is demonstrated, urgent sur-
gery is required.
The plain abdominal radiograph in an infant
with intestinal malrotation may be normal. In
acute midgut volvulus, the radiographic appear-
ances are notoriously variable and may include
the following:
• A relatively normal bowel gas pattern.
• A dilated stomach and duodenum (“double
bubble” appearance) with a paucity of gas dis-
tally (Fig. 63.6).
• Multiple dilated loops of bowel.
• A nearly gasless abdomen (Fig. 63.7).
M. D Stringer and P. R. Mishra
65-year-old woman with a chronic history of recurrent
abdominal pain. The small bowel lies on the right of the
abdomen and the large bowel on the left
In an individual with normal gut rotation, an
upper gastrointestinal contrast study shows that
the duodenojejunal flexure lies to the left of the
midline ascending to the level of the pylorus (to
the left of the L1 vertebral pedicle) and passing
posteriorly on the lateral projection (Fig.  63.8).
Gastric distension, splenomegaly, or scoliosis
may cause the duodenojejunal flexure to be dis-
placed slightly inferiorly or medially but a con-
trast enema in such cases will show the cecum in
the right iliac fossa. In typical intestinal malrota-
tion, an upper GI contrast study performed by an
experienced radiologist shows a duodenojejunal
flexure to the right of the midline below the level
of the pylorus (Fig. 63.9). Proximal small bowel
loops are also seen to the right of the midline. In
the presence of a midgut volvulus, there is incom-
plete duodenal obstruction often in the third part
of the duodenum (Fig. 63.10) with a “beaked” or
“corkscrew” appearance (Fig.  63.11). Delayed
63  Intestinal Malrotation 857

Fig. 63.7  A gasless abdomen in a neonate with intestinal

malrotation and midgut volvulus

Fig. 63.6  A plain radiograph in a neonate with bilious

vomiting showing a dilated stomach and duodenum with a
paucity of gas distally. The baby had malrotation and a
viable midgut volvulus

films or a contrast enema typically show an

abnormally sited cecum.
An upper GI contrast study performed under
fluoroscopy by a pediatric radiologist has a sensi-
tivity of >95% for excluding malrotation, but
false-negative and false-positive results can occur
(Sizemore et al. 2008; Stephens et al. 2012). In a
stable child, barium is the preferred contrast agent
because of its density, but in a sick infant, an iso-
tonic water-soluble contrast agent is usually
selected. An upper GI contrast study is less accu-
rate at detecting midgut volvulus (Stephens et al.
2012). If the upper GI study is equivocal and there Fig. 63.8 Upper gastrointestinal contrast study in a
is a high index of clinical suspicion, it can be help- 10-month-old infant showing normal intestinal rotation
ful to identify the position of the cecum on the with the duodenojejunal flexure (DJF) to the left of the
midline ascending to the level of the pylorus (to the left of
small bowel follow-through or by contrast enema.
the L1 vertebral pedicle)
858
Fig. 63.9  An upper gastrointestinal contrast study show-
ing the duodenojejunal flexure to the right of the midline
in a child with symptomatic intestinal malrotation
A barium enema is inferior to an upper gastro-
intestinal contrast study and, by itself, cannot be
relied on to exclude malrotation. This is because
a high and mobile cecum can be present in some
healthy individuals without malrotation, and mal-
M. D Stringer and P. R. Mishra
rotation can sometimes occur despite the pres-
ence of the cecum in the right lower quadrant.
The cecum often lies in a higher relatively unfixed
position in neonates and can continue to descend
postnatally (Mirjalili et al. 2017).
An ultrasound scan by an experienced sonog-
rapher can be helpful in diagnosis particularly
when there is a midgut volvulus. Normally, the
superior mesenteric vein (SMV) lies to the right
of the superior mesenteric artery (SMA), but in
malrotation this relationship is reversed, i.e., the
SMA is to the right of the vein (Fig. 63.12). Lesser
degrees of inversion of these vessels (where they
lie anteroposterior to one another) are much less
specific (Orzech et al. 2006). A normal anatomic
relationship between the SMA and SMV does not
exclude malrotation, and an upper gastrointestinal
contrast study remains the gold standard investi-
gation. Ultrasound scanning is both specific and
sensitive for the diagnosis of midgut volvulus,
when color Doppler examination may show a
“whirlpool” flow pattern of the SMV around the
SMA (Fig. 63.13) together with a distended fluid-
filled proximal duodenum (Zhang et al. 2017).
Double contrast CT scan has been used in
teenagers and adults to diagnose intestinal malro-
tation (Fig.  63.5b) and, rarely, magnetic reso-
nance imaging in circumstances where ionizing
radiation would be hazardous such as in the fetus
(Biyyam et  al. 2009). Laparoscopy is useful in
patients where imaging studies have not yielded a
definitive answer (Hsiao and Langer 2011;
Graziano et al. 2015; Sloan et al. 2020).
63  Intestinal Malrotation 859

a b

Fig. 63.10 (a) Anteroposterior and (b) lateral views of an upper gastrointestinal contrast study in intestinal malrotation
with volvulus showing obstruction of the third part of the duodenum

a b

Fig. 63.11 (a, b) Upper gastrointestinal contrast studies showing examples of the “corkscrew” or “twisted ribbon”
appearance of the duodenum and proximal jejunum in infants with intestinal malrotation and midgut volvulus
860
SMV
SMA
Fig. 63.12  Ultrasound scan of the abdomen in an infant
with intestinal malrotation showing the reversed relation-
ship of the superior mesenteric artery (SMA) and vein
Fig. 63.13  An abdominal ultrasound scan in a 5-day-old
baby showing the whirlpool sign of midgut volvulus. The
superior mesenteric vein (SMV) lies anterior to the supe-
rior mesenteric artery (SMA) in this image
M. D Stringer and P. R. Mishra
(SMV). The Doppler arterial waveform confirms the posi-
tion of the SMA. (Kindly provided by Dr. William
Ramsden, St. James’s University Hospital, Leeds)
63.9 Differential Diagnosis
In term neonates with bilious vomiting, a surgical
cause is found in up to 40%, and intestinal malro-
tation accounts for a significant proportion of
these (Godbole and Stringer 2002; Mohinuddin
et  al. 2015). Other causes of bilious vomiting
include intestinal obstruction from other condi-
tions such as intestinal atresia, meconium ileus,
and Hirschsprung disease. In premature infants,
necrotizing enterocolitis may cause bile-stained
vomiting or bilious gastric aspirates. No surgical
cause for bilious vomiting is found in about 60%
of babies (Lilien et al. 1986; Godbole and Stringer
2002; Drewett et al. 2016), but intestinal malrota-
tion should always be excluded.
63  Intestinal Malrotation
63.10 Treatment
Operation should not be delayed once the diag-
nosis of classical intestinal malrotation is made,
although the degree of urgency depends upon the
presence of midgut volvulus. A well child with
typical malrotation but no evidence of midgut
volvulus should undergo surgery as soon as prac-
tically possible. A sick infant with volvulus and
systemic signs requires urgent resuscitation and
surgery without prior radiologic imaging other
than a plain abdominal radiograph. Delay
increases the risk of potentially fatal catastrophic
intestinal infarction. Resuscitation should there-
fore proceed while making arrangements for
emergency surgery. Preoperative resuscitation
should consist of adequate volume replacement,
nasogastric decompression, correction of electro-
lyte and acid-base disturbances, and the adminis-
tration of broad-spectrum intravenous
antibiotics.
Asymptomatic patients with incidentally dis-
covered typical malrotation (found at radiography
or surgery) should, regardless of their age, have
prompt corrective surgery if a narrow base to the
midgut mesentery is suspected. However, minor
degrees of incomplete rotation are relatively com-
mon, and the management of these patients is
controversial (Mehall et  al. 2002; McVay et  al.
2007; Graziano et al. 2015). For example, a child
with minor nonspecific abdominal pain investi-
gated by an upper gastrointestinal contrast study
and found to have a duodenojejunal flexure in a
low and/or midline position and a cecum in the
right lower quadrant is at a very low risk of volvu-
lus. In such cases, surgery is unlikely to be benefi-
cial and potentially harmful (e.g., risking adhesive
bowel obstruction). Children and adults with
atypical and chronic symptoms do not usually
benefit from the correction of atypical malrota-
tion. In some of these, malrotation is probably the
consequence of a primary abnormality in gut
motility rather than the root cause of symptoms
(Coombs et  al. 1991; McVay et  al. 2007). In
selected cases where the risk of midgut volvulus
remains uncertain, laparoscopy can be used to
clarify the anatomy and risk (Waldhausen and
Sawin 1996; Graziano et al. 2015).
861
63.10.1 Ladd Procedure
The operation aims to prevent the development of
future midgut volvulus by three mechanisms:
widening the base of the small bowel mesentery,
positioning the bowel in a non-rotated position
and creating benign adhesions. The operation is
conventionally performed through a transverse
right upper quadrant abdominal incision extended
just to the left of the midline. Ladd procedure
consists of the following operative steps designed
to relieve intestinal obstruction and prevent the
risk of recurrent volvulus:
1. Reduce any volvulus: Volvulus occurs in a
clockwise direction around the SMA, and thus
reduction is achieved by a gentle counter-
clockwise derotation of the gut (Fig.  63.14).
The entire gut must be gently delivered from
the abdomen in order to examine its orienta-
tion and fixation adequately. Chylous ascites
from lymphatic obstruction may be present in
chronic midgut volvulus (Fig. 63.15).
2. Divide Ladd bands (Fig.  63.16a), mobilize
and straighten the duodenum, and exclude any
intrinsic duodenal obstruction (which may
involve passing a large bore gastric tube via
the nose or mouth through the stomach and
into the distal duodenum).
3. Widen the base of the small bowel mesentery
by carefully dividing overlying peritoneal
adhesions (Fig. 63.16b).
Fig. 63.14  Chronic midgut volvulus in a 3-year-old child
with recurrent abdominal pain and vomiting. Note the
thickened small bowel mesentery and patchy venous
congestion
862 M. D Stringer and P. R. Mishra

Fig. 63.15 Mesenteric
chylous congestion in a
5-week-old infant with
intestinal malrotation
and volvulus after
derotation of the
volvulus. Note the high
midline cecum. The
child’s head is to the left
of the picture

a b

Fig. 63.16 (a) The division of Ladd bands which extend peritoneal bands and the anterior peritoneal layer crossing
from the cecum and ascending colon across the duodenum the superior mesenteric vessels. (Reproduced from  Spitz
to the posterior peritoneum below the liver. (b) Splaying L. Malrotation. In Newborn Surgery. second Edition. Puri
the root of the small bowel mesentery by dividing fibrous P (ed.), Arnold, London, 2003, p. 438, Figs. 61.7 and 61.9)

4. Position the small bowel on the right side of appendix now lie in the left upper quadrant,
the abdomen and the colon on the left so that most surgeons perform an appendicectomy
the duodenojejunal flexure and ileocecal junc- (by excision or inversion), although this was
tion are wide apart. Because the cecum and not originally described by Ladd (1936) and a
63  Intestinal Malrotation
Fig. 63.17  Patchy small bowel infarction visible after
derotation of the intestine in an infant with midgut volvu-
lus associated with biliary atresia
case can be made for retaining the appendix
provided the family are aware of its abnormal
location (Davidson et al. 2017).
If there is an acute midgut volvulus, the viabil-
ity of the derotated bowel must be carefully
assessed (Fig.  63.17). The untwisted bowel is
covered with a warm moist pack for several min-
utes before assessing viability. As much bowel as
possible should be preserved, but the clearly
infarcted gut is resected. The decision regarding
primary intestinal anastomosis or stoma forma-
tion will depend on the condition and length of
the residual bowel and the stability of the infant.
If there are long segments of small bowel with
dubious viability, a planned second-look laparot-
omy 24–48 h after derotation is a useful strategy.
In some cases, the entire midgut is infarcted
when traditionally the choices have been limited
to resection and long-term management of intes-
tinal failure (including the consideration of even-
tual small bowel transplantation) or palliative
care only. However, in the last decade, two
reports have described successful restoration of
intestinal perfusion by thrombolysis after midgut
volvulus complicated by severe intestinal isch-
emia. Kiely et  al. (2012) documented two neo-
nates with severe intestinal ischemia due to
volvulus treated successfully by derotation, digi-
tal massage of the mesentery, and systemic
thrombolysis. Tissue-type plasminogen activator
(tPA) was given systemically by infusion starting
6  h postoperatively and continued for 18  h in
863
combination with fresh frozen plasma (to increase
plasminogen and potentiate the action of tPA).
Low-dose anticoagulation with unfractionated
heparin was also given for 24 h. The resolution of
intestinal ischemia was noted at a “second-look”
laparotomy 36–48 h later. Both infants recovered
well, but one required resection of a jejunal stric-
ture 6 weeks later. Messaoudi et al. (2015) used
systemic thrombolysis 24  h after a second-look
laparotomy at which 50 cm of necrotic jejunum
was resected, but the ischemic ileum was left in
situ. At a third laparotomy 48 h later, the remain-
ing small bowel was reperfused, and the infant
progressed to a full recovery. No serious bleeding
complications occurred in either report.
Laparoscopic surgery can be used success-
fully in elective cases without acute midgut vol-
vulus and in the authors' practice is particularly
applicable to older children (Fig.  63.18).
Retrospective studies of selected patient cohorts
suggest that laparoscopic procedures may be
associated with less postoperative pain, earlier
return of bowel function, shorter hospital stay,
improved cosmesis, and a reduced incidence of
adhesive bowel obstruction although operative
times may be longer (Graziano et  al. 2015;
Catania et al. 2016; Ooms et al. 2016). However,
these purported advantages are controversial
(Isani et  al. 2018; Svetanoff et  al. 2020).
Conversion to an open operation is required if the
Ladd procedure cannot be completed safely or
adequately, which is more likely in neonates and
in those with midgut volvulus (Reddy et al. 2018;
Arnaud et al. 2019; Svetanoff et al. 2020). There
are some concerns about an increased risk of
recurrent volvulus after the laparoscopic Ladd
procedure (3–5% vs 0.5–1.4% after open sur-
gery), as the more pronounced adhesions induced
by open surgery may assist fixation of the surgi-
cally widened base of the small bowel mesentery
(Catania et al. 2016; Zhu et al. 2019).
Reverse rotation and mesocolic hernias are
also corrected by dividing peritoneal attachments
to free up the bowel. Elements of the Ladd proce-
dure are necessary to minimize any future risk of
volvulus. Care should always be taken to avoid
injury to mesenteric vessels during these
dissections.
864
a b
Fig. 63.18 (a) Ladd bands in a 12-year-old girl with
severe developmental delay secondary to a chromosomal
disorder. Intestinal malrotation had been discovered at
laparoscopy for a tubo-ovarian abscess, and she under-
went a laparoscopic Ladd procedure a few weeks later.
Note the Ladd bands extending from the centrally posi-
tioned cecum across the duodenum to the gallbladder and
the small bowel lying on the right of the abdomen. (b)
63.11 Complications
The outcome of children operated on for intesti-
nal malrotation has steadily improved. Once the
postoperative ileus has resolved, the vast major-
ity recover uneventfully and have no further
symptoms. A few have one or more of the follow-
ing specific complications:
Mortality secondary to intestinal infarction
accounts for most of the 2–10% overall mor-
tality associated with surgery for midgut vol-
vulus secondary to intestinal malrotation
(Rescorla et al. 1990; Malek and Burd 2005).
Some of these patients have residual short
bowel syndrome and require chronic paren-
teral nutrition with its attendant risks and the
prospect of small bowel transplantation. Delay
in recognizing and treating midgut volvulus is
a major adverse determinant of outcome.
Recurrent midgut volvulus: This occurs in fewer
than 2% of patients after open surgery (Raitio
et al. 2016; Zhu et al. 2019) and is either due to
a failure to carry out the steps of the Ladd pro-
cedure adequately or to a lack of sufficient post-
operative adhesion formation. Although fixation
M. D Stringer and P. R. Mishra
Laparoscopic view of a widened base of the small bowel
mesentery (dashed line) extending from the duodenum on
the left of the image to the cecum and site of appendicec-
tomy (arrow). The mesenteric vessels are clearly visible in
this 13-year-old girl who was found to have intestinal mal-
rotation on a background of recurrent acute abdominal
pain
procedures have been devised to overcome the
latter, these are rarely if ever necessary if the
Ladd procedure is performed correctly.
Adhesive small bowel obstruction: This is a rela-
tively common and often underestimated risk,
occurring in approximately 5–15% of patients
after an open Ladd procedure (Murphy and
Sparnon 2006; McVay et al. 2007; El-Gohary
et al. 2010). Parents should be warned of this
complication.
Intestinal dysmotility: A minority of patients have
ongoing gastrointestinal symptoms after suc-
cessful surgery for malrotation, and intestinal
dysmotility appears to be the cause. These
symptoms may include constipation, diarrhea,
abdominal pain, and vomiting and occur more
often in older children, those with “atypical”
malrotation, and in those in whom the primary
indication for surgery was not volvulus
(Coombs et  al. 1991; McVay et  al. 2007;
Nehra and Goldstein 2011; Dekonenko et al.
2019).
Intussusception: Postoperative small bowel
intussusception is a rare cause of intestinal
obstruction within the first week or two after
surgery.
63  Intestinal Malrotation 865

Rarer complications include chylous ascites et  al. 2020). Most of these patients have non-­
(from widening the base of the small bowel mes- rotation rather than typical malrotation and there-
entery), wound infection, and prolonged postop- fore a low risk of midgut volvulus. In addition,
erative ileus. intra-abdominal adhesions that develop after sur-
gical closure may help to stabilize the gut.
Despite this recommendation, it is important to
63.12 Asymptomatic Intestinal recognize that there is a range of anatomic
Malrotation in Special arrangements of the gut after repair of CDH, and
Circumstances: Heterotaxy if the gut appears typically malrotated or kinked,
Syndrome, Congenital then corrective surgery is appropriate. Regardless
Diaphragmatic Hernia, of whether a Ladd procedure is performed, fami-
and Anterior Abdominal Wall lies should be informed about recognizing poten-
Defects tial symptoms of intestinal obstruction and
midgut volvulus.
Heterotaxy syndrome is characterized by abnor- The management of abnormalities of intesti-
mal lateralization of abdominal and thoracic nal rotation associated with anterior abdominal
organs together with isomerism of the atrial wall defects has also been subjected to scrutiny,
appendages. The condition is also known as situs attempting to define the future risk of volvulus in
ambiguus because the internal viscera are untreated cases compared to the long-term risks
arranged in a pattern somewhere between situs after a prophylactic Ladd procedure. As with
solitus and situs inversus. Bilateral right- or left-­ CDH, most affected individuals have non-­
sided situs ambiguus denotes bilaterally sym- rotation rather than typical malrotation and are
metrical atria and bronchopulmonary anatomy therefore at low risk of midgut volvulus. This is
and is classified as right or left isomerism. particularly true of gastroschisis where non-­
Affected individuals often have complex congen- rotation does not require intervention
ital heart disease, which is the main prognostic (Abdelhafeez et al. 2015; Ward et al. 2017). For
determinant. Caval abnormalities, asplenia (with patients with omphalocele, some studies have
right atrial isomerism) or polysplenia (with left highlighted the risk of midgut volvulus and advo-
atrial isomerism), and abnormal intestinal rota- cated a prophylactic Ladd procedure (Wakhlu
tion are frequently present. There has long been and Wakhlu 2000; Abdelhafeez et  al. 2015;
debate about the management of intestinal rota- Fawley et al. 2017), while more recent analyses
tional abnormalities in this disorder. In particular, have suggested that such intervention actually
is surgical correction warranted to prevent mid- increases the risk of volvulus and is harmful
gut volvulus, or is an expectant approach safe? (Ward et al. 2017; Lauriti et al. 2019).
There is now general agreement that symptomatic
patients with upper gastrointestinal obstructive
symptoms should undergo imaging and correc- 63.13 Conclusions
tive surgery by the Ladd procedure, while asymp- and Controversies
tomatic patients can be safely managed
expectantly, reserving prompt investigation and Intestinal malrotation is commonly encountered
treatment if symptoms develop (Cullis et  al. by pediatric surgeons. Bilious vomiting in
2016; Ryerson et al. 2018; White et al. 2018). infancy demands appropriate investigation and
A similar conclusion that a prophylactic Ladd exclusion of malrotation. Children with typical
procedure is unnecessary has emerged for intestinal malrotation are at risk of life-­
abnormalities of intestinal rotation associated threatening midgut volvulus and should be diag-
with congenital diaphragmatic hernia (CDH) nosed and treated promptly. The Ladd procedure
(Graziano et al. 2015; Ward et al. 2017; Heiwegen is an effective operation for treating the condi-
866
tion but carries a long-term risk of adhesive
intestinal obstruction. Several controversies and
uncertainties remain: the importance of Ladd
bands in causing duodenal obstruction; the opti-
mum management of incomplete malrotation;
and the best strategy for treating infants with
midgut infarction.
References
Abdelhafeez AH, Schultz JA, Ertl A, Cassidy LD, Wagner
AJ (2015) The risk of volvulus in abdominal wall
defects. J Pediatr Surg 50(4):570–572. https://doi.
org/10.1016/j.jpedsurg.2014.12.017
Aboagye J, Goldstein SD, Salazar JH, Papandria
D, Okoye MT, Al-Omar K, Stewart D, Lukish J,
Abdullah  F (2014) Age at presentation of common
pediatric surgical conditions: Reexamining dogma. J
Pediatr Surg. 49(6):995-9. https://doi.org/10.1016/j.
jpedsurg.2014.01.039
Arnaud AP, Suply E, Eaton S, Blackburn SC, Giuliani S,
Curry JI, Cross KM, De Coppi P (2019) Laparoscopic
Ladd’s procedure for malrotation in infants and
children is still a controversial approach. J Pediatr
Surg 54(9):1843–1847. https://doi.org/10.1016/j.
jpedsurg.2018.09.023
Bartholmot C, Faure JM, Grosjean F, Couture A, Forgues
D, Fuchs F, Prodhomme O (2019) Prenatal diagno-
sis of antenatal midgut volvulus: specific ultrasound
features. Prenat Diagn 39(1):16–25. https://doi.
org/10.1002/pd.5392
Biyyam DR, Dighe M, Siebert JR (2009) Antenatal diag-
nosis of intestinal malrotation on fetal MRI.  Pediatr
Radiol 39(8):847–849. https://doi.org/10.1007/
s00247-­009-­1226-­5
Bourgouin S, Schlienger G, Deledalle FX, Thouard H
(2017) Discordant intestinal malrotation in identi-
cal twins. ANZ J Surg 87(9):E110–E111. https://doi.
org/10.1111/ans.13013
Brereton RJ, Taylor B, Hall CM (1986) Intussusception
and intestinal malrotation in infants: Waugh’s syn-
drome. Br J Surg 73(1):55–57. https://doi.org/10.1002/
bjs.1800730123
Catania VD, Lauriti G, Pierro A, Zani A (2016) Open ver-
sus laparoscopic approach for intestinal malrotation
in infants and children: a systematic review and meta-­
analysis. Pediatr Surg Int 32(12):1157–1164. https://
doi.org/10.1007/s00383-­016-­3974-­2
Coe TM, Chang DC, Sicklick JK (2015) Small bowel
volvulus in the adult populace of the United States:
results from a population-based study. Am J Surg
210(2):201–210.e2. https://doi.org/10.1016/j.
amjsurg.2014.12.048
Coombs RC, Buick RG, Gornall PG, Corkery JJ, Booth
IW (1991) Intestinal malrotation: the role of small
intestinal dysmotility in the cause of persistent symp-
M. D Stringer and P. R. Mishra
toms. J Pediatr Surg 26(5):553–556. https://doi.
org/10.1016/0022-­3468(91)90705-­x
Cullis PS, Siminas S, Losty PD (2016) Is screening of
intestinal foregut anatomy in heterotaxy patients
really necessary?: a systematic review in search of the
evidence. Ann Surg 264(6):1156–1161. https://doi.
org/10.1097/SLA.0000000000001563
Davidson JR, Eaton S, De Coppi P (2017) Let sleeping
dogs lie: to leave the appendix at the time of a Ladd
procedure. J Pediatr Surg S0022-3468(17):30570–
30575. https://doi.org/10.1016/j.jpedsurg.2017.09.003
Dekonenko C, Sujka JA, Weaver K, Sharp SW, Gonzalez
K, St Peter SD (2019) The identification and treatment
of intestinal malrotation in older children. Pediatr
Surg Int 35(6):665–671. https://doi.org/10.1007/
s00383-­019-­04454-­9
Devane SP, Coombes R, Smith VV, Bisset WM, Booth
IW, Lake BD, Milla PJ (1992) Persistent gastrointes-
tinal symptoms after correction of malrotation. Arch
Dis Child 67(2):218–221. https://doi.org/10.1136/
adc.67.2.218
Drewett M, Johal N, Keys C, Hall J, N, Burge D. (2016)
The burden of excluding malrotation in term neonates
with bile stained vomiting. Pediatr Surg Int 32(5):483–
486. https://doi.org/10.1007/s00383-­016-­3877-­2
El-Gohary Y, Alagtal M, Gillick J (2010) Long-term
complications following operative intervention for
intestinal malrotation: a 10-year review. Pediatr
Surg Int 26(2):203–206. https://doi.org/10.1007/
s00383-­009-­2483-­y
Fawley JA, Abdelhafeez AH, Schultz JA, Ertl A, Cassidy
LD, Peter SS, Wagner AJ (2017) The risk of midgut
volvulus in patients with abdominal wall defects: a
multi-institutional study. J Pediatr Surg 52(1):26–29.
https://doi.org/10.1016/j.jpedsurg.2016.10.014
Filston HC, Kirks DR (1981) Malrotation– the ubiqui-
tous anomaly. J Pediatr Surg 16(4 Suppl 1):614–620.
https://doi.org/10.1016/0022-­3468(81)90015-­4
Forrester MB, Merz RD (2003) Epidemiology of
intestinal malrotation, Hawaii, 1986-99. Paediatr
Perinat Epidemiol 17(2):195–200. https://doi.
org/10.1046/j.1365-­3016.2003.00480
Godbole P, Stringer MD (2002) Bilious vomiting in
the newborn: how often is it pathologic? J Pediatr
Surg 37(6):909–911. https://doi.org/10.1053/
jpsu.2002.32909
Graziano K, Islam S, Dasgupta R, Lopez ME, Austin
M, Chen LE, Goldin A, Downard CD, Renaud E,
Abdullah F (2015) Asymptomatic malrotation: diag-
nosis and surgical management: an American pedi-
atric surgical association outcomes and evidence
based practice committee systematic review. J Pediatr
Surg 50(10):1783–1790. https://doi.org/10.1016/j.
jpedsurg.2015.06.019
Heiwegen K, de Blaauw I, van Ling J, Botden SMBI (2020)
Malrotation in congenital diaphragmatic hernia: is it
really a problem? Eur J Pediatr Surg 30(5):434–439.
https://doi.org/10.1055/s-­0039-­1692165
Hsiao M, Langer JC (2011) Value of laparoscopy in chil-
dren with a suspected rotation abnormality on imag-
63  Intestinal Malrotation
ing. J Pediatr Surg 46(7):1347–1352. https://doi.
org/10.1016/j.jpedsurg.2010.12.008
Husberg B, Salehi K, Peters T, Gunnarsson U, Michanek
M, Nordenskjöld A, Strigård K (2016 Mar) Congenital
intestinal malrotation in adolescent and adult patients:
a 12-year clinical and radiological survey. Springerplus
1(5):245. https://doi.org/10.1186/s40064-­016-­1842-­0
Isani MA, Schlieve C, Jackson J, Elizee M, Asuelime
G, Rosenberg D, Kim ES (2018) Is less more?
Laparoscopic versus open Ladd’s procedure in chil-
dren with malrotation. J Surg Res 229:351–356.
https://doi.org/10.1016/j.jss.2018.04.016
Kantor JL (1934) Anomalies of the colon: their roentgen
diagnosis and clinical significance. Résumé of ten
years’ study. Radiology 23(6):651–662
Kiely EM, Pierro A, Pierce C, Cross K, De Coppi P
(2012) Clot dissolution: a novel treatment of midgut
volvulus. Pediatrics 129(6):e1601–e1604. https://doi.
org/10.1542/peds.2011-­2115
Kirby CP, Freeman JK, Ford WD, Davidson GP, Furness
ME (2000) Malrotation with recurrent volvulus pre-
senting with cholestasis, pruritus, and pancreati-
tis. Pediatr Surg Int 16(1–2):130–131. https://doi.
org/10.1007/s003830050040
Ladd WE (1936) Surgical diseases of the alimentary tract
in infants. N Engl J Med 215:705–708
Lauriti G, Miscia ME, Cascini V, Chiesa PL, Pierro A,
Zani A (2019) Intestinal malrotation in infants with
omphalocele: a systematic review and meta-analysis. J
Pediatr Surg 54(3):378–382. https://doi.org/10.1016/j.
jpedsurg.2018.09.010
Lilien LD, Srinivasan G, Pyati SP, Yeh TF, Pildes RS
(1986) Green vomiting in the first 72 hours in normal
infants. Am J Dis Child 140(7):662–664. https://doi.
org/10.1001/archpedi.1986.02140210060026
Long L, Zhen C, Yandong W, Ning D, Qi L, Qing G
(2018) Congenital chylous ascites in infants: another
presentation of intestinal malrotation. J Pediatr
Surg 53(3):537–539. https://doi.org/10.1016/j.
jpedsurg.2017.05.031
Malek MM, Burd RS (2005) Surgical treatment of mal-
rotation after infancy: a population-based study. J
Pediatr Surg 40(1):285–289. https://doi.org/10.1016/j.
jpedsurg.2004.09.028
Martin V, Shaw-Smith C (2010) Review of genetic factors
in intestinal malrotation. Pediatr Surg Int 26(8):769–
781. https://doi.org/10.1007/s00383-­010-­2622-­5
Martinez-Leo B, Chesley P, Alam S, Frischer JS, Levitt
MA, Avansino J, Dickie BH (2016) The association of
the severity of anorectal malformations and intestinal
malrotation. J Pediatr Surg 51(8):1241–1245. https://
doi.org/10.1016/j.jpedsurg.2016.04.008
McVay MR, Kokoska ER, Jackson RJ, Smith SD (2007)
Jack Barney Award. The changing spectrum of intes-
tinal malrotation: diagnosis and management. Am
J Surg 194(6):712–717. https://doi.org/10.1016/j.
amjsurg.2007.08.035
Mehall JR, Chandler JC, Mehall RL, Jackson RJ, Wagner
CW, Smith SD (2002) Management of typical and atyp-
867
ical intestinal malrotation. J Pediatr Surg 37(8):1169–
1172. https://doi.org/10.1053/jpsu.2002.34465
Messaoudi N, Vanlede K, Vervloessem D, Huyghe M,
Leyman P (2015) Thrombolysis for severe intesti-
nal ischemia due to midgut volvulus in a neonate.
Eur J Pediatr Surg Rep 3(2):74–77. https://doi.
org/10.1055/s-­0035-­1558460
Mirjalili SA, Tarr G, Stringer MD (2017) The length of
the large intestine in children determined by computed
tomography scan. Clin Anat 30(7):887–893. https://
doi.org/10.1002/ca.22941
Mishra PR, Stringer MD (2021) Intestinal malrotation in
extremely premature infants: a potential trap. Pediatr
Surg Int. 37(11):1607–1612. https://doi.org/10.1007/
s00383-021-04969-0
Mohinuddin S, Sakhuja P, Bermundo B et al (2015 Jan)
Outcomes of full-term infants with bilious vomit-
ing: observational study of a retrieved cohort. Arch
Dis Child 100(1):14–17. https://doi.org/10.1136/
archdischild-­2013-­305724
Murphy FL, Sparnon AL (2006) Long-term complications
following intestinal malrotation and the Ladd’s proce-
dure: a 15 year review. Pediatr Surg Int 22(4):326–
329. https://doi.org/10.1007/s00383-­006-­1653-­4
Nehra D, Goldstein AM (2011) Intestinal malrotation:
varied clinical presentation from infancy through
adulthood. Surgery 149(3):386–393. https://doi.
org/10.1016/j.surg.2010.07.004
Ooms N, Matthyssens LE, Draaisma JM, de Blaauw I,
Wijnen MH (2016) Laparoscopic treatment of intestinal
malrotation in children. Eur J Pediatr Surg 26(4):376–
381. https://doi.org/10.1055/s-­0035-­1554914
Orzech N, Navarro OM, Langer JC (2006) Is ultra-
sonography a good screening test for intestinal
­malrotation? J Pediatr Surg 41(5):1005–1009. https://
doi.org/10.1016/j.jpedsurg.2005.12.070
Pachl M, Eaton S, Kiely EM, Drake D, Cross K, Curry JI,
Pierro A, DeCoppi P (2015) Esophageal atresia and mal-
rotation: what association? Pediatr Surg Int 31(2):181–
185. https://doi.org/10.1007/s00383-­014-­3641-­4
Powell DM, Othersen HB, Smith CD (1989) Malrotation
of the intestines in children: the effect of age on pre-
sentation and therapy. J Pediatr Surg 24(8):777–780.
https://doi.org/10.1016/s0022-­3468(89)80535-­4
Raitio A, Green PA, Fawkner-Corbett DW, Wilkinson DJ,
Baillie CT (2016) Malrotation: age-related differences
in reoperation rate. Eur J Pediatr Surg 26(1):34–37.
https://doi.org/10.1055/s-­0035-­1563677
Reddy AS, Shah RS, Kulkarni DR (2018) Laparoscopic
Ladd’s procedure in children: challenges, results, and
problems. J Indian Assoc Pediatr Surg 23(2):61–65.
https://doi.org/10.4103/jiaps.JIAPS_126_17
Rescorla FJ, Shedd FJ, Grosfeld JL, Vane DW, West
KW (1990) Anomalies of intestinal rotation in child-
hood: analysis of 447 cases. Surgery 108(4):710–715.
Discussion 715–6. PMID: 2218883
Ryerson LM, Pharis S, Pockett C, Soni R, Fruitman D,
Guleserian KJ, Nater M, Raynor SC, Mackie AS,
Dicken B (2018) Heterotaxy syndrome and intestinal
868
rotation abnormalities. Pediatrics 142(2):e20174267.
https://doi.org/10.1542/peds.2017-­4267
Salehi Karlslätt K, Pettersson M, Jäntti N et al (2019) Rare
copy number variants contribute pathogenic alleles
in patients with intestinal malrotation. Mol Genet
Genomic Med 7(3):e549. https://doi.org/10.1002/
mgg3.549
Sizemore AW, Rabbani KZ, Ladd A, Applegate KE (2008)
Diagnostic performance of the upper gastrointestinal
series in the evaluation of children with clinically
suspected malrotation. Pediatr Radiol 38(5):518–528.
https://doi.org/10.1007/s00247-­008-­0762-­8
Sloan K, Alzamrooni A, Stedman FE, Ron O, Hall NJ
(2020) Diagnostic laparoscopy to exclude malrotation
following inconclusive upper gastrointestinal contrast
study in infants. Pediatr Surg Int 36(10):1221–1225.
https://doi.org/10.1007/s00383-­020-­04729-­6
Spigland N, Brandt ML, Yazbeck S (1990) Malrotation
presenting beyond the neonatal period. J
Pediatr Surg 25(11):1139–1142. https://doi.
org/10.1016/0022-­3468(90)90749-­y
Stephens LR, Donoghue V, Gillick J (2012) Radiological
versus clinical evidence of malrotation, a tortuous
tale–10-year review. Eur J Pediatr Surg 22(3):238–
242. https://doi.org/10.1055/s-­0032-­1308707
Stewart DR, Colodny AL, Daggett WC (1976) Malrotation
of the bowel in infants and children: a 15 year review.
Surgery 79(6):716–720. PMID: 1273757
Svetanoff WJ, Sobrino JA, Sujka JA, St Peter SD,
Fraser JD (2020) Laparoscopic Ladd procedure
M. D Stringer and P. R. Mishra
for the Management of Malrotation and Volvulus.
J Laparoendosc Adv Surg Tech A 30(2):210–215.
https://doi.org/10.1089/lap.2019.0602
Wakhlu A, Wakhlu AK (2000) The management of
exomphalos. J Pediatr Surg 35(1):73–76. https://doi.
org/10.1016/s0022-­3468(00)80017-­2
Waldhausen JH, Sawin RS (1996) Laparoscopic
Ladd’s procedure and assessment of malrotation. J
Laparoendosc Surg 6(Suppl 1):S103–S105. PMID:
8832939
Ward EP, Wang A, Thangarajah H, Lazar D, Bickler S,
Fairbanks T, Kling KM (2017) Preemptive Ladd pro-
cedure in congenital diaphragmatic hernia and abdom-
inal wall defects does not reduce the risk of future
volvulus. J Pediatr Surg 52(12):1956–1961. https://
doi.org/10.1016/j.jpedsurg.2017.08.067
White SC, Dean PN, McGahren ED, Gangemi JJ, Vergales
J (2018) Malrotation is not associated with adverse
outcomes after cardiac surgery in patients with het-
erotaxy syndrome. J Pediatr Surg 53(8):1494–1498.
https://doi.org/10.1016/j.jpedsurg.2018.01.008
Zhang W, Sun H, Luo F (2017) The efficiency of
sonography in diagnosing volvulus in neonates
with suspected intestinal malrotation. Medicine
(Baltimore) 96(42):e8287. https://doi.org/10.1097/
MD.0000000000008287
Zhu H, Zheng S, Alganabi M, Peng X, Dong K, Pierro A,
Shen C (2019) Reoperation after Ladd’s procedure in
the neonatal period. Pediatr Surg Int 35(1):117–120.
https://doi.org/10.1007/s00383-­018-­4382-­6
 Jejuno-Ileal Atresia
A. J. W. Millar, S. Cox, and A. Numanoglu
64.1 Introduction
Mural defects causing anatomical discontinuity
or narrowing of the small bowel can morphologi-
cally be divided into either atresia or stenosis and
cause about a third of all cases of neonatal intes-
tinal obstruction. Atresia refers to the complete
occlusion of the intestinal lumen and accounts
for more than 90% of cases, while stenosis is
defined as a partial intra-luminal occlusion with
incomplete obstruction often resulting in delay in
diagnosis until outside the neonatal period. There
is a tendency to be born prematurely or small for
dates. Other congenital anomalies outside the
gastrointestinal tract are uncommon. In a small
number of cases, a genetic cause has been identi-
fied. Antenatal and early postnatal diagnoses,
along with improvements in neonatal care, have
been important in the current overall survival of
over 90% of newborns with jejunoileal atresia in
most centres. However, the change in surgical
technique from an anastomosis of the blind-­
ending bulbous proximal atresia to the distal
bowel, to resection of the dilated bowel and pri-
mary end to end anastomosis, has been the prime
factor in the dramatic improvement in survival. It
A. J. W. Millar (*) · S. Cox · A. Numanoglu
University of Cape Town and Red Cross War
Memorial Children’s Hospital,
Cape Town, South Africa
e-mail: alastair.millar@uct.ac.za; sharon.cox@uct.ac.za;
alp.numanoglu@uct.ac.za
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_64
64
was Louw, Nixon and Benson who led the way,
all changing their technique more or less over the
same time period but for different reasons, which
led to one of the great success stories of paediat-
ric surgery (Benson 1955; Louw 1959; Nixon
1960). Additional innovative surgical techniques
to maximize the utilization of short residual
bowel length have further improved long-term
outcomes (Bianchi 1980; Kimura et  al. 1996;
Kim et al. 2003).
64.2 Historical Overview
Significant changes in surgical techniques and
outcome of intestinal atresia and stenosis have
occurred since Spriggs, in 1912, postulated that
strangulation of a segment of foetal gut was most
likely the causative factor producing intestinal
atresia (Spriggs 1912). Clinical and morbid ana-
tomical data were presented by Louw 1952 and
subsequently proven by Barnard and Louw in
canine experiments, which produced anomalies
identical to congenital intestinal atresia in humans
(Louw 1952; Louw and Barnard 1955; Louw
et al. 1981a, b). The clarification of the vascular
hypothesis and realization that the proximal
dilated bowel remained with dysfunctional peri-
stalsis directly influenced and brought the change
to the surgical approach from previous attempts at
surgical management, which included anastomo-
sis of the bulbous blind end to the distally col-
869
870 A. J. W. Millar et al.

lapsed bowel, exteriorization of both ends as an 64.3 Incidence

initial stomas, Bishop-Koop, Santulli, or side-to-
side anastomosis, to liberal back resection of the
proximal blind-ending bulbous atretic bowel to
the bowel of near normal diameter and primary
end-to-end anastomosis of proximal and distal
bowel with more equal lumen diameter (Benson
1955; Louw 1959; Nixon 1960; Nygaard 1967;
Grosfeld et  al. 1979). This change in technique
during the mid-1950s resulted in the elimination
of blind-loop syndromes and anastomotic dys-
function with an immediate reduction in m­ ortality
rates from 69% to 33% at Great Ormond Street,
London, and 90–28% at the Red Cross War
Memorial Children’s Hospital, Cape Town (Louw
1966). Subsequent technical advances, improve-
ment in neonatal care, anaesthesiology and nutri-
tional support have further impacted upon and
improved current day survival to over 90% (Tables
64.1 and 64.2) (Cywes et al. 1982).
Jejunoileal atresia has a prevalence rate of
approximately 0.54–1.1 per 10,000 live births,
with a third of infants either born prematurely or
small for dates (Best et  al. 2012). The male-to-­
female ratio is equal in most reports. Hereditary
forms and familial patterns of atresia are excep-
tionally rare and may be on a basis of autosomal
recessive or dominant transmission. A genetic
basis, however, has been established for types III
(b) and IV multiple atresias (Mishalany and
Najjar 1968; Puri and Fujimoto 1988; Lambrecht
and Kluth 1998). Associated chromosomal and
extra-abdominal anomalies (7%) are well docu-
mented but not as common as in duodenal
atresias.
64.4 Etiopathogenesis

Table 64.1  Types of 351 cases with intestinal atresia A localized intrauterine vascular accident can
seen from 1959 to 2017 at Red Cross War Memorial cause ischaemic necrosis, liquefaction of tissues
Children’s Hospital, Cape Town, South Africa and subsequent resorption of the affected devital-
Type Jejunum Ileum Total (%) (%)(%)(5)(%) ized segment(s). The ischaemia hypothesis is
Stenosis 21 13 34 (9) supported by evidence of incarceration or snaring
Type I 64 20 84 (24) of bowel in an exomphalos or gastroschisis and
Type II 20 13 33 (9)
foetal events such as intussusception, midgut vol-
Type III(a) 28 24 52 (15)
Type III(b) 65 0 65 (19) vulus, trans-mesenteric internal herniation and
Type IV 69 14 83 (24) thromboembolic occlusion resulting in atresia.
Total 267 84 351 Meconium ileus and Hirschsprung’s disease
should also be considered as possible underlying
etiological factors in ileal atresia (Santulli and
Table 64.2 Chronological improvement in outcomes
with changes in surgical techniques, as well asimproved
Blanc 1961).
anaesthetic and postoperative care of neonates Additional hypotheses based on careful clini-
Authors Years of study N Survival (%) cal observations, morphological studies and
Evans 1950 1498 9.3 experimentation include Tandler’s concept of
Gross 1940–1952 71 51 failure of recanalization of the solid cord stage of
Benson et al. 1945–1959 38 55 intestinal development, obliterative embryologi-
De Lorimer 1957–1966 587 65 cal events at Meckel’s point with excessive
Nixon and Tawes 1956–1967 62 62 resorption of the vitelline duct with adjacent
Louw 1959–1967 33 94
ileum, epithelial occlusions and foetal inflamma-
Martin and Zerella 1957–1975 59 64
Cywes et al. 1959–1978 84 88
tory diseases. The localized nature of the defect
Danismead et al. 1967–1981 101 77 would explain the low incidence of coexisting
Smith and Glasson 1961–1986 84 61 abnormalities of the extra-abdominal organs.
Vecchia et al. 1972–1997 128 84 Jejunoileal atresias have rarely been described in
Rode et al. 1959–2007 318 92 identical twins. A rare autosomal recessive pat-
1990–2007 115 94 tern of transmission has been documented, and
64  Jejuno-Ileal Atresia 871

pathological findings could support the concept
that a developmental process early on could have
affected the whole bowel. This occurrence is
termed hereditary multiple intestinal atresia
(HMIA) and is associated with combined immu-
nodeficiency and morphological alterations in
other organs such as the thymus, lung, spleen and
liver, leading to a poor prognosis and is quite
different to the previously described atresias
­
(Lambrecht and Kluth 1998).
64.5 Pathophysiology
With the development of the obstruction in the
developing foetus, the proximal bowel becomes
dilated as a result of onward peristalsis and pres-
sure of luminal contents (Nixon and Tawes 1971).
This dilatation may extend proximally for a vari-
able length and in jejunal atresias may extend into
the duodenum as well. The distal bowel is usually
collapsed and wormlike in appearance. The histo-
pathology of the bowel is usually normal apart
from some attenuation of muscle and nerves in
the proximal dilated segment. Secondary isch-
aemia, necrosis and bowel perforation may occur
at the atresia either from raised intraluminal pres-
sure from postnatal swallowed air and liquids or
from volvulus of the bulbous blind end. Thus, a
delay in diagnosis may contribute to morbidity
and was a frequent cause of mortality in historical
series (Louw et al. 1981a, b).
occurs at the site of the obstruction where the
bowel is often dysperistaltic and of questionable
viability, especially when treatment is delayed. In
rare instances, jejunoileal atresias have been
found to coexist with oesophageal, duodenal,
colonic or rectal atresias. The incidence of the
different types seen in 351 at our centre over a
58-year period is shown in Table 64.1.
• Stenosis (<9%) is characterized by a short
localized narrowing of the bowel without dis-
continuity or a mesenteric defect. Although
more frequently seen with duodenal fenes-
trated webs a ‘wind-sock’ effect can be cre-
ated when increased intraluminal pressure in
the proximal bowel bulges the membrane into
the distally collapsed bowel creating a conical
transition zone. The bowel is of normal length.
• Atresia type I (24%) is represented by a trans-
luminal septum or short atretic segment. The
dilated proximal bowel remains in continuity
with the collapsed distal bowel, there is no
mesenteric defect, and the bowel is of normal
length (Fig. 64.1).

64.6 Pathology

The morphological classification of jejunoileal

atresia into types I–IV has significant prognostic
and therapeutic implications (Martin and Zerella
1976; Grosfeld et al. 1979). The level of the most
proximal atresia determines whether it is classi-
fied as jejunal or ileal. Although single atresias
are most commonly encountered, 6–12% of
infants will have multiple atretic segments, and
up to 5% may have an additional colonic atresia.
The appearance of the atretic segment is deter-
mined by the type of occlusion, but in all cases a
maximum dilatation of the proximal bowel Fig. 64.1  Typical clinical picture of jejunal atresia type I
872
• Atresia type II (9%) has two blind-ending
atretic ends connected by a fibrous cord along
the edge of the mesentery. There is no
­mesenteric defect and the bowel length is not
foreshortened.
• Atresia type IIIa (15%) is similar to type II
except that the fibrous connecting cord is
absent and there is a V-shaped mesenteric
defect. The bowel length may be foreshort-
ened. Cystic fibrosis is commonly associated
with this variety.
• Atresia type IIIb (19%) (apple peel or
Christmas tree) consists of a proximal jejunal
atresia often with associated malrotation,
absence of most of the superior mesenteric
artery and a large mesenteric defect. The distal
bowel is coiled in a helical configuration
around a single perfusing artery arising from
the right colic arcades(Fig. 64.2). Additional
type I or type II atresias may be found in the
proximal and distal bowel. There is always a
significant reduction in intestinal length. A
familial incidence and atresias amongst sib-
lings and identical twins point to a more com-
plex genetic transmission with an overall
recurrence rate of 18% in subsequent siblings
born to the same parents.
Fig. 64.2  Type IIIb (apple peel) atresia
A. J. W. Millar et al.
• Atresia type IV (24%) represents multiple seg-
mental atresias like a string of sausages or a
combination of types I–III.  Bowel length is
always reduced. The terminal ileum is usually
spared; up to 25 atretic sites have been encoun-
tered in a single patient.
64.7 Management
64.7.1 Clinical Presentation and
Diagnosis
A delay in diagnosis may lead to impairment of
bowel viability or frank necrosis and perforation,
fluid and electrolyte abnormalities and an
increase incidence of sepsis (Figs. 64.3 and 64.4).
Antenatal perforation of atretic bowel may pres-
ent at birth with meconium peritonitis. The dif-
ferentiation between atresia and other forms of
intrinsic and extrinsic bowel obstruction due to
Fig. 64.3  A gangrenous type IIIb (apple peel) jejunal
atresia due to volvulus at its base where the single mar-
ginal artery supplies ileum, which is coiled in an apple
peel configuration
64  Jejuno-Ileal Atresia 873

at or soon after birth; the more distal the obstruc-

tion, the more generalized the abdominal disten-
sion. Visible loops of bowel may be observed,
but peristaltic activity is an inconsistent finding.
Proximal jejunal atresia often presents with gas-
tric distension, one or two loops of visible bowel
in the upper abdomen decompressable by naso-
gastric tube aspiration in an otherwise gasless
abdomen. Although the classic first stools passed
by these patients are small, grey in colour and
mucoid, normal meconium can occasionally be
passed, which would suggest a late gestation
insult leading to the development of atresia.
Delay in detection, increasing intra-luminal
pressure and/or secondary torsion of the proxi-
mal atretic distended bowel can lead to ischemia,
perforation and peritonitis, precipitating abdom-
inal tenderness, with oedema and erythema of
the abdominal wall. One-third of infants can
have biochemically determined nonhaemolytic
Fig. 64.4  Abdominal radiograph of a newborn showing a
pneumoperitoneum due to proximal bowel perforation
jaundice.
The diagnosis is confirmed by radiological
examination of the abdomen and chest (Fig. 64.5).
volvulus or internal hernia is the most important
consideration that requires early diagnosis. Many
cases of intestinal atresia are now diagnosed pre-
natally by ultrasonographic investigation of the
foetus, showing dilated intestine with vigorous
peristalsis, suggesting obstruction, particularly
so in pregnancies complicated by third trimester
polyhydramnios (Wax et  al. 2006). Intestinal
atresia is also suspected in foetuses with gastros-
chisis and evidence of intestinal dilatation.
However, prenatal ultrasound has a relatively
poor predictive value for bowel abnormalities
(31–42%). MRI imaging may prove to be more
accurate in the prenatal diagnosis of bowel atre-
sia (Veyrac et al. 2004). A positive family history
will help identify hereditary forms. As the prog-
nosis of intestinal atresia is excellent, there is no
need for intrauterine intervention or early induc-
tion of labour before 38 weeks of gestation.
After birth, intestinal atresia or stenosis can
present early with large intragastric volumes at
(>25 ml gastric aspirate) followed by persistent
bile stained vomiting, although in 20% of chil-
Fig. 64.5  Abdominal radiograph of a newborn with jeju-
dren, symptoms may be delayed for more than nal atresia showing a few dilated proximal small bowel
24 h. Abdominal distension is frequently present loops
874
Erect and supine whole-body chest/abdominal
radiographs done after birth will reveal distended
air-filled small intestinal loops proximal to an
obstruction with no gas visible in the rectum. In
some instances, the first abdominal radiograph
can reveal a completely opaque abdomen due to
fluid-filled obstructed bowel. Emptying the stom-
ach via a nasogastric tube and the injection of a
bolus of air will demonstrate the obstruction
without resorting to a contrast radiography.
The more distal the obstruction, the greater
are the number of air- and fluid-filled, distended
loops of bowel (Fig.  64.6). The bowel immedi-
ately proximal to the site of obstruction may have
the appearance of a very large air- and fluid-filled
loop. A prone lateral view may distinguish
between distal small bowel and colonic obstruc-
tions. Occasionally scattered intra-peritoneal or
scrotal calcification or a large meconium pseudo-
cyst or hydrocele may be encountered or seen
radiographically, signifying intrauterine bowel
perforation, meconium spill with dystrophic
calcification.
Fig. 64.6  Abdominal radiograph of a newborn with dis-
tal small bowel stenosis showing large air-filled intestinal
loops
A. J. W. Millar et al.
In the presence of a radiologically determined
complete obstruction, a contrast enema can be
performed to confirm the level of obstruction
(small or large bowel), document the calibre of
the colon, exclude an associated colonic atresia
and locate the position of the caecum as an indi-
cation of malrotation (Fig.  64.7). Care must be
taken not to rupture the microcolon, as the latter
signifies an unused colon with the obstruction
located in the small bowel. Where the atresia is
formed late in intrauterine life, the bowel distal to
the atresia has a more normal calibre. With
incomplete proximal small bowel obstruction, an
upper gastrointestinal contrast study is indicated
to demonstrate the site and nature of the obstruc-
tion and to exclude midgut volvulus.
The clinical and radiological presentation of
jejunoileal stenosis will be determined by the
level and degree of stenosis. The diagnosis is
often delayed for several weeks to months, and
investigations may be inconclusive due to sub-
clinical symptoms and findings.
Fig. 64.7  Contrast enema confirming a normally rotated
and patent colon with evidence of proximally dilated
small bowel loops
64  Jejuno-Ileal Atresia
64.7.2 Differential Diagnosis
A spectrum of other diseases can present with
symptoms and signs of neonatal intestinal
obstruction closely mimicking jejunoileal atresia.
These include atresia of the distal duodenum,
colonic atresia and long-segment Hirschsprung’s
disease as well as those with intraluminal con-
tents causing obstruction such as the inspissated
mucous pellets found with meconium ileus and
hypoperistalsis as in small left colon syndrome
and meconium plug syndrome. Extrinsic factors
like midgut volvulus and incarcerated hernia and
nonmechanical (functional) causes such as ileus
from sepsis or early necrotizing enterocolitis
should also be considered.
64.7.3 Surgical Management:
Making the Infant ‘Safe
for Surgery’
These babies must be resuscitated and haemody-
namically stabilized before transfer. Preoperative
management is directed at optimizing the status of
the infant, which apart from the atresia could be
compromised by delayed diagnosis, hypothermia,
fluid and electrolyte imbalance, associated abnor-
malities, prematurity, obstetrical-related compli-
cations, systemic sepsis and compromised bowel.
Basic factors required include a warm humidified
environment, gastric decompression to prevent
aspiration, fluid management (maintenance,
replacement of deficits, and ongoing losses), cor-
rection of haematological and biochemical abnor-
malities and prophylactic antibiotics.
64.7.4 Anaesthesia
Neonates tolerate surgery well provided their
special needs, physiological limitations and dis-
ease processes are duly taken into account. The
major anaesthetic considerations are related to
prematurity, fluid and electrolyte homeostasis,
abdominal distension, the risk of aspiration and
associated life-threatening congenital anomalies.
Invasive monitoring is indicated in sick or unsta-
875
ble infants, and a central line may be required for
intravenous feeding postoperatively. The anaes-
thetic management is dictated by the condition of
the infant and the available facilities. Light gen-
eral and epidural anaesthesia may avoid the need
for postoperative ventilation.
64.7.5 Surgical Strategy
The operative management of intestinal atresia is
individualized and determined by the pathologic
findings, associated abnormalities, the length of
the undamaged bowel and the general condition
of the infant (Louw 1966; Millar and Cox 2020).
64.7.5.1 Standard Surgical Procedure
A supra-umbilical transverse incision provides
excellent exposure of the entire gastrointestinal
tract. A laparoscopic or periumbilical approach
may also be used with exteriorization of the
atretic segment through a small umbilical inci-
sion (Banieghbal and Beale 2007; Lima et  al.
2009; Murphy et al. 2009). The bowel is exterior-
ized to determine the site and type of atresia and
to exclude further downstream atresias or steno-
ses and associated lesions such as incomplete
rotation or meconium ileus. The appearance of
the atretic segment depends upon the type of
occlusion, but in all cases, maximum dilatation
and hypertrophy of the proximal bowel occurs at
the point of the first obstruction. This segment
often has reduced peristaltic activity and is of
questionable viability, whilst the bowel distal to
the obstruction is collapsed, tiny and wormlike in
appearance. The intestinal content in more proxi-
mal atresias is milked backwards into the stom-
ach from where it is aspirated. In the more distal
atresias, the intestinal contents can be milked into
the bulbous segment to be resected. If a volvulus
is present, it should be derotated. It is essential to
exclude other distal small and large bowel atre-
sias by injecting normal saline down the lumen of
the distal bowel. More than one atresia can occur
in up to 20% of cases. The total length of small
bowel is then accurately measured along the
antimesenteric border because residual bowel
length has prognostic significance and will deter-
876
mine the method of reconstruction. If the total
usable bowel length is deemed adequate
(>80  cm  +  ileocaecal valve), 5–15  cm bulbous
hypertrophied proximal bowel is resected back to
the near normal calibre of the bowel. In perform-
ing the resection, one should preserve as much of
the mesentery as possible for later use to fill in
any gaps in the small bowel mesentery left after
end-to-end anastomosis. Having decided on the
site of transection, the bowel is then divided at
right angles leaving an opening of approximately
0.5–1.5 cm in diameter. The blood supply should
be adequate to ensure a safe anastomosis. If,
however, an extensive cutback resection is con-
traindicated because of insufficient residual
bowel length, the bulbous portion alone or any
compromised bowel should be resected. The
proximal bowel should then be tapered obliquely
hand-sewn or using a GIA stapler leaving the
proximal bowel opening a similar size to the dis-
tal bowel lumen to facilitate an end-to- end pri-
mary anastomosis.
Proximal bowel resection is followed by very
limited distal small bowel resection over a length
of 2–3 cm. The resection line should be slightly
a b
Fig. 64.8  End-to-end (a) or end-to-back (b) single-layer bowel anastomosis
A. J. W. Millar et al.
oblique towards the anti-mesenteric border (fish-­
mouth) to ensure that the openings of the proxi-
mal and distal bowels are of approximately equal
size to facilitate easy end-to-end or rarely an end-­
to-­back (Denis-Browne) single-layer anastomo-
sis; 5/0 or 6/0 absorbable suture material is used
(Fig. 64.8). Alternatively, an extra-mucosal end-­
to-­end anastomosis can be performed, placing
sutures on the proximal bowel further apart so as
to accommodate the discrepancy in the bowel
lumen diameters. The mesentery is approximated
with interrupted sutures, which may be difficult
with large mesenteric defects. A side-to-side
anastomosis should not be performed as it can
lead to a blind loop syndrome. There is little
place for routine gastrostomy or transanasto-
motic feeding tubes where facilities for paren-
teral nutrition are available.
64.7.5.2 Special Considerations
Atresia type I and stenosis are best dealt with by
primary resection and end-to-end anastomo-
sis. Procedures such as simple transverse
enteroplasties, excision of membranes, bypass
techniques or side-to-side anastomosis are no
64  Jejuno-Ileal Atresia
longer utilized. They fail to remove the abnor-
mal dysfunctional segments of intestine, thus
increasing the risk of the blind loop syndrome
and dysmotility.
Atresia types II and IIIa are managed in the same
manner as type I with back resection and pri-
mary end-to-end anastomosis. The conserva-
tion of bowel length is mandatory.
Multiple membranous diaphragms (type I atre-
sias) can be successfully perforated by trans-
luminal bouginage done along the entire
length of the affected small bowel (Romao
et al. 2011).
High jejunal atresia: With type IIIb or high jeju-
nal atresia, the proximal bowel should be der-
otated, the ligament of Treitz, if present,
should be taken down, and resection of the
bulbous portion may be extended into the sec-
ond part of the duodenum taking care to stay
well clear of the ampulla of Vater. This is fol-
lowed by an antimesenteric tapering duode-
nojejunoplasty (Kling et  al. 2000). Bowel
tapering can safely be done over a length of
between 20 and 35  cm. This is done to con-
serve bowel length, to reduce disparity in
anastomotic size and to improve duodenal and
proximal jejunal prograde peristaltic function.
At completion, the bowel is left in a position
of derotation with the duodenum-jejunum
positioned on the infant’s right side, the mes-
entery broad-based and the caecum lying
anterior, to the left of the midline in the upper
abdomen. These additional maneuvers induce
rapid return of prograde intestinal function,
and the neonates are usually able to tolerate
graded to full oral intake within 14 days.
With type IIIb atresias, the distal ‘apple peel’
component of the bowel is gently laid out so
that the inner margin of mesentery is clearly
visible from its right colon origin to the atretic
blind end. The division of avascular restricting
mesenteric bands along the free edge of the
distally coiled narrow mesentery may be
required, thereby releasing kinking and inter-
ference with the bowel blood supply. The
large mesenteric defect may be left open, but
where possible the preserved mesentery can
877
be used to fill in the defect. Furthermore, to
prevent kinking of the marginal artery after
the completion of the anastomosis, the bowel
is replaced carefully into the peritoneal cavity
in the position of non-rotation. There is no
need to remove the appendix.
Multiple type IV atresias, present in 20% of
cases, are often localized to a segment facili-
tating an en-bloc resection with a single anas-
tomosis in preference to multiple anastomosis.
If multiple anastomoses are deemed necessary
because of insufficient bowel length, it is use-
ful to railroad each bowel segment to be anas-
tomosed onto a silastic feeding tube before
completing the anastomoses to avoid torsion
and ensure correct orientation and continuity
of the bowel. This tube may be left in situ for
a time to serve as a stent until bowel function
has returned. It is always important to ensure
the preservation of the maximum bowel length
to avoid the short bowel syndrome; for this
reason the following techniques may be
necessary.
Tapering enteroplasty is indicated when the isch-
emic insult has resulted in an atresia with
markedly reduced intestinal length (<80 cm);
where long-length resection of abnormal or
multiple atretic segments are required; to
equalize disparity in anastomotic lumen size;
for the correction of a previous failed inver-
sion plication procedure and to improve func-
tion in a persistently dilated non-functioning
megaduodenum following surgery for upper
jejunal atresia (Thomas Jr. and Carter 1974).
The inversion plication enteroplasty method has
the theoretical advantages of reducing the risk
of leakage from a long antimesenteric suture
line, conserving mucosal surface area, and
may even facilitate the return of bowel peri-
stalsis (Fig.  64.9). More than half of the
antimesenteric bowel circumference may be
enfolded into the lumen over an extended
length without causing an obstruction. The
drawback of this method is that the holding
suture line can unravel within a few months,
precipitating persistent motor dysfunction. As
an alternative, antimesenteric seromuscular
878
a b
Fig. 64.9 (a–b) Inversion plication duodenojejunoplasty for conserving length in high small bowel atresia
stripping combined with inversion plication
may prevent this complication and preserves
maximal mucosal surface for absorption
(Kimura et al. 1996).
Intestinal atresia and gastroschisis: The intesti-
nal peel, often encountered in gastroschisis,
hinders the detection and surgical manage-
ment of an associated atresia. Two options are
available, namely, the primary resection of the
atretic segment with immediate anastomosis,
which is only indicated if this can be done
when there is little or no peel. If the identifica-
tion of the atretic segment is difficult and the
bowel very oedematous, the best approach is
to reduce the eviscerated bowel with the atre-
sia left undisturbed, awaiting resorption of the
peel and oedema, when a safe primary resec-
tion and anastomosis can be performed, which
is usually 14 to 21 days later (van Hoorn et al.
1985).
Exteriorization or stomas: Although primary
anastomosis is always preferred, a temporiz-
ing proximal stoma or chimnied anastomosis
(Bishop-Koop method) or a second-look
exploration 48–72  h later may be advisable
A. J. W. Millar et al.
when the vascular integrity of the intestine is
questionable. This may be relevant in cases
with late diagnosis, type III (b) atresias or with
gross intraperitoneal faecal contamination.
64.8 The Short Bowel Syndrome
Insufficient bowel length either as a result of the
primary injury, excessive removal of residual
bowel or an ischaemic insult to the remaining
bowel as a postoperative complication can lead to
the short bowel syndrome with long-term
sequelae for growth and development (Wales
2004).
Bowel lengthening procedures are best not per-
formed during the initial surgery. Four currently
applicable techniques are longitudinal intestinal
lengthening and tapering method or LILT (Bianchi
1980), serial transverse enteroplasty or STEP
technique (Kim et  al. 2003), spiraling intestinal
lengthening and tailoring (SILT) (Cserni 2011,
van Praagh 2022) and the myoenteropexy method
of Kimura. Although they theoretically increase
intestinal length and reduce lumen diameter,
64  Jejuno-Ileal Atresia
thereby improving both peristalsis and maintain-
ing absorptive capacity, the outcome ultimately
depends on residual small bowel length, the pres-
ence of the ileocaecal valve and any complicating
associated liver disease (Frongia et  al. 2013).
Maximal bowel adaptation should first be attained
utilizing both TPN and enteral feeding as adjunc-
tive methods of therapy before these additional
surgical techniques are contemplated.
64.9 Postoperative Care
Standard methods are used. Nasogastric decom-
pression is usually required for 4–6 days after the
operation (longer for high jejunal atresias).
Therapeutic antibiotics are continued for
5–7 days or longer, and an oral antifungal agent is
given prophylactically. Graduated oral intake is
commenced when the neonate is alert, sucks
well, and there is evidence of prograde gastroin-
testinal function, i.e. clear gastric effluent of low
volume, a soft abdomen and when flatus or faeces
have been passed per rectum.
Surveillance for alimentary dysfunction should
continue until the infant has established and stabi-
lized normal gastrointestinal function. If at any
time there is a suspicion of a leak at the anastomo-
sis (suggested by vomiting, abdominal distention,
tenderness and evidence of sepsis), a plain erect
or decubitus radiograph of the abdomen should be
taken. If this reveals free air in the abdomen more
than 24 h after the operation, laparotomy should
be performed immediately and the leaking site
sutured or the anastomosis redone. Other compli-
cations encountered include wound sepsis and
dehiscence, anastomotic kinking, ischaemia with
late-onset anastomotic stenosis, adhesive obstruc-
tion and the short bowel syndrome.
After massive intestinal length loss, TPN or
other forms of enteral support should be contin-
ued with until maximum bowel adaptation is
reached. It is imperative that graduated enteral,
preferably breast milk feeding, should be intro-
duced as soon as possible to stimulate and
enhance this process, which can take many
months (Sigalet 2001).
879
64.10 Conclusion
Following surgical correction of the anomaly, the
majority of children grow and develop normally,
the end result being influenced by the residual
intestinal length and absorptive and peristaltic
function of the residual bowel, associated anoma-
lies and the successful management of the short
bowel syndrome.
The quality of life and functional status are
usually very good and are not affected by intes-
tinal atresia, although cystic fibrosis, the short
bowel syndrome and TPN-induced liver failure
may alter prognosis. Factors contributing to the
<10% mortality rate include delay in presenta-
tion, the type of atresia, proximal bowel infarc-
tion with peritonitis, anastomotic dysfunction,
missed distal atresias, the short bowel syn-
drome, pneumonia, sepsis and associated
abnormalities.
References
Banieghbal B, Beale PG (2007) Minimal access approach
to jejunal atresia. J Pediatr Surg 42(8):1362–1364
Benson CD (1955) Resection and primary anastomosis
of the jejunum and ileum in the newborn. Ann Surg
142(3):478–485
Best KE, Tennant PW, Addor MC, Bianchi F, Boyd P,
Calzolari E, Dias CM, Doray B, Draper E, Garne E,
Gatt M, Greenlees R, Haeusler M, Khoshnood B,
McDonnell B, Mullaney C, Nelen V, Randrianaivo
H, Rissmann A, Salvador J, Tucker D, Wellesly D,
Rankin J (2012) Epidemiology of small intestinal atre-
sia in Europe: a register-based study. Arch Dis Child
Fetal Neonatal Ed 97(5):F353–F358
Bianchi A (1980) Intestinal loop lengthening--a technique
for increasing small intestinal length. J Pediatr Surg
15(2):145–151
Cserni T, Takayasu H, Muzsnay Z, Varga G, Murphy F,
Folaranmi SE, Rakoczy G (2011) New idea of inestinal
lenghtening and tailoring. Ped Surg Int 27:1009–1113
Cywes S, Davies MRQ, Rode H (1982) Congenital
jejuno-ileal atresia and stenosis. In: Cardiovascular,
respiratory, gastrointestinal and genitourinary malfor-
mations. Springer, Heidelberg, pp 109–130
Frongia G, Kessler M, Weih S, Nickkholgh A, Mehrabi
A, Holland-Cunz S (2013) Comparison of LILT and
STEP procedures in children with short bowel syn-
drome–a systematic review of the literature. J Pediatr
Surg 48(8):1794–1805
880
Grosfeld JL, Ballantine TV, Shoemaker R (1979)
Operative management of intestinal atresia and ste-
nosis based on pathologic findings. J Pediatr Surg
14(3):368–375
Kim HB, Fauza D, Garza J, Oh JT, Nurko S, Jaksic
T (2003) Serial transverse enteroplasty (STEP): a
novel bowel lengthening procedure. J Pediatr Surg
38(3):425–429
Kimura K, Perdzynski W, Soper RT (1996) Elliptical
seromuscular resection for tapering the proximal
dilated bowel in duodenal or jejunal atresia. J Pediatr
Surg 31(10):1405–1406
Kling K, Applebaum H, Dunn J, Buchmiller T, Atkinson
J (2000) A novel technique for correction of intesti-
nal atresia at the ligament of Treitz. J Pediatr Surg
35(2):353–355. discussion 356
Lambrecht W, Kluth D (1998) Hereditary multiple atre-
sias of the gastrointestinal tract: report of a case and
review of the literature. J Pediatr Surg 33(5):794–797
Lima M, Ruggeri G, Domini M, Gargano T, Mazzero G,
Landuzzi V, Sciutti R (2009) Evolution of the surgi-
cal management of bowel atresia in newborn: lapa-
roscopically assisted treatment. Pediatr Med Chir
31(5):215–219
Louw JH (1952) Congenital intestinal atresia and severe
stenosis in the newborn; a report on 79 consecutive
cases. S Afr J Clin Sci 3(3):109–129
Louw JH (1959) Congenital intestinal atresia and stenosis
in the newborn. Observations on its pathogenesis and
treatment. Ann R Coll Surg Engl 25:209–234
Louw JH (1966) Congenital atresia and stenosis of the
small intestine. The case for resection and primary
end-to-end anastomosis. S Afr J Surg 4(2):57–64
Louw JH, Barnard CN (1955) Congenital intes-
tinal atresia observations on its origin. Lancet
266(6899):1065–1067
Louw J, Cywes S, Davies M, Rode H (1981a) Congenital
jejuno-ileal atresia: observations on its pathogenesis
and treatment. Eur J Pediatr Surg 33(05):3–17
Louw JH, Cywes S, Davies MR, Rode H (1981b)
Congenital jejuno-ileal atresia: observations on its
pathogenesis and treatment. Z Kinderchir 33(1):3–17
Martin LW, Zerella JT (1976) Jejunoileal atresia–pro-
posed classification. J Pediatr Surg 11(3):399–403
Millar AJW, Numanoglu A, Cox SG (2020) Jejunoileal
atresia and Stenosis. In: Puri P (ed) Pediatric surgery,
vol 1. Springer Nature, Germany, pp 913–923
Mishalany HG, Najjar FB (1968) Familial jejunal atre-
sia–3 cases in one family. J Pediatr 73(5):753–755
Murphy FJ, Mohee A, Khalil B, Lall A, Morabito A,
Bianchi A (2009) Versatility of the circumumbili-
A. J. W. Millar et al.
cal incision in neonatal surgery. Pediatr Surg Int
25(2):145–147
Nixon HH (1960) An experimental study of propul-
sion in isolated small intestine, and applications
to surgery in the newborn. Ann R Coll Surg Eng
27(2):105–124
Nixon HH, Tawes R (1971) Etiology and treatment of
small intestinal atresia: analysis of a series of 127
jejunoileal atresias and comparison with 62 duodenal
atresias. Surgery 69(1):41–51
Nygaard K (1967) Changes in the intestinal flora after
resections and bypass-operations on the small intes-
tine in rats. The effect of different types of anastomo-
sis. Acta Chir Scand 133(7):569–583
Puri P, Fujimoto T (1988) New observations on the patho-
genesis of multiple intestinal Atresias. J Pediatr Surg
23(3):221–225
Romao RL, Ozgediz D, de Silva N, Chiu P, Langer J,
Wales PW (2011) Preserving bowel length with a
transluminal stent in neonates with multiple intestinal
anastomoses: a case series and review of the literature.
J Pediatr Surg 46(7):1368–1372
Santulli TV, Blanc WA (1961) Congenital atresia of
the intestine: pathogenesis and treatment. Ann Surg
154:939–948
Sigalet DL (2001) Short bowel syndrome in infants and
children: an overview. Semin Pediatr Surg 10(2):49–55
Spriggs N (1912) Congenital intestinal occlusion. Guys
Hosp Rep 66:143–218
Thomas CG Jr, Carter JM (1974) Small intestinal atre-
sia: the critical role of a functioning anastomosis. Ann
Surg 179(5):663–670
van Hoorn WA, Hazebroek FW, Molenaar JC (1985)
Gastroschisis associated with atresia--a plea for delay
in resection. Z Kinderchir 40(6):368–370
van Praagh JB, Hofker HS, Haveman JW (2022)
Comparing bowel lenghtening procedures: which,
when, and why? Curr Opin Organ Transplant
27:112–118
Veyrac C, Couture A, Saguintaah M, Baud C (2004)
MRI of fetal GI tract abnormalities. Abdom Imaging
29(4):411–420
Wales PW (2004) Surgical therapy for short bowel syn-
drome. Pediatr Surg Int 20(9):647–657
Wax JR, Hamilton T, Cartin A, Dudley J, Pinette MG,
Blackstone J (2006) Congenital jejunal and ileal atre-
sia: natural prenatal sonographic history and asso-
ciation with neonatal outcome. J Ultrasound Med
25(3):337–342
 Meconium Ileus
Valeria Solari and Massimo Rivosecchi
65.1 Introduction
Meconium ileus, the earliest clinical manifesta-
tion of CF, occurs in as much as 20% of patients
with CF. The clinical features are mainly due to
the presence of abnormal, inspissated, and adhe-
sive meconium. In affected neonates, CF causes
intestinal obstruction in the midileum to terminal
ileum, thus leading to progressive abdominal dis-
tension and a failure to pass the meconium.
Approximately 40% of patients with meconium
ileus have complications of intestinal volvulus,
atresia, or bowel necrosis (Escobar et al. 2005). If
perforation occurs earlier in utero, then intestinal
reabsorption and formation of pseudocysts or
abdominal calcifications may be observed. The
colon is typically “unused” or “microcolon.”
Relief of the intestinal obstruction can be
achieved through a hyperosmolar contrast enema
under fluoroscopic guidance. Failed meconium
disimpaction or complex meconium ileus neces-
sitates surgery (Carlyle et al. 2012). When con-
servative management with enema irrigation
fails, operative intervention is necessary.
Currently, there is no consensus regarding the
V. Solari (*)
Department of Pediatric Surgery, Klinik Donaustadt,
Vienna, Austria
e-mail: valeria.solari@gesundheitsverbund.at
M. Rivosecchi
Department of Pediatric Surgery, “Bambino Gesù”
Children’s Hospital, Palidoro, Rome, Italy
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_65
65
ideal surgical strategy for managing meconium
ileus. However, the long-term outcomes and sur-
vival of patients with meconium ileus and CF
have significantly improved over the past few
decades and are no worse than those in patients
with CF without meconium ileus (Johnson et al.
2010).
65.2 Historical Overview
The first description of an infant with meconium
ileum, accompanied by a description of the histo-
logical pancreatic changes, was reported by
Landsteiner in 1905 (Landsteiner 1905). He sug-
gested that a lack of pancreatic secretion causes
the thickening of the meconium, resulting in
bowel obstruction. In 1929, Kornblith described
meconium ileus with congenital stenosis of the
main pancreatic duct (Kornblith and Otani 1929).
Dodd suggested that the failure of pancreatic
enzymes to reach the gut during perinatal life was
at least partially responsible for meconiums ileus
(Dodd 1936). The term CF of the pancreas was
coined in 1936 by Fanconi, who described the
association of chronic pulmonary disease with
pancreatic insufficiency in infants (Fanconi et al.
1936). In 1938, Andersen first described the char-
acteristics of CF in the pancreas and correlated
the findings with the lung and intestinal disease
in newborns with meconium ileus (Andersen
1938). She also suggested that CF is a recessive
881
882
disease and initially used pancreatic enzyme
replacement therapy. Farber recognized signifi-
cant multiple organ involvement with tenacious
thick mucus, hence his terminology of mucovis-
cidosis (Farber et  al. 1943). Di Sant’Agnese’s
description of sweat abnormalities in CF in 1953
led to the development of the diagnostic sweat
test. For several decades, meconium ileus and CF
remained as pediatric diseases with high mortal-
ity rates. The first significant improvement in
treatment came in 1948, when Hiatt and Wilson
described the intraoperative method of disimpac-
tion of the inspissated meconium with saline via
enterostomy tube (Hiatt and Wilson 1948). This
was followed in 1969 by Noblett’s use of hyper-
osmolar enema with Gastrografin in the manage-
ment of uncomplicated meconium ileus (Noblett
1969). Over the years, several operative methods
followed. Shwachman encouraged Gross to treat
meconium ileus; the Mikulicz side-by-side enter-
ostomy was first reported by Gross in 1953.
Distal chimney enterostomy was described by
(Bishop and Koop 1957) and was followed by a
description of proximal enterostomy by (Santulli
and Blanc 1961). In 1970, O’Neill reported suc-
cessful tube enterostomy with and without bowel
resection. In 1989, scientists discovered the CF
transmembrane conductance regulator (CFTR)
gene and the most prevalent CF-causing mutation
(F508del) (Kerem et  al. 1989; Riordan et  al.
1989; Rommens et  al. 1989). The first CFTR-­
modulating drug entered the market in 2012
(Ramsey et al. 2011).
65.3 Incidence
Meconium ileus is seen in approximately 20% of
patients with CF.  The incidence rates for CF in
Europe are similar to those seen across the United
States and Canada; approximately 1  in 2000 to
3000 infants of northern European descendent
are diagnosed at birth, with an allele frequency as
high as 1  in 29 (Ratjen et  al. 2015). CF is less
common in other ancestral groups, affecting
approximately 1  in 17,000 African Americans,
1  in 13,5000 Hispanic Whites, and 1  in 31,000
Asian Americans (Kopp et  al. 2015). The inci-
V. Solari and M. Rivosecchi
dence is far lower in India, at only 1 in 40,000–
100,000 births. However, CF is thought to be
underdiagnosed in Africa and in Asia. In many
countries, infants are not tested for CF at birth,
and national registries are lacking data. Many
countries lack a national CF registry, and studies
are conducted by individuals, thus hindering
accurate determination of the prevalence.
A patient with two copies of the most com-
mon F508del mutation has a 25% risk of present-
ing with meconium ileus, whereas a patient with
a F508del paired with another mutation has a
17% risk. A family history of CF is present in
10–20% of patients with meconium ileus, and the
recurrence rate of meconium ileus has been found
to be 39% among siblings with CF (Allan et al.
1981; Blackman et al. 2006).
65.4 Etiopathogenesis
The development of meconium ileus is based on
an autosomal recessive mutation in both copies
of the CFTR gene, located on the long arm of
chromosome band 7q31.2. If both parents are
carriers of the CFTR gene mutation, there is a
one in four chance (25%) that their offspring will
develop the disease. More than 2000 different
mutations of the CFTR gene have been identified
to date, but only approximately 400 are thought
to cause disease. The most common is F508del
(previously termed ∆F508), a three base-pair
deletion on exon 11 that results in the loss of phe-
nylalanine 508; this mutation accounts for ~70%
of pathogenic CFTR variants globally.
Approximately 90% of patients with CF in the
United States have one or two F508del alleles,
and the worldwide frequency is 70% (Sharma
and Cutting 2020). Most mutations have no
known specific effects on the clinical phenotype
or function of the protein product. A five-class
system was developed for classifying CFTR
mutations on the basis of functional conse-
quences, with the aim of grouping patients with
similar prognoses. Infants with meconium ileus
and pancreatic insufficiency frequently carry
severe class I-III CFTR mutations on both alleles
(F508del, G542X, W1282X, R553X, and
65  Meconium Ileus
G551D), thus resulting in absent or nonfunc-
tional CFTR at the cell surface. Patients with
milder mutations (class IV or V) on at least one
allele maintain some residual ion channel activity
with sufficient exocrine pancreatic function.
However, this classification is increasingly rec-
ognized to be an oversimplification. The develop-
ment of meconium ileus is unlikely to be
associated with environmental factors, because it
appears early in life and exclusively in patients
with severe CFTR mutations. Studies on twins
and siblings have shown that monozygous twins
have a higher concordance of meconium ileus
than do dizygous twins and siblings, thereby sug-
gesting that non-CFTR modifier genes, such as
chromosome 12p13.3, which also contribute to
the development of meconium ileus (Blackman
et al. 2006). In contrast, genetic modifier effects
do not have a role in the development of later pre-
sentations of distal intestinal obstruction syn-
drome (DIOS) in CF (Smith et  al. 2009).
However, despite multiple modifier genes
enhance or reduce the risk of meconium ileus,
their clinical significance has remained limited.
65.5 Pathophysiology
The pathophysiology of meconium ileus is asso-
ciated with defects in the CFTR gene, which
encodes a chloride channel located at the apical
membrane in epithelial cells. CF is characterized
by the build-up of thick, viscous mucus in multi-
ple mucin-producing organs, such as the intes-
tines, lungs, pancreas, and reproductive organs.
For this reason, CF has been described as muco-
viscidosis, thus suggesting that the polymeric
gel-forming mucus and the glycosylation of pro-
teins play critical roles in CF. The CFTR protein
regulates the viscoelastic properties of mucus
(Ooi and Durie 2016).
Defects in CFTR gene expression lead to defi-
ciencies in cyclic adenosine monophosphate-­
dependent chloride and bicarbonate transport in
the affected epithelia, a process necessary for
fluid secretion and extracellular alkalization.
Work by Quinton and others has shown that
CFTR-dependent secretion of bicarbonate is
883
important for normal mucus release and viscosity
and for mucin unfolding in the bowel (Quinton
2008). Defective bicarbonate secretion by the
neighboring epithelial cells that surround mucin-­
secreting goblet cells makes the intraluminal
environment more acidic and dry, thus contribut-
ing to the pathophysiology of meconium ileus. In
addition, defective chelation of Ca2+ by bicarbon-
ate compromises further mucus unfolding and
crosslinking, thereby making the meconium
thick, viscid, and adhesive.
Several mouse models of CF have been cre-
ated, including the most common mutations,
F508del and G551D (Rosen et  al. 2018). One
model involves a replacement strategy that inter-
rupts the CFTR gene and generates a complete
“knockout” that does not produce normal CFTR
protein products. In the so-called residual func-
tion models, an insertional strategy results in the
production of a low amount of CFTR
mRNA.  Although the intestinal CF pathology
resembles that of meconium ileus, it develops
postnatally and therefore clearly differs from the
human meconium ileus seen at birth. Meconium
ileus is fatal in pig and ferret models, with a pen-
etrance of 50–100% and 100%, respectively—
values much higher than those in humans
(Meyerholz et  al. 2010). The various animal
models are characterized by sticky meconium,
severe neonatal intestinal obstruction, microco-
lon, goblet cell hyperplasia, and crypt dilata-
tion—symptoms similar to those in the human
manifestation. However, no single animal model
completely replicates the complexity of human
CF.  Gene-targeted animal models have aided in
the understanding of the gastrointestinal manifes-
tations of CF and are essential in redirecting
efforts using new paradigms to develop novel
therapeutic strategies.
65.6 Pathology
In meconium ileus, the dilated loops of the small
bowel are filled with inspissated tenacious, sticky,
green meconium, which may also be tarry or
gritty. The dilated proximal ileum is filled with
semiliquid meconium and its walls thicker.
884
Distally, the meconium becomes thicker, adhe-
sive, dark green, putty-like, and firmly adherent
to the bowel walls. The distal ileum is filled with
small inspissated “rabbit pellets” of meconium,
which are stained gray and have a beaded appear-
ance. Little or no meconium is passed into the
colon. Typically, the colon is of small caliber and
is “unused” or “microcolon.” In complicated
cases of meconium ileus, intestinal perforation
and secondary meconium peritonitis and calcifi-
cation can occur. Spontaneous healing of ileal
perforation in utero can lead to resorption of the
involved portion of the bowel and intestinal atre-
sia. When the peristalsis is vigorous, the twisting
of the ileal tract full of dense meconium may
result in volvulus, with a high risk of
perforation.
Perforation of the small bowel in utero results
in a meconium leak and produces a sterile chemi-
cal peritonitis, which results in calcification than
is visible radiographically, and intestinal adhe-
sion by dense fibrosis. Collections of meconium
may be walled off, forming meconium pseudo-
cysts. The passage of spilled meconium into the
inguinal canal can cause meconium periorchitis.
Perinatal meconium peritonitis manifests as a
green exudate overlying the serosal surface of the
adjacent bowel and peritoneal cavity.
Histological findings in the bowel reveal
dilated crypts with accumulation of secretions,
prominent goblet cells in the mucosa, the pres-
ence of a thick mucus layer, and accumulation of
meconium adherent to the mucosal surface.
Intestinal specimens from infants with compli-
cated meconium ileus also show submucosal
inflammation and fibrosis, as seen more promi-
nently in DIOS ilea (Smith et al. 2009). Thus, the
transmural inflammation present at birth in
patients with CF may be a factor contributing to
intestinal dysmotility and the later development
of DIOS.
65.7 Diagnosis
The diagnosis of meconium ileus is made on the
basis of clinical history and physical examina-
tion, as well as imaging. Nearly 25% of patients
V. Solari and M. Rivosecchi
with meconium ileus have a family history of
CF. Prenatally, a fetus is tested if it is considered
to be at high risk of CF, such as when both par-
ents are known to be carriers of at least one
pathogenic CFTR variant. This testing can be
performed on chorionic villus specimens at
10–12  weeks and on amniocentesis samples at
16–18 weeks. Clinical practice is currently shift-
ing toward noninvasive prenatal testing using
cell-free fetal DNA from maternal peripheral
blood. When both parents are carriers of the CF
gene mutation, there is a 1 in 4 chance (25%) of
CF in the child, a risk 625 times greater than that
in the general population (0.04%). Newborn
screening methods differ among countries.
Prenatal sonographic characteristics associ-
ated with meconium ileus are hyperechoic
masses, dilated bowel loops and non-­visualization
of the gallbladder. Approximately 0.8–13.3% of
fetuses with echogenic bowel are estimated to
have CF (Scotet et  al. 2010; De Oronzo 2011).
The positive predictive value of hyperechoic
bowel CF is 52% in fetuses with high CF risk, but
only 6.4% in fetuses with low CF risk (Bahado-­
Singh et al. 1994). Hyperechogenic bowel in the
second and third trimesters is considered a nor-
mal variant, and as much as 65% of cases resolve
on sonographic follow-up and do not have CF or
meconium ileus (Ruiz et  al. 2009). Echogenic
bowel has been associated with a variety of other
conditions such as trisomy, prematurity, fetal
demise, and maternal infections. Dilated bowel
loops on prenatal ultrasound have been reported
even less frequently in association with CF than
hyperechoic bowel. Additional findings associ-
ated with meconium ileus are fetal ascites and the
presence of intraabdominal cysts. Maternal poly-
hydramnios are frequently observed in prenatal
diagnosis of complicated forms of meconium
ileus, presumably as a result of the bowel
obstruction.
The diagnosis of CF relies on both the clinical
presentation, as well as evidence of CFTR dys-
function. Newborn screening is based on elevated
levels of trypsinogen (IRT) detected on dry blood
spots on the Guthrie card. IRT, a pancreatic
enzyme precursor released into the blood stream
after pancreatic damage, is used as a CF screen-
65  Meconium Ileus
ing method in newborns. However, patients
should have their diagnosis of CF confirmed or
refuted, either by a sweat test or the presence of a
CFTR mutation, according to approved proce-
dural guidelines in established international pro-
tocols such as the Clinical and Laboratory
Standards Institute Guidelines (Farrell et  al.
2017). CFTR dysfunction, such as abnormal
sweat chloride, is diagnosed at a concentration of
greater than 60 mmol/L, whereas 39–59 mmol/L
is considered to indicate intermediate cases that
should be followed up in specialist CF centers.
Newborns with corrected gestational age above
36 weeks, whose body weights are greater than
2 kg, and who have positive CF newborn screens
or positive prenatal gene tests should receive
sweat chloride testing as soon as possible after
10  days of age, ideally before 4  weeks of age.
Genotyping has become a key element of the
diagnostic workup, particularly since the intro-
duction of CFTR-modulating therapies that are
specific to certain mutations.
A unique clinical characteristic of meconium
ileus is that neonates present immediately after
birth with abdominal distension, which is pro-
duced before the patients swallow air. The bowel
loops are doughy with visible peristaltic waves
and finger pressure over a loop causes indenta-
tions—the so-called putty sign (Fig.  65.1a).
Bowel obstruction also results in delayed passage
of meconium and bilious vomiting or bile-stained
gastric fluid. Complicated meconium ileus pres-
a b
Fig. 65.1 (a) A newborn with intestinal obstruction and a
family history of cystic fibrosis. A Gastrografin enema
failed, and the patient underwent laparotomy for meco-
885
ents in utero with bowel obstruction or signs of
bowel perforation and/or necrosis, such as
intraabdominal calcifications, or signs of perito-
nitis such as a discolored or edematous abdomi-
nal wall and abdominal tenderness. Meconium
pseudocysts and signs of peritoneal irritation
may be present. Bowel perforation and peritonitis
can cause hypovolemia and/or sepsis. After birth,
an abdominal radiograph of uncomplicated
meconium ileus shows a similar appearance in
erect or supine position without air-fluid levels,
despite the obstructed bowel, because the air can-
not layer above the thick and viscous meconium.
Radiographs may also show multiple dilated
bowel loops of various sizes, owing to the con-
figuration of different segments of the bowel. A
“ground-glass” appearance (“Neuhauser’s sign”)
or fine, granular “soap bubbles” (“Singleton’s
sign”) are often seen in the right half of the abdo-
men, owing to trapped air in the sticky
meconium.
When the meconium ileus is complicated, the
abdominal radiograph may show calcification as
a result of meconium peritonitis due to fetal
bowel perforation. A double bubble image or air-­
fluid levels can be seen in association with ileal
atresia after a volvulus with ischemic damage. If
the intestinal perforation occurs early in the ante-
natal period, the X-ray appearance of a round rim
of calcification indicates a meconium pseudo-
cyst. In complicated meconium ileus, calcified
meconium pseudocysts or complications, such as
nium ileus. (b) Passage of a large meconium plug in a
newborn with meconium plug syndrome
886
volvulus, atresia, necrosis, or perforation, can be
seen. Plain radiography is the initial imaging
method used; however, there is a need for further
evaluation though a contrast study, as it can also
be therapeutic. The radiological features of
meconium ileus on contrast enema are well
described, including a typically “unused” colon
or “microcolon” of normal length that may be
empty or may contain several pellets of inspis-
sated meconium—a feature of uterine underuti-
lization. A reflux of contrast material into the
terminal ileum will show meconium pellets and
more proximally dilated loops of the small bowel.
65.8 Differential Diagnosis
Clinical and radiological features are important
in diagnosing meconium ileus but are not pathog-
nomonic. Other differential diagnoses of neona-
tal bowel obstructions are jejunoileal atresia,
Hirschsprung’s disease, neonatal small left colon,
and meconium plug syndrome. Meconium ileus
accounts for as much as 25% of cases of intesti-
nal obstruction in neonates. Ileal atresia shows
air-fluid levels and may be associated with meco-
nium ileus. Hirschsprung’s disease, particularly
total colonic aganglionosis, can also mimic
meconium ileus. Colonic Hirschsprung’s disease
shows a transitional zone, and, in patients with
total colonic aganglionosis, the contrast medium
refluxes into the terminal ileum and demonstrates
air-fluid levels. Ultimately, the histopathological
findings of a rectal biopsy can confirm agangli-
onosis. Neonatal small left colon syndrome, a
functional disease with signs of bowel obstruc-
tion, can also be included in the differential diag-
nosis of meconium ileus. The transient
dysmotility of the descending colon produces
radiological features of a small descending colon
with an abrupt transition zone at the splenic flex-
ure. A contrast enema can be therapeutic with the
passage of a meconium plug. Although its
pathogenesis is unknown, neonatal small left
­ rhea and osmotic diuresis occur; therefore,
colon syndrome is often associated with maternal
diabetes, hyperthyroidism, drug abuse, or
eclampsia. Meconium plug syndrome is often a
benign cause of intestinal obstruction that clears
V. Solari and M. Rivosecchi
after rectal stimulation or a contrast enema.
However, this syndrome may be associated with
Hirschsprung’s disease, with an incidence of
13%, when the plug is found on contrast enema.
The clinical presentation of meconium plug syn-
drome (Fig. 65.1b) is similar to meconium ileus,
and it has been reported to be associated with CF
(Olsen et al. 1982; Keckler et al. 2008). Patients
with meconium plug syndrome should be tested
for CF, and a rectal biopsy should be performed
to exclude Hirschsprung’s disease, particularly
when normal bowel function does not occur after
the passage of the plug. Other conditions may
mimic bowel obstruction, such as delayed peri-
stalsis associated with prematurity, hypothyroid-
ism, and transient ileus in sepsis. Meconium ileus
may occur even in the absence of CF (Fakhoury
et al. 1992; Gorter et al. 2010).
65.9 Management
The management of meconium ileus consists of
nonoperative and operative management of sim-
ple and complicated disease. The first steps of the
treatment include nasogastric tube decompres-
sion, antibiotic prophylaxis, and correction of
dehydration, electrolytes, and hypothermia. After
the initial assessment, a plain radiograph should
exclude the diagnosis of perforation and peritoni-
tis. An initial contrast enema with a water-soluble
agent can exclude other causes of distal bowel
obstruction and complications such as atresia.
Nonoperative management of simple meconium
ileus involves hypertonic enemas such as
Gastrografin performed under fluoroscopic guid-
ance. Gastrografin is a hyperosmolar
(1900 mOsm/L) water-soluble, radiopaque solu-
tion that contains 0.1% polysorbate 80 and 37%
organically bound iodine. The high osmolar
properties pull fluid into the bowel lumen, and
the solvent properties help release the viscous
meconium. After administration, osmotic diar-
patients must undergo aggressive fluid resuscita-
tion via an IV line to avoid fluid hypovolemia,
electrolyte imbalance, and cardiovascular col-
lapse. Since its first successful use in neonates
65  Meconium Ileus
with uncomplicated meconium ileus by (Noblett
1969), hypertonic contrast enema has become the
standard of nonoperative care. Importantly the
enema should be performed under fluoroscopy,
and the contrast agent must reach the meconium
impacted ileum to relieve the obstruction.
Spontaneous passage of the meconium should
follow. An abdominal radiograph should be per-
formed 8–12 h later to monitor the obstruction;
the enema can be repeated in the event of incom-
plete evacuation. A pediatric surgeon should be
present during the procedure. Serial Gastrografin
enemas can be repeated at 12–24-h intervals.
Although other agents have been used,
Gastrografin remains the most common. If the
contrast agent does not reflux into the dilated
bowel or no evacuation occurs after successful
refluxing into the obstructed ileum, a surgical
intervention should be planned. The success rate
for neonates with uncomplicated meconium ileus
conservatively managed with enemas is reported
to be as high as 83%, but recently the rate
decreased to as low as 36% (Copeland et  al.
2009). These findings may be attributable to the
use of agents with a lower osmotic activity or less
aggressive attempts.
65.9.1 Operative Management
Indications for operative interventions are com-
plications of meconium ileus, such as ileal atre-
sia, volvulus, perforation, meconium cysts,
peritonitis, bowel necrosis, or a combination of
various conditions. In the remaining patients, sur-
gery becomes necessary when nonoperative
treatments with enema solubilizing agents fail or
are incomplete. The goals of operative manage-
ment are to decompress the bowel and relieve
obstruction by eliminating the meconium, estab-
lish intestinal continuity, preserve maximal
bowel length, and allow for prompt enteral
nutrition.
Several surgical procedures are available for
the management of patients with meconium
ileus. However, their application remains conten-
tious, and there is no consensus in the literature
regarding the ideal surgical strategy for meco-
887
nium ileus. Enterotomy, with or without resec-
tion, was described by O’Neill in 1970 and has
been used extensively in the past (Mak et  al.
2000). The bowel irrigations are injected through
an enterotomy, and the meconium is milked dis-
tally into the colon or removed through the enter-
otomy. A similar method is a T-tube enterostomy,
in which the enterostomy is attached to the
abdominal wall, thus creating a fistula allowing
for tube irrigation with normal saline or n-­
acetylcysteine, and instillation of pancreatic
enzymes. The enterostomy can be closed directly
at laparotomy or may heal spontaneously after
removal. Resection and primary anastomosis, as
suggested by Swenson in 1962, is the procedure
of choice for many surgeons. Advocates for this
procedure argue that it allows for a faster recov-
ery and avoids high-output stoma losses, poor
nutritional status associated with better extraint-
estinal prognosis, and a second laparotomy for
stoma closure. This approach comes with a risk
of anastomotic leakage and peritonitis or stric-
tures that may require a second laparotomy, par-
ticularly in patients in unstable condition with
poor bowel perfusion. The complication rate has
been reported to be approximately 20–31%,
including adhesive bowel obstruction (Farrelly
et  al. 2014; Jawaheer et  al. 2007; Karimi et  al.
2011). The adhesions encountered at re-­
laparotomy have been reported to be more com-
mon after surgery for MI than in other neonatal
surgical conditions, thus suggesting a different
underlying inflammatory response after MI
(Choudhry and Grant 2006). Another option is to
create a stoma that can be closed later, thereby
giving patients time to recover. Several stoma
types have been advocated for over the years. The
formation of a temporary relieving stoma in the
form of a Mikulicz double-barreled enterostomy
was promoted by Gross (Fig. 65.2). In that case,
there is no risk of an intraabdominal anastomotic
leak, and complete evacuation of the inspissated
meconium is not required. The bowel can be
opened after closure of the abdominal wall, thus
avoiding intraabdominal contamination.
Another well-described surgical option is
bowel resection and the formation of a “distal
chimney enterostomy,” with an end-to-side ileal
888 V. Solari and M. Rivosecchi

Mikulicz

Fig. 65.2  The Mikulicz double-barreled ileostomy

Modified Bishop-Koop

Fig. 65.4  The modified Bishop-Koop ileostomy

Bishop-Koop

Fig. 65.3  The Bishop-Koop ileostomy

anastomosis, the so-called Bishop-Koop enteros-

tomy (Fig. 65.3).
A variation in this technique has been
described, wherein the proximal segment is anas-
Santulli
tomosed obliquely to the distal stump (Fig. 65.4).
The Bishop-Koop method remains preferred by Fig. 65.5  The Santulli enterostomy
many, because the bowel contents pass into the
distal bowel, thus allowing for more fluid and an end-to-side fashion to the proximal ileum at a
nutrient absorption. Santulli has described a level close to the subfascial plane, whereas the
­distal chimney enterostomy, which is essentially proximal ileum forms the enterostomy. This pro-
the reverse of the distal chimney enterostomy cedure allows for easier postoperative proximal
(Fig. 65.5). The distal ileal end is anastomosed in bowel decompression and irrigation. An intraop-
65  Meconium Ileus
eratively placed distal tube can be used for later
instillation of solubilizing agents. The Santulli
stoma is a high-output enterotomy and should be
closed relatively sooner to avoid unnecessary
electrolyte imbalances, dehydration, and nutri-
tional losses.
Complicated meconium ileus almost always
requires a laparotomy, except for in utero perfo-
rations that have spontaneously healed without
postnatal sequelae. During laparotomy, the
entire bowel should be inspected, because other
surgical conditions can be encountered, such as
atresia, volvulus, peritonitis, necrotic bowel, or
meconium cysts. During surgery, critical bowel
resection and documenting the remaining bowel
length are mandatory. After surgery, the residual
intraluminal meconium must be evacuated with
the instillation of 2% or 4% acetylcysteine via a
nasogastric tube or directly into the intestine via
a stoma or enterostomy. Stomas should be
closed as soon as possible to avoid electrolyte
imbalances and unnecessary nutritional losses.
In the postoperative period, particularly after
stoma formation, parenteral nutrition is neces-
sary until oral or enteral tube feeding is fully
established, and weight gain is satisfactory. To
minimize the risk associated with long-term
parenteral nutrition, such as cholestasis and
liver failure, the selection of appropriated lipids
with low osmolarity medium-chain triglycerides
and concentrated hydrolyzed protein should be
used. Parenteral nutrition can be prolonged and
must include the maintenance of salt, electro-
lytes, and vitamins. The short-term use of urso-
deoxycholic acid can improve bile flow and
prevent cholestasis. Patients with CF are sus-
ceptible to sodium losses, which can be particu-
larly high through stoma outputs and low-dietary
intake from breast milk, and may be associated
with poor growth. Sodium deficiency should be
assessed on an individual basis through the
measurement of the urinary sodium:creatinine
ratio. The early introduction of oral or enteral
feeding is beneficial. Breastfeeding or feeding
of expressed breast milk should be encouraged.
Pancreatic enzyme replacement therapy should
be started when enteral feeding is commenced.
Pancreatic insufficiency is confirmed by obtain-
889
ing a fecal elastase from a rectally collected
sample and not from an enterostomy.
Postoperatively, the diagnosis of CF must be
confirmed as the cause of meconium ileus. Over
time, the outcomes of patients with meconium
ileus and CF have improved, with a survival rate
approaching 100% for both operative and non-
operative management. Historically, patients
with CF and meconium ileus have had poorer
outcomes than those without meconium ileus.
However, advances in surgical and medical
management have improved the outcomes of the
former, and their survival rates are currently
similar to those of CF patients without meco-
nium ileus.
65.10 Conclusion
Pediatric surgeons play a critical role in the initial
management of patients with meconium ileus
and CF. Although patient care must be individu-
alized, relieving the bowel obstruction, preserv-
ing bowel length, and supporting the infant’s
physiology and nutritional status are essential.
Every patient with meconium ileus must be eval-
uated for CF regardless of newborn screening or
ancestry. Neonates presenting with meconium
ileus, particularly those undergoing surgery with
resection and/or stomas, are at greater risk of
nutritional deficits later in life. These patients
should receive multidisciplinary management in
a specialist center for CF and neonatal surgery to
improve long-term outcomes.
Continued success in the management of
patients with meconium ileus and CF-related
gastrointestinal problems will depend on con-
tinuing advances in the understanding of the
underlying disease mechanism.
References
Allan JL, Robbie M, Phelan PD et  al (1981) Familiar
occurrence of meconium ileus. Eur J Pediatr
135:291–192
Andersen DH (1938) Cystic fibrosis of the pancreas and
its relation to celiac disease: a clinical and pathologic
study. Am J Dis Chil 56:344–399
890
Bahado-Singh R, Morotti R, Copel JA et  al (1994)
Hyperechoic fetal bowel: the perinatal consequences.
Prenat Diagn 14:981–987
Bishop HC, Koop CE (1957) Management of meconium
ileus, resection, Roux-en-Y anastomosis and ileos-
tomy irrigation with pancreatic enzymes. Ann Surg
50:835–836
Blackman SM, Deering-Brose R, McWilliams R et  al
(2006) Relative contribution of genetic and nongenetic
modifiers to intestinal obstruction in cystic fibrosis.
Gastroenterology 131(4):1030–1039
Carlyle BE, Borowitz DS, Glick PL (2012) A review of
pathophysiology and management of fetuses and neo-
nates with meconium ileus for the pediatric surgeon. J
Pediatr Surg 47(4):772–781
Choudhry MS, Grant HW (2006) Small bowel obstruc-
tion due to adhesions following neonatal laparotomy.
Pediatr Surg Int 22(9):729–732
Copeland DR, St Peter SD, Sharp SW et  al (2009)
Diminishing role of contrast enema in simple meco-
nium ileus. J Pediatr Surg 44:2130–2132
De Oronzo MA (2011) Hyperechogenic fetal bowel: an
ultrasonographic marker for adverse fetal and neonatal
outcome? J Prenat Med 5(1):9–13
Dodd K (1936) Intestinal obstruction due to meconium
ileus in a newborn infant. J Pediatr 9:486–491
Escobar MA, Grosfeld JL, Burdick JJ, Powell RL, Jay
CL, Wait AD et  al (2005) Surgical considerations in
cystic fibrosis: a 32-year evaluation of outcomes. J
Surg 138(4):560–572
Fakhoury K, Durie PR, Levison H et al (1992) In the past
meconium ileus was believed to be associated entirely
to CF. Meconium ileus in the absence of cystic fibro-
sis. Arch Dis Child 67:1204–1206
Fanconi G, Uehlinger E, Knauer C (1936) Das Coeliakie-­
syndrom bei angeborener zystischer Pankreas fibro-
matose und Bronchiektasien. Wien Med Wchnschr
86:753–756
Farber S, Shwachman H, Maddock CL (1943) Pancreatic
function and disease in early life. I.  Pancreatic
enzyme activity and the celiac syndrome. J Clin Invest
20:827–833
Farrell PM, White TB, Ren CL, Hempstead SE, Accurso
F, Derichs N, Howenstine M, McColley SA, Rock
M, Rosenfeld M, Sermet-Gaudelus I, Southern KW,
Marshall BC, Sosnay PR (2017) Diagnosis of cystic
fibrosis: consensus guidelines from the cystic fibrosis
foundation. J Pediatr 181:S4–S15
Farrelly PJ, Charlesworth C, Lee S, Southern KW, Baillie
CT (2014) Gastrointestinal surgery in cystic fibrosis: a
20-year review. J Pediatr Surg 49(2):280–283
Gorter RR, Karimi A, Sleeboom C et al (2010) Clinical
and genetic characteristics of meconium ileus in
newborns with and without cystic fibrosis. J Pediatr
Gastroenterol Nutr 50:569–572
Hiatt RB, Wilson PE (1948) Celiac syndrome; therapy of
meconium ileus, report of eight cases with a review of
the literature. Surg Gynecol Obstet 87:317–327
Jawaheer J, Khalil B, Plummer T, Bianchi A, Morecroft
J, Rakoczy G, Bruce J, Bowen J, Morabito A (2007)
V. Solari and M. Rivosecchi
Primary resection and anastomosis for complicated
meconium ileus: a safe procedure? Pediatr Surg Int
23(11):1091–1093
Johnson JA, Bush A, Buchdahl R (2010) Does presenting
with meconium ileus affect the prognosis of children
with cystic fibrosis? Pediatr Pulmonol 45:951–958
Karimi A, Gorter RR, Sleeboom C, Kneepkens CM,
Heij HA (2011) Issues in the management of sim-
ple and complex meconium ileus. Pediatr Surg Int
27(9):963–968
Keckler SJ, Peter SD, Spilde TL et al (2008) Current sig-
nificance of meconium plug syndrome. J Pediatr Surg
43(5):896–898
Kerem B, Rommens JM, Buchanan JA, Markiewicz D,
Cox TK, Chakravarti A et  al (1989) Identification
of the cystic fibrosis gene: genetic analysis. Science
245:1073–1780
Kornblith BA, Otani S (1929) Meconium ileus with con-
genital stenosis of the main pancreatic duct. Am J Path
5:249–261
Kopp BT, Nicholson L, Paul G, Tobias J, Ramanathan C,
Hayes D Jr (2015) Geographical variations in cystic
fibrosis: an analysis of the U.S. CF Foundation regis-
try. Pediatr Pulmonol 50(08):754–762
Landsteiner K (1905) Darmverschluss durch eingedick-
tes Meconium Pankreatitis. Zentralbl Allg Pathol
16:903–907
Mak GZ, Harberg FJ, Hiatt P, Deaton A, Calhoon R, Brandt
ML (2000) T-tube ileostomy for meconium ileus: four
decades of experience. J Ped Surg 35(02):349–352
Meyerholz DK, Stoltz DA, Pezzulo AA, Welsh MJ (2010)
Pathology of gastrointestinal organs in a porcine model
of cystic fibrosis. Am J Pathol 176(3):1377–1389
Noblett HR (1969) Treatment of uncomplicated meco-
nium ileus by gastrografin enema: a preliminary
report. J Pediatr Surg 4:190–197
Olsen MM, Luck SR, Lloyd-Still J et  al (1982) The
Spectrum of meconium disease in infancy. J Pediatr
Surg 17:479–481
Ooi CY, Durie PR (2016) Cystic fibrosis from the gas-
troenterologist’s perspective. Nat Rev Gastroenterol
Patol 13(3):175–185
Quinton PM (2008) Cystic fibrosis: impaired bicarbonate
secretion and mucoviscidosis. Lancet 372:415–417
Ramsey BW, Davies J, McElvaney NG et  al (2011) A
CFTR potentiator in patients with cystic fibrosis and
the G551D mutation. N Engl J Med 365:1663–1672
Ratjen F, Bell SC, Rowe SM, Goss CH, Quittner AL, Bush
A (2015) Cystic fibrosis Nat Rev Dis Primers 1:15010
Riordan JR, Rommens JM, Kerem B, Alon N, Rozmahel
R, Grzelczak Z et al (1989) Identification of the cystic
fibrosis gene: cloning and characterization of comple-
mentary DNA. Science 245(4922):1066–1073
Rommens JM, Iannuzzi MC, Kerem B, Drumm ML,
Melmer D, Dean M et al (1989) Identification of the
cystic fibrosis gene: chromosome walking and jump-
ing. Science 245(4922):1059–1065
Rosen BH, Chanson M, Gawenis LR, Liu J, Sofoluwe A,
Zoso A et al (2018) Animal model systems for study-
ing cystic fibrosis. J Cyst Fibros 17(2):S28–S34
65  Meconium Ileus
Ruiz MJ, Thatch KA, Fischer JC et  al (2009) Neonatal
outcomes associated with intestinal abnormali-
ties diagnosed by fetal ultrasound. J Pediatr Surg
44:71–74
Santulli T, Blanc WA (1961) Congenital atresia of the
intestine: pathogenesis and treatment. Ann Surg
154(6):939–948
Scotet V, Dugueperoux I, Audrezet MP et al (2010) Focus
on cystic fibrosis and other disorders evidenced in
fetuses with sonographic findings of echogenic bowel:
891
16-year report from Brittany. France Am J Obstet
Gynecol 203(592):e1–e6
Sharma N, Cutting GR (2020) The genetics and genomics
of cystic fibrosis. J Cyst Fibros 19:S5–S9
Smith VV, Schaeppi MG, Bisset M, Kiparissi F, Jaffe A,
Milla PJ, Lindley KJ (2009) Lymphocytic leiomyosi-
tis and myenteric ganglionitis are intrinsic features of
cystic fibrosis: studies in distal intestinal obstruction
syndrome and meconium ileus. J Pediatr Gastroenterol
Nutr 49:42–51
 Duplications of the Alimentary
Tract
K. Hughes, A. Mortell, and Prem Puri
66.1 Introduction
Duplications of the alimentary tract were first
described by Calder (1733). These rare spherical
or tubular structures can occur anywhere in the
tract from the mouth to anus. In 1937, William
Ladd introduced the term ‘alimentary tract dupli-
cation’, in the hope of clarifying the nomencla-
ture, which had previously included descriptive
terms, such as enteric or enterogenous cysts,
giant diverticula; ileal, jejunal or colonic duplex
and an unusual Meckel’s diverticulum. Ladd pro-
posed that the unifying term ‘alimentary tract
duplications’ be applied to congenital anomalies
that involved the mesenteric side of the associ-
ated alimentary tract and shared a common blood
supply with the native bowel. Ladd described
three common characteristics of these lesions, the
presence of a well-developed smooth muscle
coat, an epithelial lining consistent with the ali-
mentary tract and attachment to some portion of
K. Hughes · A. Mortell
Children’s Health Ireland (CHI), Dublin, Ireland
P. Puri (*)
Department of Pediatric Surgery, Beacon Hospital,
and University College Dublin, Dublin, Ireland
e-mail: prem.puri@ucd.ie
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_66
66
the alimentary tract. Most duplications may sim-
ply be called ‘enterogenous cysts’, since there is
actual doubling of the alimentary tract in only
very few cases and therefore deserving of the
name ‘duplication’.
66.2 Incidence
Duplications of the alimentary tract are rare.
Schalamon et al. (2000) reported an overall inci-
dence of 1:4500. Table  66.1 summarizes the
larger published series of duplications. In many
cases, the numbers of patients reported represent
up to 40 years of work in these centres. Only a
small percentage of the total reported actually
present in the neonatal period; however, more
than 80% of the cases present before the age of 2
(MacPherson 1993). According to Kumar et  al.
(2005), synchronous alimentary tract duplica-
tions can be found in up of 20% of patients.
893
 894

Table 66.1  Incidence and location of duplication cysts

Author Total number No. of neonates Oesophageal Cervical Mediastinal Thoracoabdominal Gastric Duodenal Jejunal/ileal Caecal Colonic Rectal
Carachi 21 10 – – – – – – – – – –
Erginel et al. 40 – 2 – – – 2 3 24 4 5 –
Favara et al. 37a – 9 (3) (4) (2) 3 4 20 – 4 –
Guérin et al. 114 – 16 – – – 14 7 47 25 4 1
Gross 68 20 17 (1) (13) (3) 2 4 32 – 9 4
Okur et al. 32 – 3 – – (2) 2 5 14 4 3 1
Scarpa et al. 41b – 43 (3) (35) (5) – – – – – –
Sieber 25c – 5 – (5) – 4 2 16 – 5 –
Sujka et al. 35 – 5 – – – 3 2 14 7 2 –
Grosfeld et al. 20 – 4 – (4) – 1 – 9 – 4 –
Wren 22d – 5 – (3) (2) 1 2 12 – 3 3
Lister and Vaos 32 24 3 – (3) – 1 – 20 – 5 3
Holcomb et al. 96e 36 24 (1) (20) (3) 8 2 47 – 20 –
a
One patient had three duplications
b
Two patients each had two duplications
c
One patient had two, one had three, and one had five duplications
d
Two patients each had two duplications
e
96 patients had 106 duplications
K. Hughes et al.
66  Duplications of the Alimentary Tract
66.3 Etiopathogenesis
Numerous theories have been developed to
account for the multitude of gastrointestinal (GI)
tract duplications. In 2000, Stern and Warner out-
lined the most widely held theories regarding GI
duplication. They proposed that, embryologi-
cally, duplications could be categorized into fore-
gut, midgut and hindgut (Fig.  66.1). Foregut
duplications include the pharynx, respiratory
tract, oesophagus, stomach and the first portion
and proximal half of the second portion of the
duodenum. Midgut duplications include the dis-
tal half of the second part of the duodenum, the
jejunum, ileum, caecum, appendix and the
ascending colon and the proximal two thirds of
the transverse colon. The hindgut is composed of
duplications of the distal third of the transverse
colon, the descending and sigmoid colon, the rec-
tum and anus and components of the urological
system. In one series published by Ildstad et al.
Fig. 66.1 Diagram
demonstrating the
common locations and
incidence of
gastrointestinal
duplications
Midgut 50%
Jejunal 10%
IIeal 35%
Appendiceal 2%
Cecal 3%
895
(1988), 39% of duplications involved the foregut,
whereas 61% represented duplications of both
mid and hindgut.
66.3.1 Partial Twinning
Certain duplications appear to represent partial
twinning, particularly the tubular duplications of
the terminal ileum and colon (Ravitch et  al.
1953). There is a wide spectrum of abnormalities,
from complete twinning of the lower trunk and
extremities to mere doubling of the lumen of
hindgut structures. These lesions are often asso-
ciated with duplication of the lower urinary tract
(McPherson et  al. 1969; Ravitch et  al. 1953).
Many rare examples of abortive cephalic twin-
ning have also been described (Dykes et  al.
1988). When there is complete doubling of the
colon, one or both lumens may open as a fistula
into the perineum or into the genitourinary tract
Foregut 36%
Oesophageal 19%
Thoracoabdominal 4
Gastric 9%
Duodenal 4%
Hindgut 14%
Colonic 7%
Rectal 5%
896
and may be associated with an imperforate anus.
Doubling of the anus, vagina and bladder have all
been detailed and often can be associated with
other severe deformities, such as double spines or
two heads.
66.3.2 Split Notochord
The most satisfactory of several theories of the
origin of GI duplications is that relating to the
development of the neurenteric canal. In 1943,
Saunders noted that thoracic duplications are fre-
quently associated with abnormalities of the cer-
vical and thoracic vertebrae. These duplications
may be attached to the vertebral bodies or con-
nected to the spinal canal. These findings gave
rise to the Bentley and Smith ‘split notochord
theory’ (Bentley and Smith 1960). Initially, the
embryo has two layers, ectoderm and endoderm.
The mesoderm forms between the two, but for a
short time, these two layers remain adherent. A
transient opening (the notochordal plate) appears,
connecting the neural ectoderm with the intestinal
endoderm. This notochordal plate normally
migrates dorsally and becomes ‘pinched’ off from
the endoderm by the ingrowth of mesodermal
cells from each side. If the notochordal plate fails
to migrate as a result of adhesions to the endoder-
mal lining, the spinal canal cannot close ventrally
and a tract resembling a diverticulum is estab-
lished with the primitive gut. This tract may
remain open, leaving a fistula between the gut and
the spinal canal, or closed, leaving only a fibrous
tract. However, in the majority of cases, it disap-
pears completely, leaving only the duplication of
the GI tract. This theory explains the formation of
thoracic and caudal duplications, which may be
associated with vertebral anomalies. However, the
absence of spinal defects in many alimentary tract
duplications makes this theory less tenable as a
unifying model of origin.
66.3.3 Embryonic Diverticula
and Recanalization Defects
Lewis and Thyng (1970) found tiny bands of
intestinal epithelium protruding into the subepi-
K. Hughes et al.
thelial connective tissue in human embryos
(4–23 mm) and in animal embryos. The identifi-
cation of numerous diverticula in the intestines of
embryos led to the proposal of extension of the
diverticula into duplications. The frequent ileal
position of these diverticula is congruous with
the frequent ileal location of human GI duplica-
tions. Although this theory could explain duplica-
tions in the absence of spinal anomalies, it fails to
account for the variability of the mucosal lining
and specifically, for the frequency of heterotopic
gastric mucosa. Furthermore, the diverticula
identified in this pathological series were located
throughout the bowel circumference as opposed
to the general locations of duplications on the
mesenteric side of the intestine.
The occurrence of tubular duplications would
also not be explained by this theory. Bremer
(1944) believed that the abnormal recanalization
of the intestinal lumen after the solid stage of
development of the primitive gut in the sixth–sev-
enth week resulted in duplications. Such duplica-
tions, however, would not be confined to the
mesenteric side of the bowel. Also opposing this
theory is the finding that the solid stage of devel-
opment in the human does not usually extend
beyond the duodenum.
In 1961, Mellish and Koop proposed an envi-
ronmental theory, which held that trauma or
hypoxia could induce duplications and twinning
in lower orders. Based on the work of Louw
(1959), they concluded that vascular insuffi-
ciency could lead to the recognized GI duplica-
tions seen in humans. Additionally, intrauterine
vascular accidents are known precipitators of the
other congenital anomalies, such as gastrointesti-
nal atresias.
66.4 Pathology
As defined by Ladd 1937, duplications are hollow
structures that involve the mesenteric side of the
associated GI tract. They tend to share a common
muscular wall and blood supply with the attached
mature bowel, although each has its own separate
lining (Smith 1960). They are usually isolated
lesions and are more often cystic than tubular with
a variable size. The lesions have a muscular coat
66  Duplications of the Alimentary Tract
in two layers and are usually lined with epithe-
lium similar to that found in the associated por-
tion of the alimentary tract. The duplications,
however, are occasionally lined with heterotopic
epithelium; colonic mucosa has been described at
the base of the tongue, and the sinuses lined with
gastric mucosa have been found near the anus
(Wrenn 1992). Duplications containing gastric
mucosa are at risk of peptic ulceration, perfora-
tion and haemorrhage. Patches of ectopic gastric
mucosa along the GI tract may represent the mild-
est manifestation of duplication abnormalities.
Ectopic pancreatic tissue has been reported in
duplications of the stomach, ileum and colon
(Holcomb et al. 1989). The contents of a duplica-
tion vary with the type of epithelial lining of the
structure. If an opening is present, the duplication
contents will be similar to that of the adjacent
intestinal tract. The communication between the
two structures is rare, and the cysts usually contain
chyle or mucous. Multiple duplications can occur
in the same patient (Wrenn 1992). There is an
increased incidence of other associated anomalies
such as vertebral anomalies, myelomeningocoele,
imperforate anus, exomphalos, malrotation of the
bowel, genital anomalies, polysplenic syndrome
and duodenal atresia (Kremer et  al. 1970). No
genetic tendency has been demonstrated.
66.5 Diagnosis and Management
The presentation of duplications of the alimen-
tary tract is highly variable; therefore they are
best thought of in the context of position, as anat-
omy and associated symptoms will differ between
locations (Puri 2003). Duplications associated
with the foregut may have associated respiratory
symptoms, whereas duplications in the midgut
and hindgut can present with obstructive symp-
toms, perforation, nausea, emesis or haemor-
rhage (Scarpa et al. 2018; Sujka et al. 2018;
Erginel et al. 2017). The presentation can be with
acute or chronic symptoms, and there are also
reports of the diagnosis being made incidentally
in late adulthood in patients who have remained
asymptomatic (O’Donnell et al. 2005).
897
As with the detection of other congenital
abnormalities, the frequency of prenatal diagno-
sis of duplications of the alimentary tract has
increased significantly. This has coincided with
the development of higher resolution prenatal
ultrasound scans. Two sonographic signs, in par-
ticular, are highly suggestive of enteric duplica-
tion: the presence of a ‘double wall‘ and the
presence of peristalsis (Thakkar et  al. 2015).
With these identifying features and higher-­
resolution imaging, duplication cysts have been
detected as early as 12 weeks of gestation (Chen
et al. 2002).
With regard to the timing of surgery, a
recent systematic review indicated that almost
half of the infants diagnosed prenatally become
symptomatic early in life and are resected.
However, in asymptomatic patients, Fahy et al.
(2019) suggest consideration of close observa-
tion and delayed prophylactic resection in
infancy.
66.6 Differential Diagnosis
The presentation of duplication cysts is highly
variable with a range of symptoms, both acute
and chronic, which are related to the location,
size, type of duplication and the presence of ecto-
pic mucosa. As a result, the list of differential
diagnoses is broad.
When considering a differential diagnosis for
duplication cysts, it is helpful to categorize them
based on their anatomical locations as foregut,
midgut or hindgut. The main differentials for
foregut duplications include bronchogenic cysts,
pericardial cysts, pulmonary sequestration and
pancreatic pseudocysts. As for midgut duplica-
tions, the alternative pathologies to be consid-
ered would be choledochal, mesenteric and
omental cysts, or a Meckel’s diverticulum. With
regard to hindgut duplications, there are less
pathologies to consider, with the main differen-
tials being intestinal lymphoma, loculated bowel
perforation, gastrointestinal stromal tumour
(GIST) or very rarely a primary colonic malig-
nancy (adenocarcinoma).
898
66.7 Management
66.7.1 Oesophageal Duplication
The oesophagus is a relatively common site for
foregut duplications (19%) with the majority
being intramural, non-communicating cystic
structures related to the right side of the oesopha-
gus. Patients often present late in childhood as
they cause relatively few symptoms; however,
cervical oesophageal duplications can cause sig-
nificant respiratory distress requiring urgent air-
way control and surgery. Some lesions do not
share a common wall with the oesophagus and
can be removed easily through open or minimally
invasive techniques as demonstrated by Perger
et al. (2006). Plain radiographs may show an air-
or fluid-filled structure adjacent to the oesophagus
although this is not usually enough to confirm the
diagnosis. Contrast studies can provide useful
information regarding the mass effect of the lesion
and whether or not their lumens c­ommunicate.
Ultrasound and computed tomography (CT) are
useful for establishing a diagnosis and also for
ruling out multiple lesions, which can be present
in 10–20% of cases. Technetium scans (99mTc)
may reveal the presence of heterotopic gastric
mucosa in the case of gastrointestinal bleeding.
Treatment
The surgical approach depends largely on the
location of the cyst (Fig.  66.2). Excision of the
mass is usually curative. Winslow et  al. (1984)
advise that if complete excision is not possible,
then stripping of the mucosa from the cyst wall
will allow for the obliteration of the cyst cavity
and resolution of symptoms. Cervical oesopha-
geal duplications can be removed through a
supraclavicular incision, with particular attention
being paid to the vagus and phrenic nerves as
well as the thoracic duct to avoid unnecessary
damage. Intrathoracic duplications are resected
through a standard posterolateral thoracotomy or
a thoracoscopic approach (Scarpa et al. 2018). A
chest drain may be left in situ but is not always
required.
K. Hughes et al.
Fig. 66.2  Diagram demonstrating resection margin for
simple cystic oesophageal duplication cyst
66.7.2 Thoracoabdominal
Duplication
Thoracoabdominal duplications are rare repre-
senting only 4% of all gastrointestinal duplica-
tions. They often lie separate from the oesophagus,
more often on the right than the left side, but may
be attached to other important structures such as
the aorta, azygous vein and tracheobronchial
tree. They frequently lie in the posterior medias-
tinum and pass through the diaphragm to com-
municate with the stomach, duodenum or small
bowel. The imaging studies employed are similar
to those for oesophageal duplications, with spe-
cial attention being paid to the imaging of the
vertebral column/spinal cord for a possible intra-
spinal component. CT and/or magnetic resonance
imaging (MRI) are particularly useful in this
regard, especially if neurological symptoms of
spinal cord compression or bony spinal abnor-
malities are present.
Treatment
These challenging duplications require the resec-
tion of the thoracic and abdominal components
through two different open procedures (Fig. 66.3),
or alternatively, they may be dealt with by a com-
bined thoraco-laparoscopic approach. Although
the abdominal portions are often asymptomatic,
the thoracic components can cause symptoms as
66  Duplications of the Alimentary Tract 899

a result of mass effect on the lungs and airway.

The presence of gastric mucosa within the tho-
racic duplication cyst can lead to peptic ulcer-
ation and possible erosion into the lung
parenchyma, presenting with haemoptysis. This
complication may require a lobectomy. Once the
lesion is mobilized in the thorax, it is freed from
the posterior aspect of the diaphragm prior to
mobilization and removal of the abdominal
component.

66.7.3 Gastric Duplication

The stomach is one of the less common sites of

duplications, accounting for only 9% of all GI
duplications. Over 60% of cases are diagnosed
during the first year of life, with a significant
number (40%) appearing in the neonatal period
by the finding of a palpable cystic mass in the
upper abdomen accompanied by poor feeding,
vomiting, weight loss and failure to thrive Abrami
et  al. 1961. In rare cases, they undergo peptic
ulceration, and if the cyst communicates with the
stomach, haematemesis and melaena may be the
presenting feature. Rarely, a carcinoma may arise
within a gastric duplication cyst (Pruksapong
et al. 1979). Gastric outlet obstruction mimicking
hypertrophic pyloric stenosis is also a common
presentation of this duplication. Gastric duplica-
tions occur twice as often in females as in males
and are usually located along the greater curva-
ture (Table 66.2). Occasionally, these are pedun-
culated, but most are closed spherical cysts or
tubular structures.
It is often difficult to make a preoperative
diagnosis. Plain radiographs are usually non-­
contributory. A contrast study may show a com-
pression of the stomach, usually along the greater

Table 66.2  Locations of duplications of the stomach in

87 reported cases
Location No. of cases
Greater curvature 55
Lesser curvature 7
Anterior wall 9
Fig. 66.3 Diagram demonstrating open thoracic and
Posterior wall 9
abdominal approaches to a thoracoabdominal duplication
Other 7
cyst
900
curvature and in a small minority of cases may
demonstrate a connection between the stomach
and the duplication. In the presence of a commu-
nication, contrast may be retained in the duplica-
tion long after the remainder has passed from the
GI tract. Ultrasonography has been shown to be
useful in the diagnosis of gastric duplications.
Associated anomalies occur in 3% of gastric
duplications (Torma 1974). The most common is
another cyst, usually of the oesophagus. Dual
duplications of the stomach and pancreas have
been reported by Cloutier (1973). These are
thought to arise from an error in rotation of the
ventral pancreatic anlage.
Treatment
The management of gastric duplications is pri-
marily surgical because of the high incidence of
complications due to obstruction, bleeding or
peritonitis. As most duplications occur in the
greater curvature, a wedge of stomach is excised
together with the cyst and the gap closed with a
single or double layer of horizontal inverting
mattress sutures (Fig. 66.4). Partial gastrectomy
Fig. 66.4  Gastric duplication located at the greater cur-
vature. A wedge of stomach is excised together with the
cyst
K. Hughes et al.
should be avoided in children if possible, and if
necessary only 25–30% of the stomach should be
resected because of the associated long-term
complications.
When the resection of the adjoining stomach
is impractical, as in long tubular duplications of
the greater curvature, the main part of the dupli-
cation is excised, and the mucosa is stripped off.
The remaining seromuscular cuff can be sutured
over the denuded area after checking that the
common wall between the stomach and duplica-
tion has not been perforated, by insufflating the
stomach with air. The use of a linear stapler to
divide the common wall along the length of the
greater curvature has also been described.
66.7.4 Pyloric Duplications
True pyloric duplications are extremely rare, with
very few being reported in the English literature
and most of these presenting within the first week
of life (Grosfeld et al. 1970). They simulate the
symptoms and signs of hypertrophic pyloric ste-
nosis. Vomiting, weight loss and a palpable
abdominal mass are the main findings. There are
certain physical features which are consistent
with duplication: the mass is usually large and
smooth, in contrast to the smaller and often more
mobile ‘olive’ mass in hypertrophic pyloric ste-
nosis. Due to the non-specific findings on physi-
cal examination, radiographic procedures are
essential to aid in making the diagnosis. Plain
film radiography may show signs of gastric outlet
or duodenal obstruction with a lack of distal
bowel gas, or rarely calcification within a cyst
wall. Ultrasonography may demonstrate an inner
echogenic mucosal layer and outer hypoechoic
muscular layer differentiating the duplication
from a mesenteric cyst. Contrast studies may
help differentiate the duplication from pyloric
stenosis. If there is a clinical concern, then preop-
erative endoscopic retrograde cholangiopancrea-
tography (ERCP), percutaneous transhepatic
cholangiography (PTC) or magnetic resonance
cholangiopancreatography (MRCP) should be
performed to evaluate the involvement of the bili-
ary and pancreatic ducts.
66  Duplications of the Alimentary Tract
Treatment
Of the cases of pyloric duplication reported, the
majority underwent simple surgical excision
after opening the pyloric canal longitudinally and
then closing transversely (Grosfeld et al. 1970).
In cases where there is a risk of damage to pan-
creatic or bile ducts, an acceptable alternative is
to drain the cyst into the duodenum or into a
Roux limb of upper small bowel.
66.7.5 Duodenal Duplications
The duodenum is involved in only 4% of all
duplications and are usually posterior to the
duodenum and do not communicate with the
bowel lumen (Alfred et al. 1980). Vomiting sec-
ondary to partial or complete duodenal obstruc-
tion and an upper abdominal mass are present in
the majority of cases. They may present with
haematemesis or perforation as gastric mucosa
is present in 10–15% of cases (Bower et  al.
1978). Alternatively, because of their location,
they may present with biliary obstruction or
pancreatitis. If the duplication is of sufficient
size, it may appear on plain radiographs as a
large opacity in the right side of the abdomen
displacing adjacent bowel. Contrast studies may
show the duodenum to be displaced upwards
with a ‘beak-like’ appearance due to the com-
pression of the lumen, and the presence of con-
trast within the cyst confirms communication
between lumens. Ultrasonography may show a
cystic lesion below the liver and a classical
‘double-layered’ appearance or ‘muscular rim
sign’.
Treatment
Due to the potential presence of gastric mucosa
in the duplication cyst, these lesions should, if
possible, be dissected from the duodenum and
excised. Intraoperative cholangiography may
help determine the relationship of the cyst to the
bile and pancreatic ducts.
In cases where the lesion is extensive, or if
complete excision of the cyst may compromise
the biliary system, then cystoduodenostomy may
be performed. The cyst may also be only partially
901
excised, stripping off all the lining mucosa and
leaving that part of the cyst which is adherent to
the duodenum or pancreas. Schils et  al. (2018)
report successful endoscopic resection of symp-
tomatic duodenal duplication cysts; however,
open or minimally invasive techniques remain
more common.
66.7.6 Duplications of the Small
Intestine
Small bowel duplications constitute 45% of all
alimentary tract duplications Stringer (2006).
The vast majority of small bowel duplications are
spherical cysts in the terminal ileum. Jejunal and
ileal cysts are found on the mesenteric side of the
bowel sharing a common muscular wall with the
adjacent bowel. They may cause obstruction by
extraluminal compression, by acting as a lead
point for intussusception, or occasionally by
causing a volvulus.
Tubular duplications have the same features as
the cystic variety but are more likely to commu-
nicate with the normal lumen of the intestine and
to contain gastric mucosa. Pancreatic mucosa has
also been described in these duplications. Tubular
duplications can range from a few millimetres to
the whole length of the small bowel. The com-
munication may be at the cephalad end which
will cause the duplication to become grossly dis-
tended with intestinal contents or, if at the caudal
end, will allow the duplication to drain freely.
Communication at several different points may
be present.
Haemorrhage occurs most often in tubular
duplications, but perforation has been reported as
well (Doyle et al. 1988). Plain abdominal radio-
graphs (Fig.  66.5) may show non-specific dis-
placement of bowel gas shadows by the cyst or
signs of intestinal obstruction. Ultrasonography
may differentiate between a mesenteric and a
duplication cyst. A contrast study may demon-
strate displacement of the bowel (Fig. 66.6).
Treatment
Cystic duplications are relatively straightforward
to deal with and usually involve simple resection
902
Fig. 66.5  Supine plain film of the abdomen in a 1-day-­
old baby showing a large soft tissue mass in the right
upper and central abdomen displacing bowel loops to the
left
of the cyst and adjacent bowel with end-to-end
anastomosis (Fig.  66.7). Tubular duplications,
when very short, can be resected as with cystic
lesions; however, the majority involve a consider-
able length of small bowel, and the resulting
resection can be complex.
When managing long tubular duplications,
Wrenn (1992) suggested coring out the mucosal
lining of the lesion through a series of seromus-
cular incisions in the wall of the duplication,
thereby leaving the wall and blood supply of the
normal small bowel intact.
Norris et al. (1986) described a technique sim-
ilar to Bianchi’s (1982) bowel lengthening proce-
dure, where the separation of two leashes of
blood vessels passing to each side of the small
intestine is accomplished. Using this method, the
entire mucosa and almost the entire muscle wall
can be excised. The remaining cuff of muscle
K. Hughes et al.
Fig. 66.6  A contrast study in the same infant demonstrat-
ing a space-occupying lesion displacing bowel. At a large
ileal duplication cyst was found
wall can be oversewn, preserving the blood sup-
ply to the normal bowel.
In Bishop and Koop (1964) described the
technique of anastomosing the distal end of the
duplication to adjacent normal intestine, allowing
free drainage of the contents. Malignant change
in the mucosa has, however, been described as a
late complication of this procedure.
66.7.7 Colonic Duplications
Colonic duplications are among the rarest
reported and are frequently diagnosed in infancy
with some reports suggesting a female predilec-
tion. McPherson et al. (1969) proposed a simple
classification of colonic duplications: type I mes-
enteric cysts, type II diverticular and the more
common type III tubular colonic duplication. A
number of aetiological factors may be involved in
66  Duplications of the Alimentary Tract
Fig. 66.7  Cystic small bowel duplication with resection
margin incorporating duplication and adjacent normal
bowel
the development of the ‘double colon’. The most
valid theory suggests the division of the hindgut
into two parts at a stage during which the anlage
possessed a multi-organ developmental potential
(McPherson et al. 1969). The hindgut anlage nor-
mally forms the distal ileum, colon, rectum, blad-
der and urethra. The division of the anlage at the
same initial stage could therefore be responsible
for the duplication of the lower urinary tract too.
Simple cysts (type I) and diverticula (type II)
occasionally result. They can be identified on
plain radiographs or on contrast studies. A con-
trast enema may demonstrate a communication
between the colon and the duplication in types II
and III.  Associated genitourinary and lumbosa-
cral spine abnormalities can also be demonstrated
on the appropriate radiographic studies, particu-
larly when dealing with type III duplications.
Isotope scans are rarely of benefit with colonic
duplications, as they rarely contain ectopic
mucosa.
The complete duplication of the colon is usu-
ally asymptomatic in the neonatal period unless
the duplication of the anus or an abnormal ori-
fice, in addition to the normal orifice in the
perineum, is present. One or both orifices at the
903
Fig. 66.8  Total colonic duplication from the caecum to
the rectum lying lateral to normal bowel
distal end of the colon may end as rectovaginal or
rectourethral fistulae.
Treatment
Surgery for colonic duplication is rarely indi-
cated in the neonatal period unless there are com-
plications, e.g. obstruction or an associated
imperforate anus. All cystic and most tubular
colonic duplications can be dealt with by simple
resection and anastomosis. With rare total colonic
duplication (Fig. 66.8), the principle aim of man-
agement is to end up with two colons draining
through one anal orifice. If one part of the colon
has already reached the perineum, then the other
colon is divided and anastomosed to its partner.
This can be achieved by using a linear stapling
device. If neither colon reaches the perineum,
then a formal pull-through procedure will be
required. Neonatal management in any of these
situations is confined to fashioning a defunction-
ing colostomy to achieve the drainage of both
colons.
904
66.7.8 Rectal Duplications
Approximately 70 cases of rectal duplication
have now been reported in the literature, com-
prising only 5% of all gastrointestinal duplica-
tions. More than 50% of these have been
examples of hindgut twinning (La Quaglia et al.
1990). Rectal duplications often present in the
neonatal period with a fistula or perineal mucosal
swelling extending to the perianal area (Fig. 66.9).
The embryogenesis of rectal duplication cysts
is attributed to a ‘pinching off’ of a diverticulum
in the 20–30 mm embryo, in contrast to the ‘cau-
dal twinning’, which occurs in the 10 mm embryo
and is associated with complex hindgut
anomalies.
The presentation of these cystic lesions varies
with their size and the resulting mass effect. In
addition, they may present with a fistula, infec-
tion or ulceration due to the presence of gastric
mucosa. The duplication cyst usually forms in
the retrorectal space (Fig.  66.10) and contains
mucous which may present with infection. No
cases of a fistula between the rectum and urinary
Fig. 66.9  Rectal duplication presenting as a perianal
mucosal swelling in the neonatal period
K. Hughes et al.
tract have been described. Malignant degenera-
tion has been reported in the rectal duplication
from the fourth decade onwards.
Treatment
The treatment of rectal duplication cyst is a surgi-
cal excision or fenestration of the common wall.
Depending on the anatomical variations, a trans-
anal or transcoccygeal (Kraske) approach can be
employed. For longer or more complicated cysts,
a longer posterior sagittal incision will provide
better exposure. As with other duplications, it is
of prime importance to remove all mucosa in the
duplication. The muscularis can be left in situ.
Associated anomalies such as presacral
tumours (16%) and anorectal malformations
(21%) are frequently described in the literature
(Jasquier et al. 2001). The management of these
lesions may be difficult and often requires preop-
erative evaluation of both the GI and genitouri-
nary tract. Continence of both systems is
imperative, and therefore, treatment strategies
must be individualized based on the findings of
each patient.
Fig. 66.10  Sagittal view of a rectal duplication cyst pres-
ent in the retrorectal space
66  Duplications of the Alimentary Tract
66.8 Conclusion
It is clear that duplications of the GI tract repre-
sent a diverse and complex group of anomalies.
The presentation of these rare lesions is highly
variable and can occur with acute or chronic
symptoms. Alternatively, the diagnosis can be
made incidentally at any age. Duplications
associated with the foregut typically have asso-
ciated respiratory symptoms, whereas duplica-
tions in the midgut and hindgut can present
with obstructive symptoms, perforation, nau-
sea, emesis or haemorrhage. These lesions are
usually cystic and tend to contain mucosa simi-
lar to the adjacent bowel; however, ectopic tis-
sue, most commonly gastric mucosa, can be
present.
Surgical treatment depends on the site and
type of the cyst. Small duplications in readily
accessible areas (i.e. small intestine) may be
excised with a portion of adjacent normal bowel.
In other locations, where resection would endan-
ger adjacent structures, simple anastomosis
between the cyst and normal intestine can be per-
formed, provided there is no gastric mucosa in
the cyst. If bleeding has been a persisting com-
plaint, one can assume the presence of ectopic
gastric mucosa. If resection is contraindicated,
the lining mucosa may be stripped from the cyst,
leaving the muscle wall in situ.
At present, there are no controlled studies to
determine if the outcomes with minimally inva-
sive surgery are superior when compared to open
surgery techniques. This is, in part, limited by the
rarity of the disorder and would require large
multicentre studies to assess fully; however, there
are increasing numbers of case studies showing
favourable results with a minimally invasive
approach.
References
Abrami G, Dennison WM (1961) Duplication of the stom-
ach. Surgery 49:794–798
Alfred BA, Armstrong P, Franken FA et  al (1980)
Calcification associated with duodenal duplications in
children. Radiology 134:647–648
905
Bentley JFR, Smith JR (1960) Development posterior
enteric remnants and spinal malformations. Arch Dis
Child 35:76–86
Bianchi A (1982) Intestinal loop lengthening–a technique
for increasing small bowel length. J Pediatr Surg
15:145–151
Bishop HE, Koop CE (1964) Surgical management
of duplication of the alimentary tract. Am J Surg
107:434–442
Bower RJ, Sieber WK, Kiesewetter WB (1978) Alimentary
tract duplication in children. Ann Surg 188:669–671
Bremer JL (1944) Diverticula and duplications of the
intestinal tract. Arch Pathol 38:132–140
Calder J (1733) Two examples of children born with pre-
ternatural conformations of the guts. In: Great medical
essays and observations, vol 1, 2nd edn, Edinburgh,
pp 203–206
Chen M, Lam YH, Lin CL et  al (2002) Sonographic
features of ileal duplication cyst at 12 weeks. Prenat
Diagn 22(12):1067–1070
Cloutier R (1973) Pseudocyst of pancreas secondary to
gastric duplication. J Pediatr Surg 8:67
Doyle SG, Doig CM (1988) Perforation of the jejunum
secondary to a duplication cyst lined with ectopic gas-
tric mucosa. J Pediatr Surg 23:1025–1026
Dykes EH, Kelleher J, Fogarty EE et al (1988) Surgical
treatment of a variant of dipygus. Pediatr Surg Int
3:200–202
Erginel B, Soysal FG, Ozbey H et al (2017) Enteric dupli-
cation cysts in children: a single-institution series
with forty patients in twenty-six years. World J Surg
41(2):620–624
Fahy AS, Pierro A (2019) A systematic review of prena-
tally diagnosed intra-abdominal enteric duplication
cysts. Eur J Pediatr Surg 29(1):68–74
Grosfeld JC, Boles ET, Reiner C (1970) Duplication of
the pylorus in the newborn: a rare cause of gastric out-
let obstruction. J Pediatr Surg 5:365–369
Holcomb GW, Gheisser A, O’Neill J et al (1989) Surgical
management of alimentary tract duplications. Ann
Surg 209:167–174
Ildstad ST, Tollerud DJ, Weiss RG et  al (1988)
Duplications of the alimentary tract. Clinical charac-
teristics, preferred treatment, and associated malfor-
mations. Ann Surg 208:184–189
Jasquier C, Dobremez E, Piolat C et al (2001) Anal canal
duplication in infants and children–a series of 6 cases.
Eur J Pediatr Surg 11:186–191
Kremer RM, Lepoff RB, Izamt RJ (1970) Duplication of
the stomach. J Pediatr Surg 5:360–364
Kumar R, Tripathi M, Chandrashekar N, Agarwala S,
Kumar A, Dasan JB et al (2005) Diagnosis of ectopic
gastric mucosa using 99Tcm-pertechnetate: Spectrum
of scintigraphic findings. Br J Radiol 78:714–720
La Quaglia MP, Feins N, Eraklis A, Hendren WH (1990)
Rectal Duplication. J Pediatr Surg 25:980–984
Ladd WE (1937) Duplications of the alimentary tract.
South Med J 30:363–371
906
Lewis FT, Thyng FW (1970) The regular occurrence of
intestinal diverticula in embryos of the pig, rabbit and
man. Am J Anat 7:505–519
Louw JH (1959) Congenital atresia and stenosis in the
newborn. Ann Roy Coll Surg Engl 25:209
MacPherson RI (1993) Gastrointestinal tract duplications:
clinical, pathologic, etiologic, and radiologic consid-
erations. Radiographics 13:1063–1080
McPherson AE, Trapnell JE, Airth GR (1969) Duplication
of the colon. Br J Surg 56:138–139
Mellish RWP, Koop CE (1961) Clinical manifestations of
duplication of the bowel. Pediatrics 27:397–407
Norris R, Bereton R, Wright V et al (1986) A new surgi-
cal approach to duplications of the intestine. J Pediatr
Surg 21:167–169
O’Donnell PL, Morrow JB, Fitzgerald TL (2005) Adult
gastric duplication cysts: a case report and review of
literature. Am Surg 71(6):522–525
Perger L, Azzie G, Watch L, Weinsheimer R (2006) Two
cases of thoracoscopic resection of esophageal dupli-
cation in children. J Laparoendosc Adv Surg Tech A
16:418–421
Pruksapong C, Donovan RJ, Pinit A et al (1979) Gastric
duplication. J Pediatr Surg 14:83–85
Puri P (2003) Duplications of the alimentary tract. In: Puri
P (ed) Newborn surgery, 2nd edn. Arnold, London,
pp 479–488
Ravitch MM (1953) Hindgut duplications; doubling
of the colon and genital urinary tracts. Ann Surg
137:294–298
Saunders RL (1943) Combined anterior and posterior
spina bifida in a living neonatal human female. Anat
Rec 87:255–275
Scarpa AA, Ram AD, Soccorso G, Singh M, Parikh
D (2018) Surgical experience and learning points
K. Hughes et al.
in the Management of Foregut Duplication Cysts.
Eur J Pediatr Surg 28(6):515–521. https://doi.
org/10.1055/s-­0037-­16072
Schalamon J, Schleef J, Höllwarth ME (2000) Experience
with gastro-intestinal duplications in childhood.
Langenbeck’s Arch Surg 385(6):402–405. https://doi.
org/10.1007/s004230000170
Schils N, Vandecaveye V, Van Overbeke L, Laleman W
(2018) Endoscopic treatment of a symptomatic duo-
denal duplication cyst. Endoscopy 50(7):E184–E185
Smith JR (1960) Accessory enteric formations: classifica-
tion and nomenclature. Arch Dis Child 35:87–89
Stern LE, Warner BW (2000) Gastrointestinal duplica-
tions. Semin Pediatr Surg 9:135–140
Stringer MD (2006) Gastrointestinal duplications. In: Puri
P, Höllwarth ME (eds) Pediatric surgery, springer sur-
gery atlas series. Springer-Verlag, Berlin, Heidelberg,
New York, pp 239–256
Sujka JA, Sobrino J, Benedict LA, Alemayehu H, Peter
SS, Hendrickson R (2018) Enteric duplication in chil-
dren. Pediatr Surg Int 34(12):1329–1332
Thakkar HS, Bradshaw C, Impey L, Lakhoo K (2015)
Post-natal outcomes of antenatally diagnosed intra-­
abdominal cysts: a 22-year single-institution series.
Pediatr Surg Int 31(2):187–190
Torma MJ (1974) Of double stomachs. Arch Surg
109:555–556
Winslow RE, Dykstra G, Scholten DJ, Dean RE (1984)
Duplication of the cervical esophagus. An unrecog-
nized cause of respiratory distress in infants. Am Surg
50(9):506–508
Wrenn E (1992) Alimentary tract duplications. In: Holder
T, Ashcraft K (eds) Pediatric surgery. WB Saunders,
Philadelphia, pp 455–456
 Necrotizing Enterocolitis
Heather L. Liebe, Henri R. Ford, Victoria Camerini,
and Catherine J. Hunter
67.1 Introduction
Necrotizing enterocolitis (NEC) is the most com-
mon life-threatening, surgical and medical emer-
gency, affecting the gastrointestinal tract
encountered in the neonatal intensive care unit.
NEC occurs in 2–5% of all preterm infants,
although the incidence and severity of NEC var-
ies inversely with gestational age. For example,
NEC occurs in approximately 10% of infants
with a gestational age of less than 32 weeks and a
birth weight of less than 1500 grams accounts on
average for 5–8% of intensive care admissions. It
has been noted that infants born earlier develop
NEC at a later chronological age. The average
age of onset of disease is 20.2  days for infants
born less than 30  weeks of gestation, whereas
disease onset is reduced to 13.8 days for infants
born at 31–33  weeks and 5.4  days for infants
born after 34 weeks of gestation.
H. L. Liebe · C. J. Hunter (*)
Children’s Hospital at Oklahoma University,
Oklahoma City, OK, USA
e-mail: Catherine-hunter@ouhsc.edu
H. R. Ford
University of Miami, Miami, CA, USA
e-mail: hford@chla.usc.edu
V. Camerini
Children’s Hospital Los Angeles,
Los Angeles, CA, USA
e-mail: vcamerini@chla.usc.edu
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_67
67
Epidemiological studies have identified mul-
tiple risk factors for NEC, although a history of
prematurity, hypoxia and asphyxia and the intro-
duction of formula feeding are characteristically
associated with infants that develop NEC. Despite
the predilection for premature infants, NEC has
also been described in term infants particularly
those with cyanotic heart disease. There is no
clear evidence to suggest that geographical ori-
gin or gender alters the incidence of NEC. Data
suggests that ethnicity may play a role in the
incidence of NEC, with Black and Hispanic
infants having an increased risk of developing
NEC.
NEC is characterized by intestinal inflamma-
tion accompanied by epithelial barrier disruption,
bacterial overgrowth and submucosal invasion. In
its most severe form, NEC is characterized by
full-thickness destruction of the intestinal wall,
leading to intestinal perforation, peritonitis, sep-
sis and death. Although the overall mortality for
patients with NEC ranges from 10 to 50%, it
approaches 100% in infants with the most severe
form of the disease, characteristically the small-
est and most premature infants. Moreover, infants
that recover from NEC may still require pro-
longed hospitalization due to complications from
disease, such as intestinal obstruction due to scar-
ring, short bowel syndrome and complete intesti-
nal failure, further impacting long-term survival,
growth and development.
907
908
67.2 Historical Overview
Descriptions of NEC date as far back as 1888,
when there are reports of intestinal perforations
in infants resulting in death. These isolated case
reports persisted until 1944, when Willi wrote a
case series of 62 neonates with ‘severe infectious
enteritis’. At this early time, aggressive operative
management was recommended, followed by
antibiotics, as the predominant theory was that
NEC was an overwhelming infection. It was not
until 1952, when Schmidt and Kaiser wrote their
case series of 85 infants that the term ‘enterocoli-
tis ulcerosa necroticans’ also known as ‘necrotiz-
ing enterocolitis’ was coined. This series failed to
find a consistent infectious aetiology and
described a series of neonates near or at term who
died from intestinal necrosis. Throughout this
time, the management of NEC was evolving to
include aggressive IV fluid resuscitation and ces-
sation of oral feeds. With the growth of special-
ized neonatal care and the opening of the first
dedicated neonatal intensive care unit in the
United States in 1960 at Yale New Haven
Hospital, there was a notable rise in the incidence
of NEC.  This was attributed to better neonatal
care, resulting in increased survival of younger
premature infants. In the 1960s, research began
in earnest to discover the pathophysiology, pre-
vention and treatment of NEC. It was during this
time that the proposal of a ‘multifactorial theory’
arose, in which risk factors, such as prematurity,
altered intestinal blood flow, bacterial coloniza-
tion and enteral feeding, were first mentioned.
The mortality of NEC at that time was estimated
by the CDC to be 40%. In 1979, the International
Classification of Disease established a code for
‘death from NEC’, which allowed for better
tracking and outcome analysis, leading to
improved clinical research. The 1980s, 1990s and
2000s brought about growth in understanding the
pathophysiology of NEC, thus leading to
improved prevention measures with better NICU
care. There was also a transition from early oper-
ative management to aggressive medical man-
agement with cessation of feeds, early antibiotics
and supportive care. While much progress has
been made in understanding NEC, it remains one
H. L. Liebe et al.
of the deadliest diseases in premature infants,
leading physicians and scientists to continue
seeking better management options to improve
the survival of infants afflicted with this terrible
disease.
67.3 Etiopathogenesis
and Pathophysiology
Epidemiological studies have identified multiple
perinatal factors that increase an infant’s risk for
the development of NEC, although prematurity
and a history of hypoxia, asphyxia and the intro-
duction of formula feeding are characteristic of
infants that develop NEC. Three main factors are
required for NEC development, including an
immature intestinal barrier immaturity, bacterial
colonization and enteral feeding. Figure  67.1
summarizes the pathogenic sequences and fac-
tors contributing to the development of NEC.
67.3.1 Immature Intestinal Barrier
The intestinal epithelium is a primary barrier
between the inside of the body and the external
environment. As such, the intrinsic function of
this epithelial layer is to protect the host. The
mucosal defence system can be divided into two
categories: non-immunologic and immunologic
defence mechanisms. Non-immunologic mecha-
nisms include peristalsis, gastric acidity, proteo-
lytic enzyme activity, mucin production and
semipermeable intestinal barrier function pro-
vided by tight junctions between the epithelial
cells lining the gut. Peristalsis is the progressive
waves of contraction and relaxation of the intes-
tine. In full-term infants and adults, migrating
motor complexes propagate these waves along
the intestine. In humans, migrating complexes
are not present until approximately 34 weeks of
gestation, which may contribute to intestinal sta-
sis in premature infants, thereby altering the
microbial ecosystem. Gastric acidity is thought
to be a first-line defence against bacterial passage
into the proximal intestine. The premature human
infant’s gastric pH is initially high and then
67  Necrotizing Enterocolitis
Fig. 67.1  Contributing factors in the development of NEC
decreases towards mature levels with increasing
age and ultimately reaching a pH < 4. Permeability
is an important factor in the ability of bacteria to
translocate, causing systemic infection, and the
premature infant intestine is more permeable in
the first 2  days of life. Mucus production pro-
vides a protective viscoelastic layer to the
epithelial intestinal lining. In humans, mucus
­ responsible for enhancing B-cell differentiation
production and composition changes with age
and increases in response to bacterial challenge.
Immunologic defence mechanisms in the gut
include both the innate and adaptive immune sys-
tems. Cells of the innate immune system include
paneth cells, macrophages, polymorphonuclear
leukocytes (PMN), dendritic cells, M cells (spe-
cialized epithelial cells overlying lymphoid
aggregates called Peyer’s patches) and epithelial
cells. Paneth cells produce a variety of antibacte-
rial substances including defensins, lysozymes,
secretory phospholipase A2 and lectins. While
PMN are not regular inhabitants of the healthy
intestine, PMN increase in number in response to
intestinal injury, and their production and func-
tion are known to be impaired in the newborn,
contributing to the inadequate immature intesti-
nal response. Dendritic cells and M cells function
to capture and present antigens from the lumen of
the bowel to T and B cells dispersed in the muco-
909
sal layer and in lymphoid aggregates along the
length of the gut. Cells of the adaptive immune
system include T and B lymphocytes and their
differentiated effector cell subsets. T lympho-
cytes differentiate as cytotoxic T cells and are
responsible for the direct killing of damaged or
infected epithelial cells, and helper T cells are
and the production of plasma cells that secrete
immunoglobulin in particular IgA.  When com-
pared to the adult, both the number and function
of B and T cells are reduced in the intestine, con-
tributing to defects in the intestinal epithelial
barrier.
67.3.2 The Role of Bacterial
Pathogens and Other
Microbes in NEC
NEC occurs in both sporadic and clustered distri-
butions, causing speculation that specific infec-
tious entities may be responsible. A variety of
bacterial and viral pathogens have been impli-
cated in the pathogenesis of NEC, including
Enterobacteriaceae, Clostridia, coagulase nega-
tive Staphylococcus and several viral species
(Table 67.1).
910 H. L. Liebe et al.

Table 67.1 Organisms implicated in necrotizing
enterocolitis
Bacteria
• Enterobacteriaceae species
    – Escherichia species
    – Salmonella species
    – Klebsiella species
    – Enterobacter species
• Clostridium species
• Staphylococcus species
Viruses
• Rotavirus
• Echovirus
• Coronavirus
• Torovirus
Fungus
• +/− Candida species
Despite the fact that no single organism
appears to be responsible for NEC, the impor-
tance of bacteria in the pathogenesis of disease
should not be discounted. It is known that NEC
does not develop in animal models kept in sterile
environments. At birth, the intestine is devoid of
bacterial flora but is rapidly colonized by bacteria
from the rectovaginal flora of the mother and bac-
teria from the surrounding environment.
Additionally, colonization by commensal bacte-
ria is required for the normal development and
maturation of the newborn intestine. Bacterial-­
host crosstalk has been shown to modulate gut
vascular development and promote immune sys-
tem development. Abnormal intestinal coloniza-
tion patterns of neonates admitted to the neonatal
intensive care unit may further increase suscepti-
bility to NEC.
67.3.3 The Role of Enteral Feeding
Rapid advancement of formula feeding has been
associated with the development of
NEC. Breastfeeding is known to protect against
NEC, and premature infants have a reduced inci-
dence of infection when fed human breast milk
rather than formula. A prospective multicentre
study of preterm infants found an almost tenfold
increase in the incidence of NEC in formula-fed
infants as compared with those who were fed
breast milk. The positive effects of breast milk
are likely due to a variety of potential antimicro-
bial products including immunoglobulins, oligo-
saccharides, lactoferrin and glycoproteins with
anti-adhesive capacity for bacteria and cytokines
present in breast milk. Additionally, breast milk
promotes intestinal colonization by probiotic
(beneficial) bacteria such as Lactobacillus spe-
cies and Bifidobacteria species. The benefit of
donor breast milk in the prevention of NEC is
controversial with one randomized control trial
showing a significant decrease in the develop-
ment of stage II or greater NEC, while several
other trials have failed to show a benefit. However,
studies are consistent in showing that infants
exclusively fed human milk have a decreased risk
of developing NEC compared to formula-fed
infants. Dosage also appears to be important with
the risk of NEC increasing by up to 21% for
every 10% increase in non-human milk in an
infant’s diet. A standard feeding schedule has
also been shown to decrease the incidence of
NEC with early trophic feeds and gradual pro-
gression of feeding being protective against
NEC.
67.3.4 Inflammatory Mediators
and NEC
A variety of inflammatory chemokines including
tumour necrosis factor alpha (TNF-α), nitric
oxide (NO), platelet-activating factor (PAF) and
several cytokines (IL-1, IL-6, IL-8 and IL-10)
have been implicated in the pathogenesis of
NEC. Elevated plasma levels of TNF-α have been
found in infants with NEC, and the local upregu-
lation of IL-1, IL-8 and inducible NO synthase
(iNOS) has been demonstrated in the intestine of
infants with NEC. Nitric oxide plays a paradoxi-
cal role in the pathogenesis of NEC. Constitutive
low-level production of NO enhances mucosal
blood flow and promotes local vascular health.
However, a sustained overproduction of NO due
to the upregulation of iNOS leads to intestinal
damage. The pathologic effects of NO are postu-
lated to be due to the strong oxidant effects per-
oxynitrate (ONOO−), resulting in enterocyte cell
67  Necrotizing Enterocolitis
death and impaired mucosal healing. ONOO−
causes enterocyte cell death and impairs mucosal
healing. The upregulation of NO has been dem-
onstrated in areas of intestinal epithelial injury in
human infants with NEC. In infants who recover
from NEC, iNOS and NO return to baseline
levels.
67.3.5 Red Blood Cell Transfusion
and Anaemia
Red blood cell (RBC) transfusions were first
described as a risk factor for the development of
NEC by McGrady et  al. in 1987 who demon-
strated 33 cases of NEC in their NICU that devel-
oped within 2–48  h after receiving a blood
transfusion. Subsequent studies confirmed that
there appeared to be a relationship between RBC
transfusion and NEC. More recently, studies have
come out that suggest anaemia, and not RBC
transfusion, may be associated with an increased
risk of NEC.  Severe anaemia was shown to be
associated with a sixfold increase in the risk of
developing NEC, while RBC transfusion did not
incur any increased risk. Furthermore, biomark-
ers for NEC have been shown to be elevated prior
to RBC transfusion in infants with severe anae-
mia who later went on to develop NEC. This has
led to an emphasis on the prevention of severe
anaemia rather than avoidance of RBC transfu-
sion to prevent NEC. Therapies, such as delayed
cord clamping in premature infants and minimiz-
ing blood draws, are used to decrease the risk of
anaemia and potentially NEC.
67.3.5.1 Maternal Factors
Several maternal risk factors increase an infant’s
risk of developing NEC, including placental
insufficiency and maternal cocaine abuse. These
factors may contribute to disease susceptibility
by compromising placental blood flow and hence
perfusion of vital organs in the foetus.
911
67.4 Pathology
Several pathological findings are associated with
NEC. It is important to note, however, that find-
ings are highly variable. The terminal ileum is the
most frequently involved site; however, NEC
may affect any part of the small or large intestine.
Pan-necrosis is a highly morbid finding associ-
ated with fulminant disease (Fig.  67.2).
Pneumatosis intestinalis is a pathognomonic
finding in NEC characterized by the presence of
air in the intestinal wall (Fig. 67.3). The aetiology
of pneumatosis intestinalis is thought to be due to
bacterial invasion, fermentation and hydrogen
production in the intestinal wall.
Histologically, intestinal tissue may demon-
strate epithelial layer sloughing, increased apop-
totic bodies within epithelial cells, tissue oedema
and submucosal air (Fig. 67.4). Evidence of the
response of the immune system is characterized
by neutrophil infiltration and hypertrophy of
intestinal lymphoid aggregates. However, ulti-
mately the diagnosis of NEC is based on clinical
findings.
Fig. 67.2  Intraoperative image of an infant with intesti-
nal necrosis of the ileum, diagnosed at the time of lapa-
rotomy for NEC
912
Fig. 67.3  Supine abdominal X-ray in a neonate showing
diffuse pneumatosis intestinalis. Arrows indicate a signet
sign
Fig. 67.4  Haematoxylin-eosin staining of the intestinal
tissue in an infant with NEC. Arrows indicate epithelial
destruction with submucosal gas collections and enlarged
lymphoid aggregates
67.5 Diagnosis
The assessment of an infant suspected of having
NEC starts with a thorough history and physical
examination. A typical presentation of NEC
would be that of a premature infant advancing on
enteral feedings who develops signs of feeding
H. L. Liebe et al.
intolerance. These signs may be subtle and first
noted by the nursing staff with signs and symp-
toms of sepsis, abdominal distension, high gas-
tric residuals and guaiac-positive stool. In more
severe cases of disease, infants may present with
a more fulminate course with abdominal disten-
sion, frankly bloody stool and cardiorespiratory
failure.
Although prompt diagnosis and intervention
are desirable tenants of disease management,
there is no clear evidence proving that early diag-
nosis and interventions alter patient outcome.
67.5.1 Clinical Features
In 1973, Bell et  al. proposed the NEC grading
system, in an attempt to standardize diagnosis,
disease severity scoring and management. A
slight modification of Bell’s original criteria is in
use today (Table 67.2). The presenting symptoms
may vary and include feeding intolerance,
abdominal distension, bloody stools, apnoea,
lethargy, temperature instability and hypoperfu-
sion. Classically, increased amounts of gastric
residual and abdominal distension and visible
loops of bowel are noted on physical exam.
Occasionally, the discoloration of the abdominal
wall may be present, and palpation of the abdo-
men may elicit tenderness with guarding. The
progression of disease is variable; some infants
have minor symptoms that resolve spontaneously
over a period of days in response to bowel rest,
while others progress over hours with respiratory
failure requiring intubation, hypotension requir-
ing fluid resuscitation and inotropic support and/
or immediate surgical intervention.
67.5.2 Laboratory Findings
Common laboratory abnormalities include
thrombocytopenia, leukocytosis, electrolytes
imbalance, metabolic acidosis, hypoxia or hyper-
capnia. Therefore, a comprehensive laboratory
analysis should be performed in infants with sus-
pected NEC.  A patient with NEC may present
with an abnormal white blood cell count (WBC).
67  Necrotizing Enterocolitis
Table 67.2 Modified Bell’s stages of necrotizing
enterocolitis
I. Suspected disease
IA.
Mild systemic signs (apnoea, bradycardia, temperature
instability)
Mild intestinal signs (abdominal distention, gastric
residuals, occult blood in stool)
IB.
Mild systemic signs (apnoea, bradycardia, temperature
instability)
Mild intestinal signs (abdominal distention, gastric
residuals, gross blood in stool)
Non-specific or normal radiological findings
II. Definite disease
IIA.
Mild systemic signs (apnoea, bradycardia, temperature
instability)
Additional intestinal signs (absent bowel sounds,
abdominal tenderness)
Specific radiologic signs (pneumatosis intestinalis or
portal venous air)
Laboratory changes (metabolic acidosis,
thrombocytopenia)
IIB.
Moderate systemic signs (apnoea, bradycardia,
temperature instability, mild metabolic acidosis, mild
thrombocytopenia)
Additional intestinal signs (absent bowel sounds,
abdominal tenderness, abdominal mass)
III. Advanced disease
IIIA.
Severe systemic illness (the same as IIB with additional
hypotension and shock)
Intestinal signs (large abdominal distention, abdominal
wall discoloration, peritonitis, intestine intact)
Severe radiologic signs (definite ascites)
Progressive laboratory derangements (metabolic
acidosis, disseminated intravascular coagulopathy)
IIIB.
Severe systemic illness (same as IIIA)
Intestinal signs (large abdominal distention, abdominal
wall discoloration, peritonitis, intestinal perforation)
Severe radiologic signs (definite ascites and
pneumoperitoneum)
Worsening laboratory derangements (metabolic
acidosis, disseminated intravascular coagulopathy)
Although it may be elevated, it is more com-
monly depressed, and a severely low white blood
cell count (<1.5  ×  109/L) has been reported in
37% of cases. The granulocytopenia results from
decreased production and increased utilization of
leukocytes. Thrombocytopenia is also commonly
913
seen in almost 90% of affected individuals. Serial
C-reactive protein levels have been shown to be
successful in differentiating benign abdominal
aetiologies such as ileus from NEC. Furthermore,
it has been suggested that persistently elevated
CRP levels indicate the need for surgical inter-
vention, although this is controversial as CRP is a
non-specific finding. NEC is associated with
­bacteraemia in approximately one third of cases,
and thus blood cultures should be obtained prior
to administration of antibiotics.
67.5.3 Radiological Diagnosis
67.5.3.1 Abdominal Series X-Rays
Serial plain films of the abdomen (anteroposte-
rior radiograph and a left lateral decubitus or
cross-table lateral film) should be obtained at the
first suspicion of disease. Several radiologic find-
ings pathognomonic of NEC can be identified on
abdominal X-ray including intramural air (pneu-
matosis intestinalis), portal vein gas, the ‘fixed
loop sign’, ‘signet sign’ and, in advanced cases,
pneumoperitoneum.
Portal venous air is noted in approximately
30% of advanced cases and occurs when intra-
mural air is absorbed into the mesenteric venous
circulation (Fig. 67.5). Portal venous air portends
a worse prognosis. The finding of a fixed loop is
referred to as a ‘signet sign’ and results from a
bowel loop that remains unchanged for 24–48 h
and is associated with transmural necrosis
(Fig.  67.3). Despite some reports indicating an
association between fixed loops and pan-­necrosis,
almost half of patients recover without surgical
intervention. Free air can be seen in severe cases
of NEC (Fig. 67.6). When a large pneumoperito-
neum can be seen as a central collection of air on
the anteroposterior film, it is referred to as the
‘football sign’ and is an indication for surgical
intervention. Pneumoperitoneum in the absence
of pneumatosis intestinalis may be more sugges-
tive of focal intestinal perforation (FIP), rather
than NEC.
Several common non-specific radiographic
findings include a gasless abdomen, non-specific
bowel gas patterns and ascites. Ascites has been
914
Fig. 67.5  Supine abdominal X-ray revealing pneumato-
sis intestinalis (white arrows) and portal venous gas (black
arrow)
Fig. 67.6  Lateral decubitus abdominal X-ray of an infant
with NEC, demonstrating intestinal free air
reported to occur in approximately 10% of infants
with NEC.
Computed tomography scans (CT) will reveal
findings similar to those seen by X-ray and is
generally unnecessary and therefore, not
indicated.
67.5.3.2 Contrast Studies
The use of contrast enemas is contraindicated in
the diagnosis of NEC. However, contrast studies
may be indicated following recovery from acute
H. L. Liebe et al.
disease. In patients with a history of NEC and
new intestinal obstruction, contrast studies may
localize a stricture secondary to scarring.
67.5.3.3 Ultrasound
Ultrasound can be useful in the diagnosis of
NEC, allowing evaluation of intestinal wall
oedema, portal gas and free fluid. However,
ultrasound remains a notoriously operator-
dependent modality, and expertise in neonatal
ultrasound may not be readily available in many
centres.
67.6 Differential Diagnosis
The variability in the presentation and severity of
NEC may mimic different conditions. Both sep-
sis and ileus can present with signs and symp-
toms of systemic infection and abdominal
distension. Radiologic imaging may help in the
diagnosis, but often does not definitively rule in
NEC unless pneumatosis intestinalis is present.
Fortunately, the therapy, to be discussed later, is
the same for both early NEC and sepsis. Advanced
cases of Hirschsprung’s enterocolitis or severe
gastroenteritis may present with intramural air. In
advanced cases other aetiologies of intra-­
abdominal catastrophe including volvulus, intus-
susception, inspissated meconium syndrome and
intestinal vascular accident may be mistaken for
fulminant NEC.
Focal or spontaneous intestinal perforation is
important to consider in the differential of
NEC. FIP is a distinct entity often characterized
by the lack of signs and symptoms of sepsis.
Despite the fact that FIP is typically seen in
infants of smaller birth weight and more extreme
prematurity, it has a higher survival rate. FIP is
also associated with the use of pharmacologic
agents including indomethacin and dexametha-
sone. Furthermore, the most commonly associ-
ated organism of sepsis in FIP is Candida species,
which is distinct from the agents associated with
sepsis in NEC. It is important to note that infants
with FIP who are culture-positive for Candida
may be more severely ill than those infants with
FIP who are culture-negative.
67  Necrotizing Enterocolitis
67.7 Management
67.7.1 Medical Management
The majority of patients with NEC are treated
non-operatively. Bell’s criteria can be used to
guide the initial management based on the sever-
ity of the disease (Fig. 67.7). Prompt resuscitative
measures including evaluation of airway, breath-
ing and circulation remain paramount in these
patients. Many patients may require ventilator
support, and tracheal intubation is preferred to
prevent aerophagia and greater intestinal disten-
sion. Most patients with NEC will be hypovole-
mic and will require adequate fluid resuscitation,
with intravenous crystalloids. If the patient is
affected by coagulopathy, then correction with
platelets and/or fresh frozen plasma is indicated.
States of acidosis should be corrected by optimal
ventilation, fluid resuscitation and, if necessary,
bicarbonate or tris[hydroxymethyl]-aminometh-
Clinical Concern for NEC
Bell’s Stage I
• ABC,s
• NPO
• Place OG tube
• Check cultures
o (prior to initiation of antibiotics)
• Serial Abdominal X-rays
• Laboratories: CBC, CRP
Improvement of Worsening of
symptoms symptoms
Medical therapy
Course determined by clinical judgment
Fig. 67.7  Algorithms of the therapeutic interventions in NEC
915
ane (THAM) administration. The use of inotropic
support may be indicated in patients with refrac-
tory hypotension despite fluid administration.
After blood cultures have been obtained, broad-­
spectrum antibiotics should be started. Antibiotics
should be tailored after blood cultures have
resulted. Currently, there is insufficient evidence
regarding the choice of a specific antibiotic regi-
men or duration, although broad-spectrum antibi-
otics based on patterns of microbial resistance
within individual neonatal intensive care units
are recommended. The use of antifungal agents
should also be considered in severely ill patients.
Appropriate antibiotic therapy has been shown to
improve the outcome and survival in infants with
NEC.
All enteric feeds should be stopped and the
stomach decompressed with an orogastric tube.
Serial measurements of abdominal girth, frequent
abdominal exams and serial abdominal radiographs
are obtained to monitor disease progression.
Bell’s Stage II and III
• ABCs
• Ventilatory support as needed
• Cardiovascular support as
needed (fluid and pressors)
• NPO
• Place OG tube
• Check cultures
o (prior to initiation of antibiotics)
• Serial Abdominal X-rays
• Laboratories: CBC, CRP
• Obtain surgical consult
Radiologic and clinical
Improvement with signs of advanced NEC
medical support
Surgery
916 H. L. Liebe et al.

67.7.2 Probiotics segments with intervening areas of questionable

viability or the entire extent of the small and large
At birth, the intestine lacks bacterial flora and, bowel (Fig.  67.2). The standard of care is to
with time, becomes colonized with a variety of remove all necrotic areas, taking great care to
bacterial species. As previously mentioned, the preserve any bowel that appears viable. After
microbial flora in the premature hospitalized resecting the necrotic regions, an area of viable
patient is different from that of a healthy full-­ intestine is used to create an ostomy with or with-
term infant. Typically, full-term infants have a out a mucus fistula. There are occasional reports
predominance of favourable gram-positive organ- of resection of isolated perforation with primary
isms including Lactobacilli and Bifidobacteria, reanastomosis of the remaining intestine; how-
whereas premature infants are largely colonized ever, this is not a widely accepted approach. In
by Enterococcus and gram-negative organisms. patients who are found to have multiple areas of
The administration of exogenous probiotics may questionable intestinal viability, a ‘second-look’
protect against NEC by altering the intestinal procedure may be planned to revaluate the bowel
ecosystem from potentially harmful microflora to in 24–36 h prior to the formation of an enteros-
a more favourable or beneficial environment. tomy. In the past, if at the time of surgery many
Clinical trials investigating the potential benefit segments of intestine were found to be involved,
of probiotics have found an overall decrease in the creation of multiple stomas was typically
the risk of NEC in infants given probiotics. A practiced. However, this strategy may result in
meta-analysis of 34 randomized control trials the sacrifice of potentially viable intestine; hence,
showed that the risk of developing NEC decreased some surgeons advocate the ‘patch, drain and
by 42% in infants who were given probiotics wait’ approach. In this case, each perforation is
when compared with placebo. A randomized pro- sutured closed, Penrose drains are placed, and
spective study by Bin-num et al. found an abso- parenteral support is continued. In cases of pan-­
lute risk reduction of 12% in the incidence of necrosis, surgical therapy is often associated with
NEC and a decline in disease severity after daily extremely poor outcomes, and intervention may
feeding supplementation with probiotics. In a be foregone.
prospective study by Lin et al., the overall patient In patients weighing less than 1000  g, lapa-
mortality was lower after probiotic administra- rotomy with intestinal resection is associated
tion, with a lower incidence of NEC following with a very high morbidity and mortality and
probiotic therapy (1.1% vs. 5.3%). Although poor neurodevelopmental outcomes. Ein et al. in
some studies show no benefit, the majority of 1977 reported a series of extremely low birth
studies are in support of probiotic use in the pre- weight patients treated with a peritoneal-drainage
vention of NEC. However, the type of probiotic, procedure. They reported that three survived, two
dosage and duration of therapy remain unclear of which did not require a subsequent operation.
and are currently areas of investigation. It was postulated that peritoneal drainage allowed
a release of intra-abdominal pressure, drainage of
infection and time for further medical optimiza-
67.7.3 Surgical Management tion and stabilization. Despite remaining some-
what controversial, this approach was adopted
Despite appropriate and timely medical manage- for years by many surgeons as both a temporizing
ment, over a third of patients with NEC require measure and a definitive therapeutic approach to
surgical intervention. In infants with a birth Bell’s stage III disease in extremely low birth
weight greater than 1500  g, laparotomy with weight patients. Moss et  al. published a multi-­
resection of affected intestine is the preferred institutional, randomized, controlled trial in
approach. In some cases, the intestine is necrotic 2006, comparing primary peritoneal drainage
in only a single segment, while in others, the dis- with laparotomy and bowel resection in infants
ease pattern is more complex involving multiple less than 1500 g. They found no statistical differ-
67  Necrotizing Enterocolitis
ence in survival, dependence on parenteral nutri-
tion or length of hospital stay. However, recent
data suggests that infants less than 1000  g may
benefit from initial peritoneal drain placement as
opposed to laparotomy. In general, infants in this
weight category who undergo initial drain place-
ment are less likely to develop short bowel syn-
drome and may have a survival benefit even when
they go on to require a subsequent laparotomy.
Hence peritoneal drainage as the initial operative
intervention has an important role in the surgical
management of extremely low birth weight
infants; however, careful individual assessment
of each patient and consultation between medical
and surgical personnel are required for determin-
ing an optimal treatment strategy.
In patients who improve after laparotomy with
ostomy formation, re-establishing intestinal con-
tinuity within 6 weeks, assuming adequate bowel
length, remains an optimal course of action.
67.8 Complications
67.8.1 Strictures
The most serious complications of acute NEC
include intestinal necrosis and perforation
which may occur in up to one third of patients.
However, some patients who initially appear to
respond well to medical management develop
gastric residuals associated with abdominal dis-
tension and bilious emesis upon resuming
enteral feedings. This scenario is suggestive of
the presence of intestinal strictures, which form
in areas of partial intestinal necrosis or ischemia
with healing, causing tissue contraction and
scarring. The most commonly affected site is at
the junction of the descending and sigmoid
colon. Radiographic imaging may confirm
bowel obstruction, with a transition zone and
air-fluid levels. If a stricture is suspected, a con-
trast enema (or an upper gastrointestinal study)
should be performed to assess intestinal patency.
If a stricture is demonstrated, surgical resection
is indicated at this time. Intestinal strictures
develop in approximately one third of patients
with a history of NEC.
917
67.8.2 Short Bowel Syndrome
In addition to intestinal strictures, nearly 25% of
patients with a history of NEC and surgical inter-
vention develop short bowel syndrome. The
intestine of a full-term neonate is approximately
250  cm; when a large segment or multiple seg-
ments of bowel is (are) resected, the amount of
the remaining intestine may not be sufficient to
adequately absorb nutrients and fluids. The
patient is then reliant upon intravenous nutrition
and is exposed to the risks of long-term paren-
teral support including line sepsis, cholestasis,
cirrhosis and liver failure. Some success in survi-
vors with short bowel syndrome has been reported
after small bowel transplant.
67.8.3 Neurodevelopmental Delay
Neurodevelopmental delay continues to be a sig-
nificant issue in survivors of NEC, as many as
50% of infants are affected. Blakely et al. found
that patients with NEC who underwent laparot-
omy instead of peritoneal drainage had a lower
risk of future developmental delay. Prior investi-
gations suggested that neurodevelopmental delay
was more related to prematurity and prolonged
hospital stay, rather that NEC itself. However,
Blakely et al. controlled for the confounding fac-
tors of prematurity and severity of disease in their
study. The Necrotizing Enterocolitis Surgery
Trial (NEST) is an ongoing randomized control
trial looking specifically at death and neurodevel-
opmental outcomes in laparotomy vs. peritoneal
drainage in infants less than 1000 g. This study is
ongoing and will better elucidate the impact of
surgery on neurodevelopmental delay.
67.9 Conclusion
NEC continues to be a formidable disease, mainly
affecting premature neonatal intestines. Despite
its severity, the majority of cases are able to be
successfully medically managed with cessation
of enteral feeds, early antibiotic administration
and avoidance of anaemia. However, 30–50% of
918
neonates progress to requiring surgical treatment
with pneumoperitoneum and peritonitis as abso-
lute indications for surgery. Although the surgical
treatment of NEC is controversial, peritoneal
drain placement is generally recommended in
very low birthweight neonates. In all other
infants, exploratory laparotomy with bowel
resection and ostomy is the recommended
approach. Long-term complications are common
and include short bowel syndrome, intestinal
strictures and neurodevelopmental delay.
Promising advancements have been made in the
field of NEC, although more research is needed
to better diagnose and treat this challenging
disease.
Further Reading
Bi LW et al (2019) Which is the best probiotic treatment
strategy to prevent the necrotizing enterocolitis in
premature infants: a network meta-analysis reveal-
H. L. Liebe et al.
ing the efficacy and safety. Medicine (Baltimore)
98(41):e17521
Cacho NT et  al (2017) Necrotizing enterocolitis and
human milk feeding: a systematic review. Clin
Perinatol 44(1):49–67
Caplan MS, Fanaroff A (2017) Necrotizing: a historical
perspective. Semin Perinatol 41(1):2–6
Hackam DJ et  al (2019) New insights into necrotizing
enterocolitis: from laboratory observation to per-
sonalized prevention and treatment. J Pediatr Surg
54(3):398–404
Moss RL et al (2006) Laparotomy versus peritoneal drain-
age for necrotizing enterocolitis and perforation. N
Engl J Med 354(21):2225–2234
O’Connor DL et al (2016) Effect of supplemental donor
human milk compared with preterm formula on
neurodevelopment of very low-birth-weight infants
at 18  months: a randomized clinical trial. JAMA
316(18):1897–1905
Patel RM et al (2016) Association of red blood cell trans-
fusion, anemia, and necrotizing enterocolitis in very
low-birth-weight infants. JAMA 315(9):889–897
Yanowitz TD et al (2019) Does the initial surgery for nec-
rotizing enterocolitis matter? Comparative outcomes
for laparotomy vs. peritoneal drain as initial surgery
for necrotizing enterocolitis in infants <1000  g birth
weight. J Pediatr Surg 54(4):712–717
 Constipation
Hannah M. E. Evans-Barns, Sebastian K. King,
Bridget R. Southwell, and John M. Hutson
68.1 Introduction, Definition
and Incidence
Constipation is a common childhood condition
that occurs in up to 30% of children (van den
Berg et al. 2006). It accounts for 3–5% of visits to
paediatricians and 10–15% of referrals to gastro-
enterologists. A positive family history may be
found in 25–50% of constipated children, and a
higher incidence has been reported in monozy-
gotic than dizygotic twins. The peak incidence of
constipation occurs at the time of toilet training
(between 2 and 4 years of age), with an increased
prevalence in boys.
Constipation may be defined as being either
organic or functional. Childhood functional con-
stipation is further defined by the Rome IV crite-
In P.  Puri and M.  Höllwarth (eds.), Pediatric Surgery:
Diagnosis and Management, https://doi.org/10.1007/978-3-
540-­69560-8_45, © Spinger-Verlag Berlin Heidelberg 2009
H. M. E. Evans-Barns · S. K. King
Department of Paediatric Surgery, The Royal
Children’s Hospital and Surgical Research Group,
Murdoch Children’s Research Institute, Melbourne,
VIC, Australia
e-mail: Hannah.Evans-Barns@rch.org.au
B. R. Southwell
Surgical Research Group, Murdoch Children’s
Research Institute, Melbourne, VIC, Australia
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_68
68
ria (Table 68.1) (Hyams et al. 2016). Constipation
may also be classified as being either acute or
chronic. Chronic constipation, likewise known as
Table 68.1  Diagnostic criteria for functional constipa-
tion (Rome IV) (Hyams et al. 2016)
Must include two or more of the following occurring at
least once per week for a minimum of 1 month with
insufficient criteria for a diagnosis of irritable bowel
syndrome:
    1. Two or fewer defaecations in the toilet per week in
a child of a developmental age of at least 4 years
    2. At least one episode of faecal incontinence per
week
    3. History of retentive posturing or excessive
volitional stool retention
    4. History of painful or hard bowel movements
    5. History of a large faecal mass in the rectum
    6. History of large diameter stools that can obstruct
the toilet
After appropriate evaluation, the symptoms cannot be
fully explained by another medical condition
J. M. Hutson (*)
University Clinic of Paediatric and Adolescent
Surgery, Medical University of Graz,
Graz, Austria
e-mail: john.hutson@rch.org.au
919
920
intractable constipation, describes those children
in whom there is no improvement in their symp-
toms after 3 months of optimal medical therapy
and/or behavioural modification (Tabbers et  al.
2014).
68.2 Historical Overview
Constipation has been cited as a cause of ill
health for millennia. The first recognised descrip-
tion appears in the Papyrus Ebers, one of the old-
est known books (circa 1550  BC) (Whorton
2000a). An ancient Egyptian medical text, this
papyrus includes the concept that disease resulted
from putrefaction of waste in the colon; sub-
stances released during this process were thought
to poison the body (Bryan and Joachim 1930).
This concept endured, persisting into the nine-
teenth century as intestinal autointoxication, with
the idea that ‘[man at] every moment of his life
runs the risk of being overpowered by poisons
generated within his system’ (Whorton 2000a;
Mathias 2018; Bouchard and Oliver 1894). As
the French physician Charles Bouchard stated,
the constipated person ‘is always working toward
his own destruction […] by intoxication’
(Bouchard and Oliver 1894).
Purported cures for constipation have encom-
passed the relatively benign—including physical
activity and machines for abdominal massage—
to a range of more invasive strategies, such as
rectal dilatators, colonic irrigation sets and surgi-
cal intervention (including colectomy) (Mathias
2018; Whorton 2000b). There have been signifi-
cant advances in contemporary understanding of
the underlying causes and management of consti-
pation; however, the precise aetiology and opti-
mal treatment approach remain the subject of
ongoing investigation (Southwell 2020a; van
Mill et al. 2019; Mugie et al. 2011).
68.3 Aetiology of Constipation
Three distinct periods of development are com-
monly associated with the emergence of func-
tional constipation. The first occurs with a change
H. M. E. Evans-Barns et al.
in diet following the introduction of solid food
during infancy, the second with the initiation of
toilet training and the final period emerging with
the commencement of schooling (Di Lorenzo
2001). Separation from parents and scheduling
changes have been identified as other common
psychosocial triggers (Di Lorenzo 2001). These
events may be associated with a distressing def-
ecation event, prompting the child to avoid evac-
uation of their bowels (‘stool withholding’) so as
not to repeat the experience (Mugie et al. 2011;
Partin et  al. 1992). This further exacerbates the
constipation.
68.4 Acute Constipation
Children with no obvious organic cause for their
constipation should be managed by a combina-
tion of dietary changes, laxatives and/or stool
softeners and/or bulking agents and behavioural
modification and toilet training. Toilet posture
education and pelvic floor muscle training by a
physiotherapist should be considered. 60% of
children presenting with constipation will
respond to treatment within 6–12  months
(Pijpers et  al. 2010; van Ginkel et  al. 2003). A
more detailed exploration of the management
approach is provided in the latter part of this
chapter.
68.5 Chronic Constipation
Despite appropriate medical management, con-
stipation associated with faecal incontinence,
recurrent faecal impaction and functional and
emotional problems persists in around 30% of
patients (van Ginkel et al. 2003). Chronic consti-
pation, not associated with an obvious organic
cause, was once thought to be entirely related to
behavioural or psychological problems; however,
this is no longer the case. Instead, some of these
children are now thought to have an organic
cause for their constipation due to low motility of
the entire colon, known as slow transit constipa-
tion (STC) (Southwell et al. 2005; Hutson et al.
2004).
68 Constipation
Children with STC need to be distinguished
from those with functional faecal retention
(FFR), in whom there is a hold-up in the rectosig-
moid colon associated with normal proximal
colonic transit. Functional faecal retention is the
most common cause of constipation and faecal
incontinence in children and typically develops at
the time of toilet training or when the child starts
school. Children with FFR should respond to
conventional medical and behavioural therapy,
whereas those with STC may require more long-­
term management.
68.6 Pathology of Slow Transit
Constipation
The gastrointestinal tract (GIT) contains its
own nerve cell bodies that form an intrinsic net-
work that is connected to the central nervous
system via the vagal, coeliac and pelvic nerves.
Enteric neurons have cell bodies within ganglia
that lie in the myenteric or submucosal plex-
uses. The cell bodies have processes that pene-
trate the muscle layers where they release their
neurotransmitters. Acetylcholine (Ach) and
tachykinins (including substance P) cause gas-
trointestinal muscular contraction, whilst relax-
ation is initiated by the release of vasoactive
intestinal peptide (VIP), nitric oxide (NO) and
ATP.
Interstitial cells of Cajal (ICC) are found in
the tunica muscularis throughout the GIT and lie
between enteric nerve terminals and smooth
muscle. Although their precise role remained
undetermined for several decades, it is now
thought that they act as a conduit for the active
transmission of electrical slow waves, as well as
serving as gastrointestinal pacemaker cells
(Fig.  68.1). Slow transit constipation has been
associated with a decrease in substance P fibres
and changes in NOS and VIP. A loss of ICC has
also been demonstrated in a range of motility dis-
orders, including STC.  The loss of ICC and/or
substance P (SP) may be a primary anomaly, but
recent evidence suggests they may be secondary
to chronic stretching of the colon with severe,
long-standing constipation.
921
Nerve fibre with
a varicosities
Neurotransmitter
AP
Enzyme
SMC
SMC SMC
Receptors
b
AP
Enzyme
ICC
SMC
SMC SMC
Fig. 68.1  Old and new models of neuromuscular trans-
mission in the gastrointestinal tract. (a) Old two-cell
model. Action potentials travelling along nerve fibres
caused the release of neurotransmitter from varicosities.
Transmitter diffused across extracellular space and bound
to receptors on muscle cells inducing contraction or relax-
ation. (b) New three-cell model. Interstitial cells of Cajal
(ICC) have receptors for transmitters and are connected to
each other and to smooth muscle cells by gap junctions.
ICC form a network among the smooth muscle cells.
Transmitter released from nerve fibres binds to receptors
on ICC, modifying excitation with changes conveyed to
adjacent ICC and muscle cells by electrical conduction.
ICC also act as pacemaker cells generating and conduct-
ing rhythmic electrical activity that produces slow waves
AP, action potential; ICC, interstitial cells of Cajal; SMC,
smooth muscle cell; filled circle, transmitter; filled square,
receptor
68.7 Pathophysiology of Slow
Transit Constipation
Motility is affected in STC, with reduced sponta-
neous propagating contractions and loss of the
expected increase in motility following a meal
challenge and/or laxative provocation (Stanton
et al. 2005; King et al. 2008; Dinning et al. 2015).
922
The underlying spontaneous rhythmic activity
(slow wave frequency) is reduced in situ (Spencer
et al. 2012); however, in vitro studies indicate that
the reduction may be neurally mediated (Dinning
et al. 2016).
Filling the colon with fluid results in colonic
elongation and inhibits motility (Dickson et  al.
2008). Slow transit constipation is associated
with elongation of the transverse colon (redun-
dant colon) (Yik et  al. 2012). Although colonic
elongation may be the result of the underlying
slow colonic transit, animal studies suggest it
may be the cause. Based on animal studies, par-
tial obstruction of the bowel leads to increased
ED2+ macrophages, TNFα mRNA, CD14,
COX2, loss of inhibitory junction potentials and
loss of muscle depolarisation. This is associated
with decreased muscle force, spontaneous rhyth-
mic activity (slow waves), presence of non-­
propagating CMMCs, slowed transit, loss of the
mucosal reflex and decreased response to nerve
stimulation (Chang et al. 2001; Won et al. 2006;
Heredia et  al. 2012; Bao et  al. 2009; Li et  al.
2013).
In addition, the longitudinal elongation of the
colon leads to the activation of an inhibitory
reflex: the longitudinal elongation inhibitory
reflex (or enteric occult reflex). Nitric oxide-­
positive descending interneurons block junction
potentials in muscle (Dickson et al. 2007, 2008).
The inhibitory effect is polarised, leading to
increased action potentials oral to the elongation
and decreased action potentials anal to elonga-
tion and blocks the pelvic nerve activation of the
colon (Dickson et al. 2008).
In humans, the loss of ICC and colonic elon-
gation have been noted. Human STC is associ-
ated with decreased ICC and ckit mRNA and
protein (Tong et al. 2004, 2005). The elongation
of human colon associated with colonic atresia
or megacolon is associated with reduced ICC
(Sutcliffe et  al. 2007). Colonic elongation is
more common in the transverse colon of chil-
dren with STC, whilst the sigmoid colon is not
elongated in constipation where the delay is at
the anorectum (anorectal retention) (Yik et  al.
2012). SP and VIP neurons are reduced in some
patients (Hutson et al. 1996; 1997; 2001; 2004;
H. M. E. Evans-Barns et al.
Uemura et al. 1998; Sutcliffe et al. 2004; King
et al. 2004; 2010; Yik et al. 2011), and TK trans-
mission is affected (Stanton et al. 2003; Tomita
2008).
Recently, the cystic fibrosis transmembrane
conductance regulator (CFTR) was found in cho-
linergic enteric neurons. It was shown to modu-
late enteric cholinergic neurotransmission. There
was downregulation of CFTR and choline acetyl-
transferase (ChAT) in the myenteric ganglia of
the STC colon. This correlated with impaired
contractile responses to electrical field stimula-
tion (Yeh et al. 2019).
Electrical nerve stimulation—applied transcu-
taneously or via electrodes implanted into spinal
nerves S2–S3—can improve motility in some
patients (Dinning et  al. 2007; 2012; Southwell
2020b). It may also increase neuron numbers in
animal models (Wang et  al. 2019). Recently, a
transcutaneous method of monitoring rhythmic
colonic activity has been developed. This should
enable non-invasive analysis of bowel motility,
including responses to treatment (Erickson et al.
2020). In turn, this may facilitate both a greater
understanding of the underlying pathophysiology
and enable the optimisation of electrical stimula-
tion parameters. Monitoring in this fashion has
demonstrated that bowel motility becomes hyper-
active after colonic surgery. The motility patterns
identified were dominated by cyclic motor pat-
terns, rather than becoming quiescent (Vather
et al. 2018). This may be similar to the increased
activity seen in control and STC colon in  vitro
(Dinning et al. 2016).
68.8 Differential Diagnosis
Constipation may be caused by endocrine, meta-
bolic, pharmacological, obstructive, systemic or
central nervous system disorders (Table  68.2).
Hirschsprung disease, caused by aganglionosis
of the bowel, must be excluded in children with
severe idiopathic constipation. This is especially
in those children in whom the constipation has
been present since birth, or if there is associated
delayed passage of meconium (beyond 24  h),
failure to thrive or a positive family history.
68 Constipation 923

Table 68.2  Causes of constipation, including differential Table 68.2 (continued)

diagnoses • Behavioural causes
• Congenital anatomic or structural defects     – Learned pattern of defaecation (may be related to
    – Anorectal malformation, including anal stenosis previous painful defaecation)
    – Anteriorly displaced anus     – Adverse life event
    – Meconium plug syndrome     – Defiant behaviour
    – Hirschsprung disease     – Intellectual disability
    – Pelvic mass
    – Abnormal abdominal musculature—Prune belly,
gastroschisis, Down syndrome
• Metabolic and endocrine disorders
68.9 Diagnosis: A Framework
    – Diabetes insipidus
    – Hypercalcaemia, hypokalaemia 68.9.1 Clinical
    – Renal tubular acidosis
    – Hypothyroidism Medical history and physical examination are
    – Dehydration essential when diagnosing constipation. A thor-
    – Multiple endocrine neoplasia type 2B
ough antenatal, perinatal and postnatal history
• Chronic intestinal pseudo-obstruction
should be obtained (Table 68.3). It is important to
• Cystic fibrosis
• Connective tissue disorders—Scleroderma, SLE, clarify what each individual family defines as
Ehlers-Danlos syndrome ‘constipation’: the occurrence of specific symp-
• Coeliac disease toms and their frequency should be determined.
• Neurologic causes Essential information includes an accurate gas-
    – Damage to the spinal cord—Meningomyelocele, trointestinal and general medical assessment, as
trauma, surgery, tumours, cauda equina syndrome,
tethered cord well as a developmental and psychosocial evalu-
    – Cerebral palsy ation. The delayed first passage of meconium,
    – Infectious polyneuritis frequent soiling, passage of large soft stools,
    – Amyotonia congenita abdominal distension and bloating are all com-
    – Muscular dystrophy mon features associated with STC.
    – Degenerative disorders A thorough physical examination is essential
    – Neurofibromatosis
in the initial assessment of a child with consti-
• Cows’ milk intolerance or other food allergies
• Other organic causes pation (Table 68.4). This should include a gen-
    – Colonic dysmotility (slow transit eral examination, as well as an abdominal
constipation—STC) examination, and external examination of the
    – Outlet obstruction (functional faecal perineum and perianal area. A rectal examina-
retention—FFR) tion should only be performed by an appropri-
• Dietary
ately experienced practitioner, and in specific
    – Poor fibre intake
    – Poor fluid intake circumstances.
    – FODMAP sugar intolerance (constipation-related In most cases, a comprehensive history and
IBS) examination may determine whether or not an
• Medication organic cause is responsible for the constipation.
    – Analgesics (codeine preparations) If the constipation is believed to be non-organic,
    – Antacids
it is rare to be able to obtain sufficient informa-
    – Anticholinergics
    – Anticonvulsants
tion in order to distinguish STC from FFR.
    – Tricyclic antidepressants Blood samples should be obtained for coeliac
    – β-blockers disease screening, thyroid function testing and
    – Iron and calcium supplements allergy testing (cows’ milk protein intolerance
    – Antispasmodics and high percentage of eosinophils in full blood
    – Diuretics count). In children of school age (>4  years),
924 H. M. E. Evans-Barns et al.

Table 68.3  Model of history taking in a child with Table 68.4 Model of examination in a child with
constipation constipation
Demographics General appearance
    Age     Failure to thrive
    Sex Routine observations
Presenting symptoms     Height and weight
    Frequency of defecation     Pulse
    Behaviour associated with defecation     Blood pressure
    Consistency of stools General examination
    Soiling     Including cardiovascular and respiratory examination
    Pain (abdominal, rectal or other) Abdominal examination
    Rectal bleeding—Associated with passage of stool,     Distension
mixed or coating     Hepatosplenomegaly
    Appetite     Abdominal mass—Including faecaloma
    Vomiting     Palpable bowel loops
    Abdominal distension Neurological and spinal examination
    Weight loss/gain     Lower limb—Tone, reflexes and power
    Toilet training     Sacral dimple/sacral hair tuft
    Onset and duration of symptoms     Obvious spinal deformity
    Previous diagnoses and treatments     Muscle (especially buttock) wasting
Current treatment     Joint hypermobility
Perinatal history Anal inspection
    Any antenatal concerns/diagnoses     Site
    Gestation     Visible stool (skin and clothing)
    Birth condition (need for NICU/special care)     Skin condition
    Time of passage of meconium     Perianal skin tags
Developmental history     Anal fissure
    Growth and attainment of developmental markers Rectal examination
Past medical history     Anal wink
    Hospital admissions (medical and surgical)     Anal tone
    Urinary symptoms     Pain
    Hypothyroidism associated symptoms (cold     Presence and consistency of stool
intolerance, coarse hair, dry skin)     Pelvic mass
Dietary history     Explosive stool on finger withdrawal
Medications     Bleeding
    Immunisations
    Allergies
Family history tion or loading is obvious on abdominal and/or
    Gastrointestinal (especially irritable bowel syndrome rectal examination, then a little more information
and coeliac disease) and other significant illnesses may be attained by means of a plain X-ray.
(including thyroid disease, cystic fibrosis, neurological
Abdominal X-rays may be used to assess the
conditions, connective tissue disorders, diabetes)
Psychosocial history presence and degree of abdominal loading, espe-
    Age-appropriate quality of life assessment cially in obese subjects or in those in whom a rec-
tal examination is refused or inappropriate;
however, interpretation may be subjective, and
breath tests are useful to determine if there is any X-ray timing in relation to defaecation may be
sugar malabsorption (Hutson et  al. 2020; misleading.
Waingankar et al. 2018).

68.9.2 Abdominal X-Ray
Plain abdominal X-rays have debatable value in
the assessment of constipation. If faecal impac-
68.9.3 Transit Studies
Colonic transit time takes between 1 and 3 days,
during which time there is extensive mixing of
stool. The quantification of transit time (from the
68 Constipation
stomach to the excretion from the anal canal)
provides an objective evaluation of the faecal
clearance. Transit time was initially measured
using plastic, non-absorbable, radiopaque mark-
ers. Transit time in different regions was deter-
mined by the ingestion of different shaped
markers over 3 to 6 days. Studies measuring nor-
mal transit in children give the upper range of
total colonic transit from 46 to 62 h. Transit rates
in children less than 5 years old are faster, whilst
children older than 6 years have a range of transit
and frequency of defecation similar to adults.
This mode of assessment of gastrointestinal tran-
sit time is widely available and was considered
the gold standard. However, it has been recog-
nised that indigestible solid particles do not move
with a meal and may not be handled by the colon
in the same manner as stool.
Consequently, nuclear transit studies (using
scintigraphy) have emerged as the preferred
method gastrointestinal transit assessment. A
tracer dose of technetium, or gallium, in 20 ml of
milk is ingested, and an initial image is obtained
at 2 h to assess gastric emptying. A further image
is acquired at 6 h to ascertain whether the tracer
has reached the colon. Subsequently, images are
obtained at 24, 30 and 48 h to document transit
through the colon. The colonic transit index may
be obtained based on the geometric mean of
intestinal activity at 6, 24, 30 and 48  h post-­
ingestion of tracer. By this means, patients with
segmental, pan-colonic (STC) or pan-intestinal
transit deficits may be distinguished from those
with normal gastrointestinal transit with FFR and
rapid transit constipation (Fig.  68.2) (Tabbers
et al. 2014; Hutson et al. 2020).
68.9.4 Rectal Biopsy
In cases of intractable constipation with a history
of delayed passage of meconium or symptoms
since birth, a diagnosis of Hirschsprung disease
needs to be eliminated by performing a rectal
biopsy (Tabbers et  al. 2014). Biopsy specimens
are obtained from 2–3 cm above the anal verge
925
and should be deep enough to include adequate
submucosa. A diagnosis of Hirschsprung disease
is supported by an absence of ganglion cells, usu-
ally in the presence of hypertrophied nerve fibres,
with a marked increase in acetylcholinesterase
activity in the lamina propria and muscularis
mucosa (Ambartsumyan et  al. 2020). A rectal
biopsy is also useful in identifying those children
with a food allergy, as recognised by increased
eosinophils in the mucosa.
68.9.5 Laparoscopic Colonic Biopsies
In children with proximal colonic delay demon-
strated by their transit study, laparoscopic sero-
muscular biopsies have been performed in
association with rectal biopsy, in some centres, in
an attempt to identify any consistent histological
anomalies (King et  al. 2005; Imaji et  al. 2000;
Oliver et  al. 2001). Biopsies collected from the
hepatic flexure, mid-transverse colon, splenic
flexure and sigmoid colon are processed for
immunofluorescence histochemistry and stained
for substance P, vasoactive intestinal peptide
(VIP), nitric oxide synthase (NOS) or cKit (a
marker for interstitial cells of Cajal (ICC)). It has
been proposed that some children with STC have
a form of intestinal neuronal dysplasia (IND).
This abnormality of intestinal innervation is sub-
tler than Hirschsprung disease and may be diag-
nosed by abnormal immunohistochemistry.
68.9.6 Colonic Manometry
Colonic manometry involves the in  vivo mea-
surement of changes in intraluminal pressure
within the colon. A multichannel water-perfused
or solid-state pressure recording catheter is sited
in the colon in either a retrograde manner, via
colonoscopy, or an antegrade manner, via a pre-­
existing appendix stoma or a nascolonic route
(Fig. 68.3) (Dinning et al. 2010).
Colonic contractile activity produces changes
in intraluminal pressure. Contractions can be
926 H. M. E. Evans-Barns et al.

a
6 hours 24 hours

30 hours 48 hours

6 hours 24 hours

30 hours 48 hours

Fig. 68.2  Nuclear transit studies demonstrating (a) slow transit constipation (STC) and (b) functional faecal retention
(FFR)
68 Constipation 927

colonic manometry have been defined and param-

eters measured. Expected frequency of propagat-
ing sequences, ratio of antegrade to retrograde
contractions, frequency of high-amplitude propa-
gating sequences, post-prandial response and
diurnal variation have been identified (Rodriguez
et al. 2017).
Although originating from the auspices of
research, manometric data have become
increasingly valuable in clinical assessment of
­
children with refractory constipation (Tabbers
et al. 2014; Rodriguez et al. 2017; Pensabene et al.
2003; Koppen et  al. 2016). Colonic manometry
may be utilised to assess any underlying motility
disorder, direct management and guide surgical
intervention (Dinning et al. 2010; Rodriguez et al.
2017; Pensabene et al. 2003; Dinning 2018; Gupta
et al. 2020). A correlation has been demonstrated
that children with STC on nuclear transit studies
have a lack of antegrade peristaltic sequences on
colonic manometry (King et al. 2008).
The development of high-resolution manom-
etry has markedly improved study resolution (see
Fig. 68.3  Abdominal radiograph showing an antegradely
inserted 8 channel manometry catheter passing from the Figs.  68.4 and 68.5) Comparative studies have
appendix (App) to the rectum (Rec), with the positions of highlighted deficiencies in assessments based on
the side holes shown (King et al. 2008) low-resolution techniques (Dinning et al. 2013).
The application of this technique to children with
propagating or non-propagating, with propagat- STC identified a potential manometric marker of
ing contractions occurring in either an antegrade neuropathy (Giorgio et  al. 2013). Although fur-
or retrograde direction (Fig. 61.4). High-­ ther research is required to support these find-
amplitude contractions are thought to represent ings, HRM in paediatric STC offers a promising
mass movement within the colon. Standards for area for future exploration.
928 H. M. E. Evans-Barns et al.

Fig. 68.4 Comparison of low- and high-resolution sequences is noted in the recording obtained using a high-­
colonic manometric recording. This study was conducted resolution catheter; these were not identified by the low-­
in a woman with slow transit constipation during sacral resolution study (Dinning et al. 2012)
nerve stimulation. The presence of retrograde propagating

a b c

Fig. 68.5  High-resolution manometry in children demonstrating both normally (a) and abnormally (b, c) propagating
contractions (Corsetti et al. 2019)
68 Constipation
68.10 Management
68.10.1 Medical
Most children improve with standard medical
treatment. This should include laxatives/stool
softeners/bulking agents, dietary changes, behav-
ioural modification and toilet training. Adequate
fibre and fluid intake should be highlighted as
part of a normal diet; however, additional fibre
supplementation is not supported by current evi-
dence (Southwell 2020a). Polyethylene glycol
(with or without electrolytes) acts as an osmotic
laxative by increasing the amount of water
retained in the faeces, thus producing stools that
are softer, bulkier and easier to pass. It is espe-
cially useful for disimpaction; however, regular
administration of polyethylene glycol is also con-
sidered the first-line therapy for ongoing man-
agement of chronic constipation (Southwell
2020a). As maintenance treatment may be
required for months to years, adequate long-term
follow-up of these children is essential to ensure
effective management (Southwell 2020a; Mugie
et al. 2011).
Neurotrophin-3, a neurotrophic factor acting
on the central nervous system, has been shown to
increase stool frequency and enhance colonic
transit. Novel therapeutic drug prucalopride is a
selective, 5-HT4 receptor agonist with a proki-
netic effect on the lower gastrointestinal tract
(Winter et al. 2013). Whilst shown to be effective
in treating adults with constipation, the evidence
in children remains limited and conflicting (van
Mill et al. 2019; Winter et al. 2013; Koppen et al.
2018). Other oral laxatives, such as lubiprostone
and linaclotide, have shown promise in adult
populations; however, there is insufficient evi-
dence to support their safety and efficacy for pae-
diatric use (Southwell 2020a).
Despite significant advances in the manage-
ment of paediatric constipation, the optimal treat-
ment approach remains the subject of ongoing
research. Several emerging fields include the role
of the microbiome in constipation, dietary influ-
ences (including food intolerances) and the appli-
929
cation of neuromodulation to this cohort
(Southwell 2020a). Although probiotics are the
subject of intense public and research interest,
evidence is currently insufficient to support their
routine use in the management of childhood con-
stipation (Southwell 2020a; Mugie et  al. 2011).
Diets excluding certain FODMAP (fermentable
oligo-, di-, and monosaccharide and polyol) sug-
ars have been shown to be effective in some chil-
dren with IBS-C (Hutson et al. 2020; Chumpitazi
and Shulman 2016; Hill et al. 2017). Rapid tran-
sit constipation is an increasingly recognised
entity, characterised by rapid transit through the
proximal colon associated with chronic constipa-
tion. Exclusion of sugar intolerances identified
through breath tests in this cohort may improve
constipation and associated symptoms, including
abdominal pain (Hutson et al. 2020).
68.10.2 Surgical
The surgical management of chronic treatment
resistant constipation initially consisted mostly
of bowel resection (with or without stoma forma-
tion). However, a less invasive approach is now
regularly taken, with the formation of a continent
appendix stoma for the administration of ante-
grade continence enemas (ACE). This approach
was first described by Malone for the manage-
ment of incontinence after the correction of an
anorectal malformation (Malone et al. 1990).
In this procedure, the appendix is brought
through the anterior abdominal wall, traditionally
in either the right iliac fossa or at the umbilicus.
The appendix is sutured to the skin to form an
appendicostomy. Antegrade continence enemas
are then performed via the stoma to flush faeces
from the caecum to the rectum. The ACE may be
performed via intermittent stomal catheterisation
or via a Chait caecostomy button.
When the colon is intermittently (every
2–3  days) evacuated in this manner, it remains
relatively empty, improving continence and soil-
ing. There is consensus that this approach is rela-
tively effective in children, leading to
930
improvements in the number of bowel move-
ments and incontinence episodes. Improved qual-
ity of life scores have also been demonstrated
(Mugie et al. 2011; Mohamed et al. 2020).
68.11 Complications
68.11.1 Disease Related
Ineffective treatment may lead to faecal impac-
tion necessitating either medical or surgical dis-
impaction. Constant soiling, if poorly managed,
may result in perianal erythema and, in severe
cases, excoriation.
68.11.2 Laxatives
There is widespread belief that the chronic use of
laxatives may lead to tolerance, habituation and
even colonic damage. These misconceptions
often lead to inappropriate prescribing practices.
When used appropriately, there are relatively few
side effects to either bulk, osmotic or even stimu-
lant laxatives.
68.11.3 Surgical
Intolerable stool leakage from an appendicos-
tomy, stomal stenosis and/or difficult catheterisa-
tion may necessitate revision. Other potential
complications include local site infection and
granulation tissue development.
68.11.4 Quality of Life
One of the biggest and least recognised compli-
cations of chronic constipation is the associated
impact on the quality of life (Clarke et al. 2008;
Youssef et al. 2005). Chronic abdominal pain and
constant faecal soiling often leads to disrupted
peer relationships, undue family stress and social
ostracism. Behavioural problems, which may be
extreme, may be the cause in some patients and,
H. M. E. Evans-Barns et al.
however, are often the result of years of living
with constipation.
68.12 Conclusion
Constipation is a common childhood condition.
Although the majority of patients will respond to
adequate standard treatment, a subgroup of these
children will continue to experience symptoms
into adulthood. This group may present a signifi-
cant management challenge. It is essential that
medical professionals remain sensitive to the
impact of childhood constipation on the patient’s
and family’s quality of life. Recent years have
seen significant progress in the management of
paediatric constipation. Several emerging treat-
ment strategies suggest the potential for a greater
selection of management options becoming
available for these children.
References
Ambartsumyan L, Smith C, Kapur RP (2020) Diagnosis of
Hirschsprung disease. Pediatr Dev Pathol 23(1):8–22
Bao YG et  al (2009) Roles of enteric nervous system
neurotransmitters and interstitial cells of Cajal in the
colon in slow transit constipation in rats. Zhongguo
Dang Dai Er Ke Za Zhi 11(6):481–485
Bouchard C, Oliver TS (1894) Lectures on auto-­
intoxication in disease; or, self-poisoning of the indi-
vidual. F. A. Davis Company, Philadelphia
Bryan CP, Joachim H (1930) The papyrus Ebers. G. Bles,
London
Chang IY et al (2001) Loss of interstitial cells of Cajal and
development of electrical dysfunction in murine small
bowel obstruction. J Physiol 536(Pt 2):555–568
Chumpitazi BP, Shulman RJ (2016) Dietary carbohy-
drates and childhood functional abdominal pain. Ann
Nutr Metab 68(Suppl. 1):8–17
Clarke MCC et al (2008) Quality of life in children with
slow transit constipation. J Pediatr Surg 43(2):320–324
Corsetti M et  al (2019) First translational consensus on
terminology and definitions of colonic motility in
animals and humans studied by manometric and
other techniques. Nat Rev Gastroenterol Hepatol
16(9):559–579
Di Lorenzo C (2001) Pediatric anorectal disorders.
Gastroenterol Clin N Am 30(1):269–287
Dickson EJ et  al (2007) An enteric occult reflex under-
lies accommodation and slow transit in the distal large
bowel. Gastroenterology 132(5):1912–1924
68 Constipation
Dickson EJ et al (2008) Polarized intrinsic neural reflexes
in response to colonic elongation. J Physiol 586(Pt
17):4225–4240
Dinning PG (2018) A new understanding of the physi-
ology and pathophysiology of colonic motility?
Neurogastroenterol Motil 30(11):e13395
Dinning PG et al (2007) Sacral nerve stimulation induces
pan-colonic propagating pressure waves and increases
defecation frequency in patients with slow-transit con-
stipation. Color Dis 9(2):123–132
Dinning PG et  al (2010) Paediatric and adult colonic
manometry: a tool to help unravel the patho-
physiology of constipation. World J Gastroenterol
16(41):5162–5172
Dinning PG et  al (2012) Pancolonic motor response to
subsensory and suprasensory sacral nerve stimulation
in patients with slow-transit constipation. Br J Surg
99(7):1002–1010
Dinning PG et al (2013) Low-resolution colonic manom-
etry leads to a gross misinterpretation of the frequency
and polarity of propagating sequences: initial results
from fiber-optic high-resolution manometry studies.
Neurogastroenterol Motil 25(10):e640–e649
Dinning PG et al (2015) Colonic motor abnormalities in
slow transit constipation defined by high resolution,
fibre-optic manometry. Neurogastroenterol Motil
27(3):379–388
Dinning PG et al (2016) High-resolution colonic motility
recordings in vivo compared with ex vivo recordings
after colectomy, in patients with slow transit constipa-
tion. Neurogastroenterol Motil 28(12):1824–1835
Erickson JC et  al (2020) Electrocolonography: non-­
invasive detection of colonic cyclic motor activity
from multielectrode body surface recordings. IEEE
Trans Biomed Eng 67(6):1628–1637
Giorgio V et al (2013) High-resolution colonic manometry
accurately predicts colonic neuromuscular pathologi-
cal phenotype in pediatric slow transit constipation.
Neurogastroenterol Motil 25(1):70–e9
Gupta A et al (2020) Surgically treated intractable consti-
pation in children evaluated with colonic manometry.
J Pediatr Surg 55(2):265–268
Heredia DJ et  al (2012) Insights from a novel model of
slow-transit constipation generated by partial outlet
obstruction in the murine large intestine. Am J Physiol
Gastrointest Liver Physiol 303(9):G1004–G1016
Hill P, Muir JG, Gibson PR (2017) Controversies and
recent developments of the low-FODMAP diet.
Gastroenterol Hepatols 13(1):36–45
Hutson JM, Chow CW, Borg J (1996) Intractable consti-
pation with a decrease in substance P-immunoreactive
fibres: is it a variant of intestinal neuronal dysplasia? J
Pediatr Surg 31(4):580–583
Hutson JM et  al (1997) Deficiency of substance
P-immunoreactive nerve fibres in children with intrac-
table constipation: a form of intestinal neuronal dys-
plasia. J Paediatr Child Health 33(3):187–189
Hutson JM et al (2001) Slow transit constipation in chil-
dren. J Paediatr Child Health 37(5):426–430
931
Hutson JM et al (2004) Chronic constipation: no longer
stuck! Characterization of colonic dysmotility as a
new disorder in children. J Pediatr Surg 39(6):795–799
Hutson JM et  al (2020) ‘Rapid transit’ constipation in
children: a possible genesis for irritable bowel syn-
drome. Pediatr Surg Int 36(1):11–19
Hyams JS et al (2016) Childhood functional gastrointes-
tinal disorders: child/adolescent. Gastroenterology
150(6):1456–1468
Imaji R et al (2000) Rectal mucosal biopsy compared with
laparoscopic seromuscular biopsy in the diagnosis of
intestinal neuronal dysplasia in children with slow-­
transit constipation. J Pediatr Surg 35(12):1724–1727
King SK, Sutcliffe JR, Hutson JM (2005) Laparoscopic
seromuscular colonic biopsies: a surgeon's experience.
J Pediatr Surg 40(2):381–384
King SK et al (2004) Paediatric constipation for adult sur-
geons—Article 2: New microscopic abnormalities and
therapies. ANZ J Surg 74(10):890–894
King SK et al (2008) 24-hour colonic manometry in pedi-
atric slow transit constipation shows significant reduc-
tions in antegrade propagation. Am J Gastroenterol
103(8):2083–2091
King SK et al (2010) Substance P and vasoactive intestinal
peptide are reduced in right transverse colon in pedi-
atric slow-transit constipation. Neurogastroenterol
Motil 22(8):883–e234
Koppen IJ et  al (2016) Childhood constipation: finally
something is moving! Expert Rev Gastroenterol
Hepatol 10(1):141–155
Koppen IJN et al (2018) Recommendations for pharma-
cological clinical trials in children with functional
constipation: the Rome foundation pediatric subcom-
mittee on clinical trials. Neurogastroenterol Motil
30(4):e13294
Li Z et  al (2013) Alterations of Cajal cells in the colon
of slow transit constipation rats. Zhonghua Wei Chang
Wai Ke Za Zhi 16(8):777–779
Malone PS, Ransley PG, Kiely EM (1990) Preliminary
report: the antegrade continence enema. Lancet
336(8725):1217–1218
Mathias M (2018) Autointoxication and historical precur-
sors of the microbiome-gut-brain axis. Microb Ecol
Health Dis 29(2):1548249–1548249
Mohamed H et al (2020) Tube cecostomy versus appendi-
costomy for antegrade enemas in the management of
fecal incontinence in children: a systematic review. J
Pediatr Surg 55(7):1196–1200
Mugie SM, Di Lorenzo C, Benninga MA (2011)
Constipation in childhood. Nat Rev Gastroenterol
Hepatol 8(9):502–511
Oliver MR et al (2001) Gastrointestinal transit and ano-
rectal manometry in children with colonic substance
P deficiency. J Gastroenterol Hepatol 16(6):624–630
Partin JC et al (1992) Painful defecation and fecal soiling
in children. Pediatrics 89(6 Pt 1):1007–1009
Pensabene L et al (2003) Colonic manometry in children
with defecatory disorders. Role in diagnosis and man-
agement. Am J Gastroenterol 98(5):1052–1057
932
Pijpers MAM et al (2010) Functional constipation in chil-
dren: a systematic review on prognosis and predictive
factors. J Pediatr Gastroenterol Nutr 50(3):256–268
Rodriguez L et  al (2017) An ANMS-NASPGHAN con-
sensus document on anorectal and colonic manometry
in children. Neurogastroenterol Motil 29(1)
Southwell BR (2020a) Treatment of childhood con-
stipation: a synthesis of systematic reviews and
meta-­analyses. Expert Rev Gastroenterol Hepatol
14(3):163–174
Southwell BR (2020b) Electro-neuromodulation for
colonic disorders- review of meta-analyses, sys-
tematic reviews and RCTs. Neuromodulation
23(8):1061–1081
Southwell B, King S, Hutson J (2005) Chronic constipa-
tion in children: organic disorders are a major cause. J
Paediatr Child Health 41(1–2):1–15
Spencer NJ et  al (2012) Characterization of motor pat-
terns in isolated human colon: are there differences in
patients with slow-transit constipation? Am J Physiol
Gastrointest Liver Physiol 302(1):G34–G43
Stanton MP et  al (2003) Cholinergic transmission to
colonic circular muscle of children with slow-­transit
constipation is unimpaired, but transmission via
NK2 receptors is lacking. Neurogastroenterol Motil
15(6):669–678
Stanton MP et al (2005) Colonic manometry via appen-
dicostomy shows reduced frequency, amplitude, and
length of propagating sequences in children with slow-­
transit constipation. J Pediatr Surg 40(7):1138–1145
Sutcliffe JR et al (2004) Paediatric constipation for adult
surgeons—Article 1: Targeting the cause. ANZ J Surg
74(9):777–780
Sutcliffe J et  al (2007) Reduced distribution of pace-
making cells in dilated colon. Pediatr Surg Int
23(12):1179–1182
Tabbers MM et  al (2014) Evaluation and treatment
of functional constipation in infants and children:
evidence-based recommendations from ESPGHAN
and NASPGHAN.  J Pediatr Gastroenterol Nutr
58(2):258–274
Tomita R (2008) Regulation of the peptidergic nerves
(substance P and vasoactive intestinal peptide) in
the colon of women patients with slow transit con-
stipation: an in vitro study. Hepato-Gastroenterology
55(82–83):500–507
Tong WD et al (2004) Decreased interstitial cells of Cajal
in the sigmoid colon of patients with slow transit con-
stipation. Int J Color Dis 19(5):467–473
Tong WD et  al (2005) Expression of c-kit messenger
ribonucleic acid and c-kit protein in sigmoid colon of
patients with slow transit constipation. Int J Color Dis
20(4):363–367
H. M. E. Evans-Barns et al.
Uemura S et  al (1998) Distributions of substance
P- and VIP-immunoreactive nerve fibres in the
colonic circular muscle in children. Pediatr Surg Int
14(1–2):66–70
van den Berg MM, Benninga MA, Di Lorenzo C (2006)
Epidemiology of childhood constipation: a systematic
review. Am J Gastroenterol 101(10):2401–2409
van Ginkel R et  al (2003) Childhood constipation: lon-
gitudinal follow-up beyond puberty. Gastroenterology
125(2):357–363
van Mill MJ, Koppen IJN, Benninga MA (2019)
Controversies in the management of functional con-
stipation in children. Curr Gastroenterol Rep 21(6):23
Vather R et  al (2018) Hyperactive cyclic motor activity
in the distal colon after colonic surgery as defined
by high-resolution colonic manometry. Br J Surg
105(7):907–917
Waingankar K et  al (2018) Dietary exclusion of fruc-
tose and lactose after positive breath tests improved
rapid-transit constipation in children. JGH Open
2(6):262–269
Wang Y et  al (2019) Colonic electrical stimulation pro-
motes colonic motility through regeneration of myen-
teric plexus neurons in slow transit constipation
beagles. Biosci Rep 39(5):BSR20182405
Whorton JC (2000a) Inner hygiene: constipation and the
pursuit of health in modern society. Oxford University
Press, New York
Whorton J (2000b) Civilisation and the colon: con-
stipation as the “disease of diseases”. BMJ
321(7276):1586–1589
Winter HS, Di Lorenzo C, Benninga MA (2013) Oral pru-
calopride in children with functional constipation. J
Pediatr Gastroenterol Nutr 57(2):197–203
Won KJ et al (2006) Motility disorder in experimentally
obstructed intestine: relationship between muscula-
ris inflammation and disruption of the ICC network.
Neurogastroenterol Motil 18(1):53–61
Yeh KM et  al (2019) Cystic fibrosis transmembrane
conductance regulator modulates enteric cholinergic
activities and is abnormally expressed in the enteric
ganglia of patients with slow transit constipation. J
Gastroenterol 54(11):994–1006
Yik YI et  al (2011) Gender differences in reduced sub-
stance P (SP) in children with slow-transit constipa-
tion. Pediatr Surg Int 27(7):699–704
Yik YI et  al (2012) How common is colonic elongation
in children with slow-transit constipation or anorectal
retention? J Pediatr Surg 47(7):1414–1420
Youssef NN et  al (2005) Chronic childhood constipa-
tion is associated with impaired quality of life: a
case-­controlled study. J Pediatr Gastroenterol Nutr
41(1):56–60
 Hirschsprung’s Disease
Prem Puri
69.1 Introduction
Hirschsprung’s disease (HSCR) is characterized
by an absence of ganglion cells in the distal
bowel, extending proximally for varying dis-
tances (Heanue and Pachnis 2007). In over 80%
of patients, aganglionosis is confined to the recto-
sigmoid. In the remaining patients, aganglionosis
extends beyond the rectosigmoid, involving the
descending colon, transverse colon, or the entire
colon with a short segment of terminal ileum.
Total intestinal aganglionosis with an absence of
ganglionic cells from the rectum to the duode-
num is the rarest form of HSCR and is associated
with high morbidity and mortality (Senyuz et al.
1989; Nemeth et al. 2001; Ruttenstock and Puri
2009; Ziegler et al. 1987). The aganglionosis is
continuous and uninterrupted until the proximal
transitional zone is reached. The length of the
transitional zone may vary and extend for several
centimeters and is characterized by hypogangli-
onosis (Coyle et al. 2019).
P. Puri (*)
Department of Pediatric Surgery, Beacon Hospital,
and University College Dublin, Dublin, Ireland
e-mail: prem.puri@ucd.ie
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_69
69
69.2 Historical Overview
The condition of “congenital megacolon” has
been recognized for centuries. The first known
description of this condition was by the ancient
Indian surgeon Sushruta in 600 BC and later by
Fredericus Ruysch, a Dutch anatomist in 1691
(Fiori 1998; Raveenthiran 2011). The most in-­
depth description of congenital megacolon was
reported by Harald Hirschsprung, a pediatrician
from Copenhagen, when he presented two infants
with this condition to the Society of Pediatrics in
Berlin in 1886 (Skaba 2007). His treatise was
entitled “Constipation in newborns due to dilata-
tion and hypertrophy of the colon.” During the
next 60 years, several cases of congenital mega-
colon were reported, and most of these children
died. It was not until Keenohan in 1948 and
Bodian et al. in 1949 confirmed the diagnosis of
Hirschsprung’s disease by recognizing the pres-
ence of the contracted and spastic rectosigmoid
and the absence of ganglion cells in the spastic
distal segment. In 1948, Swenson and Bill pub-
lished their classic paper recommending resec-
tion of the contracted and spastic rectosigmoid
with the preservation of the sphincter as the treat-
ment for Hirschsprung’s disease (Skaba 2007;
Swenson 1999). Although improvements in oper-
ative techniques and earlier diagnosis have
resulted in significantly improved morbidity and
mortality in patients with HSCR, the dogma
933
934
stated by Swenson that patients can only be cured
by the removal of the contracted aganglionic seg-
ment of bowel tissue is still valid (Swenson
1996).
69.3 Incidence
Several studies on the incidence of HSCR have
been reported. The incidence of HSCR is esti-
mated to be 1  in 5000 live births and ranges
from 1 in 2000 to 1 in 12,000 live births (Puri
and Nakamura 2019). A large survey of HSCR
cases from a population database in California
found 2464 cases of HSCR among 9.3 million
births during 1995–2013, with an incidence of
2.2 cases per 10,000 live births (Anderson
et al. 2018). Incidence varied by race, with the
highest rates seen among African Americans at
4.1 per 10,000 live births and Asian/Pacific
children at 2.5 per 10,000 live births compared
to 1.9 cases per 10,000 live births among
White/Caucasian children. The incidence of
HSCR in the UK and Ireland is reported to be
1.8 per 10,000 live births (Bradnock et  al.
2017). The incidence of HSCR is 1.91 per
10,000 live births in Sweden (Lof Granstrom
et  al. 2016). Taguchi et  al. reported that the
incidence of HSCR in Japan during 1978–2012
was 1.96 per 10,000 live births (Taguchi et al.
2017). The incidence of HSCR in Ontario,
Canada, during 1991–2013 was 2.05 per
10,000 live births (Nasr et al. 2017). In a regis-
ter based study in Europe, a total of 1322 cases
of HSCR among 12,146,210 live births were
reported, and the total prevalence was 1  in
9174 live births (Best et  al. 2014). More
recently, a nationwide survey from Taiwan
found an incidence of HSCR of 1 in 4545 live
births between 1998 and 2010 (Chia et  al.
2016). A European register-based study
revealed that there is no evidence of a signifi-
cant increased risk of HSCR in cases born to
women aged ≥35 years compared to those aged
25 to 29 (Best et al. 2014). A population-­based
case control study found an increased risk in
obese females having children with HSCR
(Lof Granstrom et al. 2016).
P. Puri
69.3.1 Gender
It has been recognized that males are more com-
monly affected than females with a male:female
ratio of 4:1 (Best et al. 2014; Taguchi et al. 2017;
Puri and Nakamura 2019). The male preponder-
ance is less evident in long-segment HSCR,
where the male:female ratio is 1:1–2 and is even
reversed in total colonic aganglionosis, where the
male:female ratio is 0.8:1 (Orr and Scobie 1983;
Puri and Nakamura 2019). The reason for these
skewed ratios is unclear; no X-linked loci have
been described in HSCR.
69.3.2 Birth Characteristics
A Californian population-based study compris-
ing of 2464 HSCR patients reported a median
gestational age of HSCR patients to be 39 weeks.
This study found that a higher population of
patients with HSCR were born preterm
(<37 weeks), and a higher population of HSCR
patients were low birth weight (<2500  g) com-
pared to patients without HSCR.  Duess et  al.
reported a prevalence rate of 6% of preterm
infants born with HSCR (Duess et  al. 2014).
Bradnock et al. reported that 12% of infants with
HSCR were born prematurely (Bradnock et  al.
2017).
69.3.3 Race
Epidemiological studies have suggested that
there may be ethnic variations in the prevalence
of HSCR. Recently, one of the largest population-­
based studies of the epidemiology of HSCR from
California from 1995 to 2012 found that the inci-
dence of HSCR varied by race (Anderson et al.
2018). The incidence was highest among African
American (4.1/10,000 births) and Asians
(2.5/10,000 births) compared to White/Caucasian
and Hispanics (1.9/10,000 births). In another
study, Chia et al. reported a total of 629 HSCR
cases with an overall incidence of 2.2 in 10,000
live births in Taiwanese children (Chia et  al.
2016). Recently, (Taghavi et al. 2018) reported a
69  Hirschsprung’s Disease
total of 246 patients with HSCR in New Zealand.
The prevalence of HSCR was 2.6 in 10,000 live
births for European, 1.5 in 10,000 among Maori,
5.5  in 10,000 among Pacific patients, 2.6  in
10,000 among Asian, and 1.8  in 10,000 among
Middle Eastern. In New Zealand, the prevalence
of HSCR was statistically significantly greater in
Pacific people (p < 0.0005) (Taghavi et al. 2018).
The proportion of children with long-segment
HSCR was also significantly greater in Pacific
and Asian populations compared to others
(p = 0.04) (Taghavi et al. 2018).
69.4 Etiopathogenesis
In the human fetus, neural crest-derived neuro-
blasts first appear in the developing esophagus at
5 weeks of gestation, and then migrate down to
the anal canal in a craniocaudal direction during
the 5th to 12th weeks. The neural crest cells first
form the myenteric plexus just outside the circu-
lar muscle layer. The mesenchymal-derived lon-
gitudinal muscle layer then forms, sandwiching
the myenteric plexus after it has been formed in
the 12th week of gestation. In addition, after the
craniocaudal migration has ended, the submu-
cous plexus is formed by the neuroblasts, which
migrate from the myenteric plexus across the cir-
cular muscle layer and into the submucosal
region; this progresses in a craniocaudal direc-
tion during the 12th–16th weeks of gestation
(Burns et al. 2009). The absence of ganglion cells
in HCSR has been attributed to a failure of migra-
tion of neural crest cells. The earlier the arrest of
migration, the longer the aganglionic segment.
Genetic factors have been implicated in the
etiology of HSCR (Moore 2017; Mc Laughlin
and Puri 2015). HSCR is known to occur in fami-
lies. The reported incidence of familial cases var-
ied from 3.6% to 7.8% in different series. A
familial incidence of 15–21% has been reported
in total colonic aganglionosis and 50% in the rare
total intestinal aganglionosis. Recently, (Moore
and Zaahl et al. 2015) reported 45 patients with
HSCR in 35 kindreds. Aganglionosis was signifi-
cantly more frequent with total colonic agangli-
onosis in 40% of familial cases.
935
Several genes and signaling molecules have
been identified which control morphogenesis and
differentiation of the enteric nervous system
(ENS) (Tam 2016; Tam et al. 2019). These genes,
when mutated or deleted, interfere with ENS
development. So far, 22 genes are known to be
involved in the development of HSCR.  One of
these genes, RET, with tyrosine kinase activity is
involved in the development of enteric ganglia
derived from vagal-neural crest cells and is
thought to the major gene causing HSCR
(Tomuschat and Puri 2015). Mutations of the
RET gene account for 50% of familial and 15 to
20% of sporadic cases of HSCR (Tam et al. 2019).
HSCR occurs as an isolated trait in 70% of
patients (Tam et al. 2019). A chromosomal abnor-
mality is associated with HSCR in 12% of
patients, trisomy 21 being by far the most fre-
quent (>90%). The clinical association between
trisomy 21 (Down’s syndrome) and HSCR is
well-established, being of the order of 5%, and
remains the most common congenital association
with HSCR (Moore 2019). Other authors have
reported the occurrence of Down’s syndrome in
4.5–16% of all patients with HSCR (Menezes
and Puri 2005; Moore 2019). Other chromosomal
abnormalities that have been described in asso-
ciation with HSCR include interstitial deletion of
distal 13q, partial deletion of 2p and reciprocal
translocation, and trisomy 18 mosaic. A number
of unusual hereditary syndromes have been
reported in patients with HSCR.  These include
Shah-Waardenburg syndrome, multiple endo-
crine neoplasia (MEN) type 2 syndrome, con-
genital central hypoventilation syndrome
(Ondine’s curse), Goldberg-Shprintzen syn-
drome, McKusick-Kaufman syndrome, Bardet-­
Biedl syndrome, and Smith-Lemli-Opitz
syndrome (Puri and Nakamura 2019).
Associated congenital anomalies have been
reported in around 20% of HSCR patients (Pini
Prato et al. 2013) and include cardiac malforma-
tions, gastrointestinal malformations, kidney and
urinary tract (CAKUT) malformations, craniofa-
cial anomalies, cleft palate, and polydactyly.
Associated congenital cardiac anomalies have
been reported in 5–8% of patients with HSCR
(Duess and Puri 2015; Hofmann et al. 2014). An
936 P. Puri

associated urological anomaly has been reported

in 4% of patients with HSCR (Bradnock et  al.
2017; Hofmann et  al. 2014). The association
between HSCR and multiple endocrine neoplasia
type 2a (MEN2A) has rarely been reported.
Coyle et al. (2014) found that while the overall
incidence of HSCR co-occurring with MEN2A is
low, both conditions occur with a relatively high
frequency in families with a RET mutation at
Exon 10. In a literature search, they found the co-­
occurrence of HSCR and MEN2A was recorded
in 84 cases (Coyle et al. 2014).

69.5 Pathophysiology

The pathophysiology of HSCR is poorly under-

stood. It has long been recognized that the
obstructive symptoms in HSCR are secondary to
the abnormal motility of the narrow distal seg-
ment, but there is still no clear explanation for the
occurrence of the spastic or tonically contracted
segment of bowel. Normal intestinal motility
requires coordinated interaction of the ENS,
smooth muscle cells (SMCs), interstitial cells of
Cajal (ICCs), and platelet-derived growth factor
receptor α-positive (PDGFRα+) cells, together
known as the SIP syncytium. Several abnormali-
ties of ENS and the SIP syncytium have been
described to explain the basis for motility dys-
function in the contracted bowel in HSCR
(Nemeth et al. 2002; Rolle et  al. 2002, b;
O’Donnell et al. 2019; Puri et al. 2020).
69.6 Pathology
The characteristic gross pathological feature in
HSCR is dilation and hypertrophy of the proxi-
mal colon with an abrupt or gradual transition to
a narrow distal bowel (Fig.  69.1). Although the
degree of dilation and hypertrophy increases with
age, the cone-shaped transitional zone from the
dilated to the narrow bowel is usually evident in
the newborn.
Histologically, HSCR is characterized by the
absence of ganglionic cells in the myenteric and
submucous plexuses and the presence of hyper-
trophied nonmyelinated nerve trunks in the space
normally occupied by the ganglionic cells. The
Fig. 69.1  Typical gross pathology in HSCR, with a tran-
sitional zone at the rectosigmoid level
aganglionic segment of bowel is followed proxi-
mally by a hypoganglionic segment of varying
length. This hypoganglionic zone is characterized
by a reduced number of ganglion cells and nerve
fibers in myenteric and submucous plexuses.
69.7 Diagnosis
The diagnosis of HSCR is usually based on clini-
cal history, imaging, and rectal biopsy and con-
firmed by histological examination of a rectal
biopsy.
69.7.1 Clinical Presentation
HSCR should be considered in any child who has
a history of constipation dating back to the new-
born period. The median age at which children
are diagnosed with HSCR has progressively
decreased over the past decades with greater
awareness of the disease. Of all the cases of
HSCR, 80–90% produce clinical symptoms and
are diagnosed in the neonatal period. Delayed
passage of meconium is the cardinal symptom in
69  Hirschsprung’s Disease 937

neonates. Over 90% of affected patients fail to

pass meconium in the first 24 h of life. The usual
presentation of HSCR in the neonatal period is
with constipation, abdominal distention and
­bile-­stained vomiting. Recently, in a European
surveillance study which collected 1039 isolated
cases of HSCR between 1980 and 2009, Best
et al. (2014) reported that the time of diagnosis
was known in 803 (77.3%) isolated cases. A con-
genital anomaly was antenatally suspected in 11
(1.4%) isolated cases. HSCR was diagnosed at
birth in 161 (20.0%) cases, in the first week in
354 (44.1%) cases, between 1 and 4 weeks in 116
(14.4%) cases, between 1 and 12  months in 95
(11.8%), and after 12 months in 25 (3.1%) cases
(Best et al. 2014). The remaining 41 cases were
postnatally diagnosed but at an unknown time. In
a prospective survey from 1997 to 2000, the
Australian Paediatric Surveillance Unit reported
that the diagnosis of HSCR in the newborn period
was made in 90.5% of patients (Singh et  al.
2003). The neonate with HSCR is usually a full-­
term baby and presents with a distended abdo-
men, feeding intolerance with bilious aspirates or Fig. 69.2  A 2-day-old infant with marked abdominal dis-
bilious vomiting, and classically, with delay in tention and failure to pass meconium. Suction rectal
biopsy confirmed HSCR
the passage of meconium (Fig.  69.2). In many
cases, a rectal examination or rectal irrigation
causes the passage of meconium and relief of ure to thrive (Puri and Tomuschat 2018). This is
acute intestinal obstruction. most common among breast-fed infants who may
Among normal full-term infants, 98% pass develop constipation around the time of weaning
meconium in the first 24 h of life, and the remain- (Puri and Tomuschat 2018). Rectal examination
der will pass their first stool within 48 h. It has of patients with HSCR may show a tight anus
always been said that over 90% of HSCR infants (Puri and Tomuschat 2018). After a careful his-
fail to pass meconium in the first 24 h of life (Puri tory and physical examination, the diagnostic
and Tomuschat 2018). However, several authors steps will include barium enema, anorectal
have found that more than 40% of HSCR new- manometry, and a rectal biopsy.
borns pass meconium in the first 24  h of life About one third of babies with HSCR present
(Singh et al. 2003). Thus, one should not be dis- with diarrhea. Diarrhea in HSCR is always a
suaded from carrying out a rectal suction biopsy symptom of enterocolitis which is the common-
in the absence of a history of delayed passage of est cause of morbidity and mortality. The classic
meconium. A prenatal history suggestive of symptoms of Hirschsprung-associated enteroco-
intestinal obstruction is rare, except in children litis (HAEC) include abdominal distention, fever,
with total colonic aganglionosis. Occasionally, a and diarrhea. However, there is a broad clinical
diagnosis of HSCR should be considered in the spectrum with which children present, and other
presence of unexplained perforation of the cecum signs or symptoms may include vomiting, rectal
or appendix, although this is a rare presentation bleeding, lethargy, loose stools, and obstipation
(Puri 2009). Some children do not become (Gosain et  al. 2017). The reported incidence of
obstructed in the neonatal period and present HAEC varies widely, ranging from 6–60% prior
later in infancy or in adulthood with severe con- to definitive pull-through operation and from
stipation, chronic abdominal distension, and fail- 25–37% post operation (Gosain et  al. 2017;
938
Nakamura et  al. 2018). While all patients with
HSCR have a risk of HAEC, several features
appear to be associated with an increased risk.
These include Down’s syndrome, long-segment
aganglionosis, prior HAEC, and obstruction due
to any cause (Gosain et al. 2017).
69.7.2 Imaging
Plain abdominal films in a neonate with HSCR
will show dilated loops of bowel with fluid levels
and airless pelvis. Occasionally, one may be able
to see a small amount of air in the un-distended
rectum and dilated colon above it raising the sus-
picion of HSCR (Fig.  69.3a). Plain abdominal
radiographs obtained from patients with total
colonic aganglionosis (TCA) may show charac-
teristic signs of ileal obstruction with air fluid
levels or simple gaseous distension of small
intestinal loops.
a b
Fig. 69.3  HSCR. (a) Abdominal radiograph in a 4-day-­
old infant showing marked dilation of large and small
bowel loops. Note the gas in the undilated rectum. (b)
P. Puri
In patients with enterocolitis complicating
HSCR, plain abdominal radiography may show a
thickening of the bowel wall with mucosal irreg-
ularity or a grossly dilated colon loop, indicating
toxic megacolon. Pneumoperitoneum may be
found in those with perforation. Spontaneous
perforation of the intestinal tract has been
reported in 3% of patients with HSCR.
Barium enema performed by an experienced
radiologist, using careful technique, should
achieve a high degree of reliability in diagnosing
HSCR in the newborn. It is important that the
infant should not have rectal washouts or even
digital examinations prior to barium enema, as
such interference may distort the transitional
zone appearance and give a false-negative diag-
nosis. A soft rubber catheter is inserted into the
lower rectum and held in position with firm
strapping across the buttocks. A balloon catheter
should not be used due to the risk of perforation
and the possibility of distorting a transitional
Barium enema in this patient reveals transitional zone at
the sigmoid level
69  Hirschsprung’s Disease
zone by distension. The barium should be
injected slowly in small amounts under fluoro-
scopic control with the baby in the lateral posi-
tion. A typical case of HSCR will demonstrate
the flow of barium from the undilated rectum
through a cone-­shaped transitional zone into the
dilated colon (Fig. 69.3b). Some cases may show
an abrupt transition between the dilated proxi-
mal colon and the distal aganglionic segment,
leaving the diagnosis in little doubt. In some
cases, the findings on the barium enema are
uncertain, and a delayed film at 24 h may con-
firm the diagnosis by demonstrating the retained
barium and often accentuating the appearance of
the transitional zone.
In TCA, the contrast enema is not pathogno-
monic and may not provide a definitive diagno-
sis. The colon in TCA is of normal caliber in
25–77% of cases.
a Air Inflation
Rectum
Internal Sphincter
External Sphincter
Rectosphincteric Reflex
b
Air Inflation Air Inflation
Rectum
Internal Sphincter
External Sphincter
Fig. 69.4  Anorectal manometry. (a) Normal rectosphincteric reflex on rectal balloon inflation. (b) Absence of recto-
sphincteric reflex and marked internal sphincter rhythmic activity in a patient with HSCR
939
69.7.3 Anorectal Manometry
In the normally innervated bowel, distension of
the rectum produces relaxation of the internal
sphincter rectosphincteric reflex. In normal per-
sons, upon distending the rectal balloon with air,
the rectum immediately responds with a transient
rise in pressure lasting 15–20 s; at the same time,
the internal sphincter rhythmic activity is
depressed or abolished, and its pressure falls by
15–20 cm, the duration of relaxation coinciding
with the rectal wave.
In patients with HSCR, the rectum often
shows spontaneous waves of varying amplitude
and frequency in the resting phase. The internal
sphincter rhythmic activity is more pronounced.
On rectal distension, with an increment of air,
there is a complete absence of internal sphincter
relaxation (Fig.  69.4a, b). Failure to detect the
Air Inflation
Rectosphincteric Reflex
Air Inflation
940
rectosphincteric reflex in premature and term
infants is believed to be due to technical difficul-
ties and not to immaturity of ganglion cells. Light
sedation, particularly in infants and small chil-
dren, may overcome technical difficulties
encountered in this age group.
69.7.4 Rectal Biopsy
The diagnosis of HSCR is confirmed on exami-
nation of rectal biopsy specimens. The introduc-
tion of a histochemical staining technique for the
detection of acetylcholinesterase (AChE) activ-
ity in suction rectal biopsy has resulted in a reli-
able and simple method for the diagnosis of
HSCR.  Full thickness rectal biopsy is rarely
indicated for the diagnosis of HSCR except in
total colonic aganglionosis. A suction rectal
biopsy should involve sampling a segment of the
rectal wall 2  cm proximal to the dentate line,
along the posterior wall of the rectum. A biopsy
taken too distally may obtain a specimen from
the physiologically aganglionic region errone-
ously, suggesting the presence of HSCR, whereas
a biopsy taken too proximally (i.e., 5+ cm) may
a b
Fig. 69.5  Acetylcholinesterase staining of suction rectal
biopsy. (a) Normal rectum showing minimal acetylcholin-
esterase staining in the mucosa, lamina propria, and mus-
P. Puri
miss very-­short-­segment HSCR. In normal per-
sons, barely detectable acetylcholinesterase
activity is observed within the lamina propria
and muscularis mucosa, and submucosal gan-
glion cells stain strongly for acetylcholinester-
ase. In HSCR, there is a marked increase in
acetylcholinesterase activity in the lamina pro-
pria and muscularis which is evident as coarse,
discrete cholinergic nerve fibers stained brown
to black (Fig. 69.5a, b).
In TCA, AChE activity in suction rectal biop-
sies presents an atypical pattern, different from
the classic one. Positive AChE fibers can be
found in the lamina propria as well as the muscu-
laris mucosae. However, cholinergic fibers pres-
ent a lower density than in classical HSCR.
In recent years, calretinin, a calcium-binding
protein that functions as a calcium sensor/modu-
lator and expressed in submucosal and myenteric
ganglion cells and mucosal nerve fibers, has been
described as an adjunctive or primary diagnostic
test in HSCR (Kapur 2019). Aganglionic seg-
ments completely lack calretinin immunoreactiv-
ity in enteric nerves (Kapur et  al. 2009; Kapur
2019). The sensitivity and specificity of calretinin
IHC are equivalent to rapid AChE, with the addi-
cularis mucosae. (b) HSCR characterized by marked
staining of cholinesterase-positive nerves in the lamina
propria and muscularis mucosae
69  Hirschsprung’s Disease
Fig. 69.6  Calretinin staining of a rectal suction biopsy
tional advantage that calretinin IHC may be
informative when inadequate tissue is available
to establish an H&E diagnosis (Fig. 69.6).
69.8 Differential Diagnosis
Several conditions must be considered where an
infant is being evaluated for HSCR.  Common
differential diagnoses include colonic atresia,
meconium ileus, meconium plug syndrome,
small left colon syndrome, low anorectal malfor-
mation, and intestinal motility disorders. Colonic
atresia gives similar plain film findings to HSCR
but is readily excluded with barium enema show-
ing complete mechanical obstruction. Distal
small bowel atresia shows gross distension of the
bowel loop immediately proximal to the obstruc-
tion with the widest fluid level in it.
In meconium ileus, the typical mottled thick
meconium may be seen. Also, clear and sharp
fluid levels are not a feature in erect or lateral
decubitus views. However, HSCR can sometimes
941
simulate meconium ileus in plain films and may
give equivocal findings on Gastrografin or bar-
ium enema.
Meconium plugs obstructing the colon can
present as HSCR with strongly suggestive history
and plain films. Small left colon syndrome with
marked distension proximal to a narrowed
descending colon also simulates HSCR at the left
colonic flexure. These two conditions usually
resolve with Gastrografin enema, but a minority
of these cases will actually have HSCR, which
should be excluded clinically.
69.9 Management
Once the diagnosis of HSCR has been confirmed
by rectal biopsy examination, the infant should
be prepared for surgery. If the newborn has
enterocolitis complicating HSCR, the correction
of dehydration and electrolyte imbalance by infu-
sion of appropriate fluids will be required. It is
essential to decompress the bowel as early as
942
possible in these babies. Deflation of the intestine
may be carried out by rectal irrigations, but some
babies may require colostomy.
In recent years, the vast majority of cases of
HSCR are diagnosed in the neonatal period.
Many centers are now performing one-stage pull-­
through operations in the newborn with minimal
morbidity rates and encouraging results. The
advantages of operating on the newborn are that
the colonic dilation can be quickly controlled by
washouts and at operation, the caliber of the pull-­
through bowel is near normal, allowing for an
accurate anastomosis that minimizes leakage and
cuff infection. A number of different operations
have been described for the treatment of
HSCR. The four most commonly used operations
are the rectosigmoidectomy developed by
Swenson and Bill, the retrorectal approach devel-
oped by Duhamel, the endorectal procedure
developed by Soave, and deep anterior colorectal
anastomosis developed by Rehbein. The basic
principle in all these procedures is to bring the
ganglionic bowel down to the anus. The long-­
term results of any of these operations are satis-
factory if they are performed correctly. Recently,
a number of investigators have described and
advocated a variety of one-stage pull-through
procedures in the newborn using minimally inva-
sive laparoscopic techniques. Many pediatric sur-
geons in recent years have performed a transanal
endorectal pull-through operation performed
without opening the abdomen and have reported
excellent results in rectosigmoid HSCR (Puri
et al. 2020).
69.9.1 Role of Colostomy
It is important to recognize a stoma may still be
indicated for children with severe enterocolitis,
perforation, malnutrition, or massively dilated
proximally bowel, as well as in situations where
there is inadequate pathology support to identify
the transition zone on frozen sections. Many sur-
geons prefer right transverse colostomy; others
advocate performing colostomy just above the
transition to the ganglionic bowel. Ileostomy is
P. Puri
indicated in patients who have TCA (Puri et al.
2020).
69.9.2 Transanal One-Stage
Endorectal Pull-Through
Operation
Over 80% of patients with HSCR have rectosig-
moid aganglionosis. A one-stage pull-through
operation can be successfully performed in these
patients using a transanal endorectal approach,
without opening the abdomen. This procedure is
associated with excellent clinical results and per-
mits early postoperative feeding, early hospital
discharge, no visible scars, and a low incidence
of enterocolitis (Ruttenstock and Puri 2010; Puri
and Coyle 2019).
A good barium enema study is essential for
the transanal one-stage pull-through operation. A
typical study of rectosigmoid HSCR will show
the flow of barium from the undilated rectum
through a cone-shaped transition zone into a
dilated sigmoid colon (Puri et al. 2020).
69.9.3 Operative Technique
The patient is positioned on the operating table in
the lithotomy position. The legs are strapped over
sandbags. A Foley catheter is inserted into the
bladder. A Denis-Browne retractor or anal retrac-
tor is placed to retract perianal skin. The rectal
mucosa is circumferentially incised using the
cautery with a fine-tipped needle, approximately
5 mm from the dentate line, and the submucosal
plane is developed. The proximal cut edge of the
mucosal cuff is held with multiple fine silk
sutures, which are used for traction (Fig.  69.7).
The endorectal dissection is then carried proxi-
mally, staying in the submucosal plane.
When the submucosal dissection has extended
for about 3 cm, the rectal muscle is divided cir-
cumferentially, and the full thickness of the rec-
tum and sigmoid colon is mobilized out through
the anus. This requires the division of rectal and
sigmoid vessels, which can be done under direct
vision using cautery.
69  Hirschsprung’s Disease
When the transition zone is encountered,
full-­
thickness biopsy sections are taken, and
frozen section confirmation of ganglion cells is
obtained. The rectal muscular cuff is split longi-
tudinally either anteriorly or posteriorly. The
colon is then divided several centimeters above
the most proximal normal biopsy site (Fig. 69.7),
and a standard Soave-Boley anastomosis is per-
formed. No drains are placed. The patient is
started on oral feeds after 24 h and discharged
home on the third postoperative day. Digital rec-
tal examination is performed 2 weeks after the
operation. Routine rectal dilatation is not per-
formed unless there is evidence of a stricture
(Puri and Tomuschat 2018; Puri and Coyle
2019).
a b
d e
Fig. 69.7  Transanal endorectal pull-through. (a) Rectal
mucosa is circumferentially incised using the needle-tip
cautery approximately 5 mm above the dentate line, and
submucosa plane is developed. (b) When the submucosal
dissection is extended proximally for about 3  cm, the
muscle is divided circumferentially, and the full thickness
943
69.9.4 Laparoscopic-Assisted
Pull-Through
Despite the fact that many of the benefits of the
pull-through procedure can be attained by trans-
anal pull-through alone, significant advantages
are realized by performing the laparoscopic dis-
section (Georgeson and Robertson 2004). Prior
to the irreversible step of endorectal dissection,
laparoscopy allows the verification of the pres-
ence of ganglion cells in the proximal colon ped-
icle by seromuscular biopsy. In TCA cases, a
different type of pull-through operation or
ostomy may be considered. Also, laparoscopic
mobilization of the intra-abdominal segment
increases the mobility of the rectum and makes
c
of the rectum and sigmoid colon is mobilized out through
the anus. (c) On reaching the transition zone, full-­
thickness rectal biopsies are taken for frozen section to
confirm ganglion cells. (d) The colon is divided several
centimeters above the most proximal biopsy site. (e) A
standard Soave-Boley anastomosis is performed
944
Fig. 69.8  The placement of trocars for laparoscopic-assisted pull-through operation
the endpoint of the endorectal dissection more
definitive (Georgeson et al. 1995).
Initially, three ports (Fig. 69.8) are placed to
visualize the transition zone and to perform biop-
sies for histologic leveling. A window is then
made between the colon and superior rectal ves-
sels, and a distal dissection of the aganglionic
colon is then performed circumferentially, keep-
ing close to the colon wall, carefully preserving
the mesenteric blood supply to the rectum. A
blunt and sharp dissection of avascular plane pos-
terior to the rectum follows. Anteriorly, the rec-
tum is dissected for about 1–2  cm below the
peritoneal reflection. It is important to avoid too
extensive a lateral dissection, where damage to
the nervi erigentes can result. Proximal mesen-
teric dissection of the colon pedicle with careful
preservation of the marginal artery depends on
the extent of the aganglionic segments; some
patients require sigmoid colon mobilization or
division of the lateral colonic fusion fascia up to
the splenic flexure. After the endoscopic dissec-
tion of the colon and rectum has been completed,
the pneumoperitoneum is released, and the
instruments are removed, and the procedure is
completed with an endorectal transanal
pull-through.
Once the anastomosis of the proximal gangli-
onated colon with the anorectal cuff is completed,
P. Puri
reinsufflation for pneumoperitoneum can be per-
formed to inspect the colon pedicle for twisting
or potential internal herniation (Georgeson et al.
1995; Puri and Coyle 2019).
69.9.5 Postoperative Care
Most infants undergoing pull-through for HSCR
can be fed the next day. The anastomosis should
be calibrated with an appropriately sized dilator
2 weeks after the procedure.
69.9.6 Complications
Early postoperative complications which can
occur after any type of pull-through operation
include wound infections, anastomotic leak,
anastomotic stricture, retraction, or necrosis of
the neorectum, intestinal adhesions, and ileus.
Late complications include constipation, entero-
colitis, incontinence, anastomotic problems,
adhesive bowel obstruction, and urogenital com-
plications (Puri and Tomuschat 2018).
69.9.6.1 Anastomotic Leak
The most dangerous early postoperative compli-
cation following the definitive abdominoperineal
69  Hirschsprung’s Disease
pull-through procedure is leakage at the anasto-
motic suture line. Factors which are responsible
for anastomotic leak include ischemia of the dis-
tal end of the colonic pull-through segment, ten-
sion on the anastomosis, incomplete anastomotic
suture lines, and inadvertent rectal manipulation.
If a leak is recognized in a patient without a
colostomy, it is imperative to perform a diverting
colostomy promptly, to administer intravenous
antibiotics, and to irrigate the rectum with antibi-
otic solution a few times daily. Delay in estab-
lishing fecal diversion is likely to result in an
extensive pelvic abscess which may require lapa-
rotomy and transabdominal drainage (Puri and
Tomuschat 2018).
69.9.6.2 Retraction of Pull-Through
Retraction of a portion, or all of the colonic seg-
ment from the anastomosis, can occur and is usu-
ally seen within 3  weeks of the operation.
Evaluation under general anesthesia is generally
necessary. In occasional patients, resuturing the
anastomosis may be feasible transanally. For
those with separation of less than 50% of the
anastomosis but with adequate vascularity of the
colon, a diverting colostomy for approximately
3  months is necessary. For patients with wide
separation at the anastomosis, early transabdomi-
nal reconstruction of the pull-through is recom-
mended (Puri and Tomuschat 2018).
69.9.6.3 Perianal Excoriation
Perianal excoriation occurs in nearly half of the
patients undergoing a pull-through procedure,
but generally resolves within 3 months with local
therapy and resolution of diarrhea. It is helpful to
begin placing a barrier cream on the perianal skin
promptly after the operation and to continue after
each movement for the first few weeks. The reso-
lution of diarrhea will often hasten the clearance
of perianal skin irritation.
69.9.6.4 Enterocolitis
As mentioned earlier, HAEC is a significant com-
plication of HSCR both in the pre- and postoper-
ative periods. HAEC can occur at any time from
the neonatal period to adulthood and can be inde-
pendent of the medical management and surgical
945
procedure performed. The incidence of enteroco-
litis ranges from 20% to 58%. Fortunately, the
mortality rate has declined over the last 30 years
from 30% to 1%. This decrease in mortality is
related to earlier diagnosis of HSCR and entero-
colitis, rectal decompression, appropriate vigor-
ous resuscitation, and antibiotic therapy. It has
been reported that routine postoperative rectal
washouts decrease both the incidence and the
severity of the episodes of enterocolitis following
definitive surgery. In episodes of recurrent entero-
colitis, which can develop in up to 56% of
patients, anal dilatations have been recom-
mended. However, prior to commencing a treat-
ment regime, a contrast enema should be
performed to rule out a mechanical obstruction.
Patients with a normal rectal biopsy may require
a sphincterotomy (Puri and Tomuschat 2018).
69.9.6.5 Constipation
Constipation is common after the definitive repair
of HSCR and can be due to residual agangliono-
sis and high anal tone. Repeated and forceful anal
dilations of Botulinum toxin injection into the
sphincter under general anesthesia may resolve
the problem. In some patients, internal sphincter
myectomy may be needed. In patients with scar-
ring, stricture, or intestinal neuronal dysplasia
proximal to the aganglionic segment, treatment is
based on the underlying cause.
69.9.6.6 Soiling
Soiling is fairly common after all types of pull-­
through operations, its precise incidence is pri-
marily dependent on how assiduously the
investigator looks for it. The reported incidence
of soiling ranges from 10% to 30%. The attain-
ment of normal postoperative defecation is
clearly dependent on the intensity of bowel train-
ing, social background, and respective intelli-
gence of the patients. Mental handicap, including
Down’s syndrome, is invariably associated with
long-term incontinence. Those patients with pre-
operative enterocolitis would also seem to have a
marginally higher long-term risk of incontinence.
In some patients in whom soiling is intractable
and a social problem, a Malone procedure may
be needed to stay clean.
946
69.10 Conclusions
The vast majority of patients treated with any one
of the standard pull-through procedures achieve
satisfactory continence and function with time
(Menezes et al. 2006). The attainment of normal
continence is dependent on the intensity of bowel
training, social background, and respective intel-
ligence of patients. Enterocolitis remains the
most common complication of HSCR. Many
patients with enterocolitis complicating HSCR
continue to have disturbances of bowel function
several years after surgery for HSCR (Menezes
and Puri 2006). Patients with total colonic agan-
glionosis continue to have problems with bowel
control although many improve with time
(Menezes et al. 2008). Mental handicap, includ-
ing Down’s syndrome, is invariably associated
with long-term incontinence (Granstrom et  al.
2015; Friedmacher and Puri 2013; Menezes and
Puri 2005).
The progress in understanding normal gut
development and motility has led to an expand-
ing field of research into developing novel thera-
pies for HSCR. During the last decade, there has
been increasing focus on the development of
novel stem cell-based therapies for the treatment
of HSCR (McCann et al. 2019). Stem cell trans-
plantation, using laboratory-cultured neural stem
cells (NSCs) to colonize aganglionic intestine
and restore intestinal motility, has been proposed
as a treatment for HSCR.  Further research is
needed to determine the optimal source of stem
cells to replace the ENS in HSCR and determine
the best way to deliver stem cells to the bowel.
References
Anderson JE, Vanover MA, Saadai P, Stark RA,
Stephenson JT, Hirose S (2018) Epidemiology of
Hirschsprung’s disease in California from 1995 to
2013. Pediatr Surg Int 34:1299–1303
Best KE, Addor MC, Arriola L, Balku E, Barisic I,
Bianchi F et al (2014) Hirschsprung’s disease preva-
lence in Europe: a register based study. Birth Defects
Res A Clin Mol Teratol 100(9):695–702
Bradnock TJ, Knight M, Kenny S, Nair M, Walker GM
(2017) British Association of Pediatric Surgeons con-
genital anomalies surveillance. Hirschsprung’s disease
P. Puri
in the UK and Ireland: incidence and anomalies. Arch
Dis Child 102:722–727
Burns AJ, Roberts RR, Bornstein JC, Young HM (2009)
Development of the enteric nervous system and its role
in intestinal motility during fetal and early postnatal
stages. Semin Pediatr Surg 18(4):196–205
Chia ST, Chen SC, Lu CL, Sheu SM, Kuo HC (2016)
Epidemiology of Hirschsprung’s disease in Taiwanese
children: a 13-year nationwide population-based
study. Pediatr Neonatol 57(3):201–206
Coyle D, Friedmacher F, Puri P (2014) The association
between Hirschsprung’s disease and multiple endo-
crine neoplasia type 2a: a systematic review. Pediatr
Surg Int 30(8):751–756
Coyle D, O’Donnell AM, Tomuschat C, Gillick
J, Puri P (2019 Nov) The extent of the transi-
tion zone in Hirschsprung disease. J Pediatr Surg
54(11):2318–2324
Duess JW, Hofmann AD, Puri P (2014) Prevalence of
Hirschsprung’s disease in premature infants: a system-
atic review. Pediatr Surg Int 30(8):791–795
Duess JW, Puri P (2015) Syndromic Hirschsprung’s dis-
ease and associated congenital heart disease: a system-
atic review. Pediatr Surg Int 31(8):781–785
Fiori MG (1998) Domenico Battini and his description
of congenital megacolon: a detailed case report one
century before Hirschsprung. J Peripher NervSyst
3(3):197–206
Friedmacher F, Puri P (2013) Hirschsprung’s disease
associated with down syndrome: a meta-analysis of
incidence, functional outcomes and mortality. Pediatr
Surg Int 29(9):937–946
Georgeson KE, Fuenfer MM, Hardin WD (1995) Primary
laparoscopic pull-through for Hirschsprung’s disease
in infants and children. J Pediatr Surg 30(7):1017–
1021. discussion 21–2
Georgeson KE, Robertson DJ (2004) Laparoscopic-­
assisted approaches for the definitive surgery
for Hirschsprung’s disease. Semin Pediatr Surg
13(4):256–262
Gosain A, Frykman PK, Cowles RA, Horton J, Levitt M,
Rothstein DH et al (2017) Guidelines for the diagnosis
and management of Hirschsprung-associated entero-
colitis. Pediatr Surg Int 33(5):517–521
Granstrom AL, Danielson J, Husberg B, Nordenskjold
A, Wester T (2015) Adult outcomes after surgery for
Hirschsprung’s disease: evaluation of bowel function
and quality of life. J Pediatr Surg 50(11):1865–1869
Heanue TA, Pachnis V (2007) Enteric nervous system
development and Hirschsprung’s disease: advances
in genetic and stem cell studies. Nat Rev Neurosci
8(6):466–479
Hofmann AD, Duess JW, Puri P (2014) Congenital anom-
alies of the kidney and urinary tract (CAKUT) associ-
ated with Hirschsprung’s disease: a systematic review.
Pediatr Surg Int 30(8):757–761
Kapur RJ (2019) Anatomic pathology of Hirschsprung’s
disease. In: Puri P (ed) Hirschsprung’s disease and
allied disorders. Springer Nature, Switzerland,
pp 255–281
69  Hirschsprung’s Disease
Kapur RP, Reed RC, Finn LS, Patterson K, Johanson J,
Rutledge JC (2009) Calretinin immunohistochemis-
try versus acetylcholinesterase histochemistry in the
evaluation of suction rectal biopsies for Hirschsprung
disease. Pediatr Dev Pathol 12(1):6–15
Lof Granstrom A, Svenningsson A, Hagel E, Oddsberg J,
Nordenskjold A, Wester T (2016) Maternal risk factors
and perinatal characteristics for Hirschsprung disease.
Pediatrics 138(1)
Mc Laughlin D, Puri P (2015) Familial Hirschsprung’s
disease: a systematic review. Pediatr Surg Int
31(8):695–700
McCann CJ, Goldstein AM, Hotta R, Thapar N, Hofstra
MW, Burns AJ (2019) Stem cell therapy for enteric
neuropathies. In: Puri P (ed) Hirschsprung’s disease
and allied disorders. Springer Nature, Switzerland,
pp 133–152
Menezes M, Corbally M, Puri P (2006) Long-term results
of bowel function after treatment for Hirschsprung’s
disease: a 29-year review. Pediatr Surg Int
22(12):987–990
Menezes M, Pini Prato A, Jasonni V, Puri P (2008) Long-­
term clinical outcome in patients with total colonic
aganglionosis: a 31-year review. J Pediatr Surg
43(9):1696–1699
Menezes M, Puri P (2005) Long-term clinical outcome
in patients with Hirschsprung’s disease and associated
Down’s syndrome. J Pediatr Surg 40(5):810–812
Menezes M, Puri P (2006) Long-term outcome of patients
with enterocolitis complicating Hirschsprung’s dis-
ease. Pediatr Surg Int 22(4):316–318
Moore SW (2015) Total colonic aganglionosis and
Hirschsprung’s disease: a review. Pediatr Surg Int
31(1):1–9
Moore SW (2017) Genetic impact on the treatment &
management of Hirschsprung disease. J Pediatr Surg
52(2):218–222
Moore SW (2019) Congenital anomalies and genetic
associations in Hirschsprung’s disease. In: Puri P (ed)
Hirschsprung’s disease and allied disorders. Springer
Nature, Switzerland, pp 175–200
Moore SW, Zaahl M (2015) Clinical and genetic correla-
tions of familial Hirschsprung’s disease. J Pediatr Surg
50(2):285–288
Nakamura H, Lim T, Puri P (2018) Probiotics for the pre-
vention of Hirschsprung-associated enterocolitis: a
systematic review and meta-analysis. Pediatr Surg Int
34(2):189–193
Nasr A, Sullivan KJ, Chan EW, Wong CA, Benchimol
EI (2017) Validation of algorithms to determine inci-
dence of Hirschsprung disease in Ontario, Canada: a
population-based study using health administrative
data. Clin Epidemiol 9:579–590
Nemeth L, Rolle U, Puri P (2002) Altered cytoskeleton in
smooth muscle of aganglionic bowel. Arch Pathol Lab
Med 126(6):692–696
Nemeth L, Yoneda A, Kader M, Devaney D, Puri P (2001)
Three-dimensional morphology of gut innervation in
total intestinal aganglionosis using whole-mount prep-
aration. J Pediatr Surg 36(2):291–295
947
O’Donnell AM, Montedonico S, Puri P (2019)
Pathophysiology of Hirschsprung’s disease. In: Puri
P (ed) Hirschsprung’s disease and allied disorders.
Springer Nature, Switzerland, pp 153–166
Orr JD, Scobie WG (1983) Presentation and incidence of
Hirschsprung’s disease. Br Med J 287(6406):1671
Pini Prato A, Rossi V, Mosconi M, Holm C, Lantieri F,
Griseri P et  al (2013) A prospective observational
study of associated anomalies in Hirschsprung’s dis-
ease. Orphanet J Rare Dis 8:184
Puri P (2009) Hirschsprung’s disease and variants. In:
Puri P, Hollwarth M (eds) Pediatric surgery: diagnosis
and management. Springer, Heidelberg, pp 453–462
Puri P, Coyle D (2019) Hirschsprung’s disease. In: Puri
P, Hollwarth M (eds) Pediatric surgery. Springer,
Germany, pp 249–259
Puri P, Nakamura H (2019) Epidemiology and clinical
characteristics of Hirschsprung’s disease. In: Puri
P (ed) Hirschsprung’s disease and allied disorders.
Springer Nature, Switzerland, pp 167–174
Puri P, Tomuschat C (2018) Hirschsprung’s disease. In:
Puri P (ed) Newborn surgery, 4th edn. CRC Press,
Boca Raton, pp 667–683
Puri P, Tomuschat C, Nakamura H (2020) Hirschsprung’s
disease. In: Puri P (ed) Pediatric surgery. Springer
Nature, Germany, pp 1011–1030
Raveenthiran V (2011) Knowledge of ancient Hindu
surgeons on Hirschsprung disease: evidence from
Sushruta Samhita of circa 1200–600 BC.  J Pediatr
Surg 46(11):2204–2208
Rolle U, Nemeth L, Puri P (2002) Nitrergic innervation of
the normal gut and in motility disorders of childhood.
J Pediatr Surg 37(4):551–567
Rolle U, Piotrowska AP, Nemeth L, Puri P (2002) Altered
distribution of interstitial cells of Cajal in Hirschsprung
disease. Arch Pathol Lab Med 126(8):928–933
Ruttenstock E, Puri P (2009) A meta-analysis of clinical
outcome in patients with total intestinal aganglionosis.
Pediatr Surg Int 25(10):833–839
Ruttenstock E, Puri P (2010) Systematic review and meta-­
analysis of enterocolitis after one-stage transanal pull-­
through procedure for Hirschsprung’s disease. Pediatr
Surg Int 26(11):1101–1105
Senyuz OF, Buyukunal C, Danismend N, Erdogan E,
Ozbay G, Soylet Y (1989) Extensive intestinal agan-
glionosis. J Pediatr Surg 24(5):453–456
Singh SJ, Croaker GD, Manglick P, Wong CL,
Athanasakos H, Elliott E et al (2003) Hirschsprung’s
disease: the Australian paediatric surveillance Unit’s
experience. Pediatr Surg Int 19(4):247–250
Skaba R (2007) Historic milestones of Hirschsprung’s
disease (commemorating the 90th anniversary of pro-
fessor Harald Hirschsprung’s death). J Pediatr Surg
42(1):249–251
Swenson O (1996) Early history of the therapy of
Hirschsprung’s disease: facts and personal observa-
tions over 50 years. J Pediatr Surg 31(8):1003–1008
Swenson O (1999) How the cause and cure of
Hirschsprung’s disease were discovered. J Pediatr
Surg 34(10):1580–1581
948
Taghavi K, Goddard L, Evans SM, Hobson A, Beasley
SW, Sankaran S et al (2018) Ethnic variations in the
childhood prevalence of Hirschsprung disease in New
Zealand. ANZ J Surg 89(10):1246–1249
Taguchi T, Obata S, Ieiri S (2017) Current status of
Hirschsprung’s disease: based on a nationwide survey
of Japan. Pediatr Surg Int 33(4):497–504
Tam PK (2016) Hirschsprung’s disease: a bridge for sci-
ence and surgery. J Pediatr Surg 51(1):18–22
Tam PKH, Tang CSM, Garcia-Barcelo MM (2019)
Genetics of Hirschsprung’s disease. In: Puri P (ed)
P. Puri
Hirschsprung’s disease and allied disorders. Springer
Nature, Switzerland, pp 121–132
Tomuschat C, Puri P (2015) RET gene is a major risk
factor for Hirschsprung’s disease: a meta-analysis.
Pediatr Surg Int 31(8):701–710
Ziegler MM, Ross AJ 3rd, Bishop HC (1987) Total intes-
tinal aganglionosis: a new technique for prolonged
survival. J Pediatr Surg 22(1):82–83
 Variant Hirschsprung’s Disease
Florian Friedmacher and Prem Puri
70.1 Introduction
There are a number of newborns and children
who present with clinical symptoms similar to
Hirschsprung’s disease (HD) despite the pres-
ence of ganglion cells in rectal biopsies. Various
terms, such as chronic idiopathic intestinal pseu-
doobstruction, intestinal hypoperistalsis syn-
drome or pseudo-HD, have been used to describe
these conditions over the years. At present, there
are only a few articles in the medical literature
that have attempted to standardize the terminol-
ogy of HD and allied intestinal disorders
(Holschneider et  al. 1994). In 1997, Prem Puri
suggested that variant HD may be a more appro-
priate description for this heterogeneous group of
functional bowel disorders (Box 70.1) in patients
who suffer from chronic constipation and abdom-
inal distension despite a ganglionic rectal biopsy
(Puri 1997). Specific histological, immunohisto-
chemical or electron microscopic investigations,
F. Friedmacher (*)
Department of Pediatric Surgery, University Hospital
Frankfurt, Goethe University Frankfurt,
Frankfurt (Main), Germany
e-mail: florian.friedmacher@nhs.net
P. Puri
Department of Pediatric Surgery, Beacon Hospital,
and University College Dublin, Dublin, Ireland
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_70
70
combined with anorectal manometry studies, are
required to distinguish between the different vari-
ants of HD (Friedmacher and Puri 2013).
Although the initial diagnostic workup and sub-
sequent management can be challenging, the
majority of these patients will have a satisfactory
long-term outcome.
Box 70.1Variants of Hirschsprung’s disease
(HD)
Intestinal neuronal dysplasia (IND)
Intestinal ganglioneuromatosis (GNM)
Isolated hypoganglionosis (HG)
Immature ganglia (IG)
Absence of the argyrophil plexus (AP)
Internal anal sphincter achalasia (IASA)
Megacystis-microcolon-intestinal
hyperperistalsis syndrome (MMIHS)
70.2 Intestinal Neuronal
Dysplasia
In 1971, William A.  Meier-Ruge first described
intestinal neuronal dysplasia (IND) as a hyperplas-
tic malformation of the enteric plexus (Meier-­Ruge
1971). A few years later, Puri et al. (1977) reported
a case of rectosigmoid aganglionosis associated
with IND of the descending and transverse colon.
Nowadays, IND can be classified into two clinical
949
950
and histological distinct subtypes (Fadda et  al.
1983). IND type A (IND A), occurring in less than
5% of all IND cases, is characterized by congenital
aplasia or hypoplasia of the sympathetic innerva-
tion. Patients with IND A typically present in the
neonatal period with episodes of abdominal disten-
sion, intestinal obstruction and diarrhea with
bloody stools. IND type B (IND B) is defined by
hyperplasia of the parasympathetic submucosal
and myenteric plexuses, accounting for over 95%
of all IND cases. Typical histological features of
IND B include hyperganglionosis, giant ganglia,
ectopic ganglion cells and increased activity of ace-
tylcholinesterase (AChE) in the lamina propria and
around submucosal blood vessels (Bruder and
Meier-Ruge 2007a). IND occurring in association
with HD is invariably IND B.
70.2.1 Epidemiology
IND occurs with an estimated incidence of 1 in
7500 newborns (Granero Cendón et  al. 2007).
However, the frequency of isolated IND cases
seems to be highly variable with reported rates
ranging between 0.3% and 40% of all rectal biop-
sies (Friedmacher and Puri 2013). IND immedi-
ately proximal to an aganglionic colon segment is
not uncommon and has been suggested as a pos-
sible cause of persistent bowel symptoms after
pull-through operation for HD (Kobayashi et al.
1995). Some authors have found IND in up to
44% of their HD patients (Montedonico et  al.
2011), whereas others have rarely encountered
IND in association with HD. The high variability
of patient age, specimen type and applied stain-
ing methods has resulted in considerable confu-
sion in the published literature regarding the
accurate diagnostic criteria (Schäppi et al. 2013).
70.2.2 Pathogenesis
The existence of IND as a distinct histopatho-
logical entity remains controversial (Schäppi
et al. 2013). Several authors have suggested that
the observed changes may be either a variant of
normal bowel development or a secondary
F. Friedmacher and P. Puri
acquired phenomenon caused by congenital
obstruction or inflammation (Martucciello et al.
2005). An underlying autoimmune mechanism
has also been proposed for IND (Friedmacher
and Puri 2013). Furthermore, there may be an
additional genetic component, as several familial
cases of IND have been found (Martucciello
et  al. 2002; Moore et  al. 1993). The strongest
evidence that IND is a real entity actually arose
from two different Hox11L1 knockout mouse
models (Puri and Gosemann 2012). In both
cases, homozygous mutant mice developed
megacolon at the age of 3 to 5  weeks, without
any further morphological changes. Histological
and immunohistochemical evaluation revealed
hyperplasia of the ganglia similar to the pheno-
type observed in human IND.  Another animal
model resulting in a phenotype similar to IND
was reported in rats with a heterozygous muta-
tion of the endothelin B receptor (Ednrb) and
showed features of hyperganglionosis, giant
ganglia and hypertrophied nerve fiber strands in
the submucosal plexus (Friedmacher and Puri
2013). However, mutational screening of human
patients with IND has not identified any altera-
tions in HOX11L1 and EDNRB genes to date
(Puri and Gosemann 2012).
70.2.3 Clinical Presentation
Most patients with IND present with chronic
constipation with or without abdominal disten-
sion, thus clinically resembling HD with absence
of internal anal sphincter relaxation on manome-
try, but with a normal contrast enema examina-
tion of the colon. It has been shown that intestinal
obstruction is the most characteristic clinical fea-
ture of IND in infants and young children
(Montedonico et  al. 2002). Furthermore, there
appears to be a high incidence of associated con-
genital anomalies, ranging from 25% to 30%
(Puri and Gosemann 2012). The most common
ones are anorectal malformations, megacystis,
intestinal malrotation, congenital short bowel,
hypertrophic pyloric stenosis, necrotizing entero-
colitis and Down syndrome (Martucciello et  al.
2002; Montedonico et al. 2002).
70  Variant Hirschsprung’s Disease 951

70.2.4 Diagnosis continues, mainly due to the lack of consensus on

The method of choice for the diagnosis of IND is
rectal suction biopsy. It is essential to include a
sufficient amount of submucosal tissue in the
biopsy specimens. Previously, the diagnosis of
IND was based on qualitative criteria, thus result-
ing in a high interobserver variation (Koletzko
et al. 1999). Therefore, the debate about the exis-
tence of IND as a distinct histopathological entity
diagnostic criteria (Martucciello et  al. 2005;
Schäppi et  al. 2013). Initially, IND was diag-
nosed on the basis of AChE immunohistochemis-
try of nerve fibers in rectal biopsies (Kobayashi
et al. 1995; Meier-Ruge et al. 1995; Montedonico
et  al. 2011) (Fig.  70.1a, b). However, as AChE
activity in the lamina propria mucosae has been
shown to be an age-dependent phenomenon that
disappears at the time of submucosal plexus mat-

a b

c d

Fig. 70.1  AChE immunohistochemistry of a normal rec- NADPH-d staining of a normal submucosal plexus (c).
tal suction biopsy (a). Rectal biopsy from a patient with Submucosal plexus of a patient with IND showing giant
IND, showing hyperganglionosis, giant ganglia and ganglia (d)
increased AChE activity in the lamina propria (b).
952
uration (Coerdt et al. 2004), more specific stain-
ing techniques were required to assess the enteric
nervous system in more detail. Thus, enzyme his-
tochemistry for lactate dehydrogenase, succinate
dehydrogenase and nitric oxide synthase has
been introduced to evaluate and diagnose IND B
quantitatively (Meier-Ruge and Bruder 2005).
Various other neuronal and glial markers, such as
nicotinamide adenine dinucleotide phosphate-­
diaphorase (NADPH-d) (Fig.  70.1c, d), neural
cell adhesion molecule (NCAM), neuron-specific
enolase (NSE), cathepsin D, protein gene product
9.5 (PGP9.5), S-100 protein, peripherin, synapto-
physin and cuprolinic blue, have also been used
(Meier-Ruge and Bruder 2005; Meier-Ruge et al.
2004; Friedmacher and Puri 2013). Cuprolinic
blue staining has been proposed as it stains the
whole population of ganglion cells (Puri and
Gosemann 2012), but only their cell bodies and
not their axons, which makes the differentiation
between the individual cell types relatively easy.
In addition, defective innervation of the neuro-
muscular junction within the affected bowel seg-
ment of patients with IND has been identified
(Friedmacher and Puri 2013). Abnormal submu-
cosal vasculature is a further histological finding
in isolated IND and IND associated with HD,
which may also be a useful diagnostic feature
(Friedmacher and Puri 2013). Furthermore, a
reduced number of c-Kit-positive interstitial cells
of Cajal (ICCs) has been demonstrated in the
myenteric plexus and muscle layers of IND cases
(Rolle et  al. 2007). More recently, a marked
reduction in the expression of phosphatase and
tensin homolog (PTEN) has been discovered in
the submucosal and myenteric plexuses of
patients with IND, which may explain the
observed motility dysfunction (Friedmacher and
Puri 2013). Due to age-dependent AChE expres-
sion and the discrepancy of applied staining tech-
niques, the most commonly used diagnostic
criteria at present are as follows: (1) more than
20% of 25 submucosal ganglia must be giant
ganglia containing 9 or more ganglion cells, and
(2) the patient must be older than 1 year, as before
that age, giant ganglia may be misinterpreted due
to the fact that immature ganglia often have an
incomplete differentiation in nerve cells (Bruder
F. Friedmacher and P. Puri
and Meier-Ruge 2007a; Meier-Ruge et al. 2004).
The majority of patients with IND do not display
any specific radiological features on contrast
enema studies other than rectosigmoid disten-
sion. The rectosphincteric reflex has often been
shown to be present, absent or atypical in these
patients (Puri and Gosemann 2012).
70.2.5 Management
In the first instance, the management of IND B
should be conservative, consisting of laxatives
and enemas (Bruder and Meier-Ruge 2007a,
2007b). Most patients have been shown to
respond well to this treatment strategy. However,
if bowel symptoms persist longer than 6 months
despite conservative bowel management, surgical
or interventional treatment options should be
considered (Puri and Gosemann 2012). Internal
sphincter myectomy has been performed by sev-
eral authors with satisfactory results, whereas
others recommend the injection of botulinum
toxin into the anal sphincter (Friedmacher and
Puri 2013). Resection of the affected bowel seg-
ment and pull-through procedure are rarely indi-
cated in infants and children with IND, but in
adolescent or adult patients, this is often the only
successful therapeutic option (Puri and Gosemann
2012). The indication for surgery should not be
determined on the basis of histopathological find-
ings alone; instead, the decision must be based on
the individual patient’s clinical symptoms and
distress.
70.2.6 Outcome
A multidisciplinary team of experienced pediat-
ric surgeons and gastroenterologists is crucial for
the long-term follow-up of patients with IND and
chronic constipation. Authors from Ireland
reported functional outcome in 33 infants and
children with IND (Friedmacher and Puri 2013):
64% had a good response to conservative man-
agement with normal bowel habits and did not
need any surgical intervention. However, 36% of
their patients underwent internal sphincter myec-
70  Variant Hirschsprung’s Disease
tomy after failed conservative treatment. Seven
out of these 12 patients had normal bowel habits
after surgery and 2 were able to stay clean with
regular enemas. Three patients continued having
persistent constipation after myectomy and sub-
sequently required resection of their redundant
and dilated sigmoid colon, which resulted in nor-
mal bowel habits. Other authors have achieved
satisfactory results in 90% of their patients within
6  months of internal sphincter myectomy
(Friedmacher and Puri 2013).
70.3 Intestinal
Ganglioneuromatosis
Intestinal ganglioneuromatosis (GNM) is charac-
terized by a diffuse proliferation of nerve fibers
with marked hyperplasia of submucosal and
myenteric ganglion cells causing the thickening
of the bowel wall (D’Amore et  al. 1991). This
extremely rare but severe neoplastic condition
leads to chronic bowel obstruction and is often
associated with multiple endocrine neoplasia
type 2B (MEN 2B), neurofibromatosis 1 or
Cowden syndrome (de Krijger et al. 1998; Moline
and Eng 2011).
70.3.1 Epidemiology
Although the exact incidence of intestinal GNM
is unknown, it has been reported that the fre-
quently associated MEN 2B syndrome occurs in
approximately 1:4,000,000 live births (Moline
and Eng 2011). Conversely, it can be estimated
that intestinal GNM is present in nearly 90% of
patients with MEN 2B (O’Riordain et al. 1995).
70.3.2 Pathogenesis
The pathogenesis of intestinal GNM involves
complex hyperplasia of peptidergic, cholinergic
and most likely adrenergic nerve fibers and neu-
rons (D’Amore et  al. 1991). Transmural GNM
mainly originates from the myenteric plexus,
while mucosal GNM predominantly affects the
953
submucosal plexus and is often associated with
neurofibromatosis (D’Amore et  al. 1991).
Furthermore, mutation analysis in patients with
MEN 2B identified a de novo germline
Met918Thr mutation in exon 16 of the rear-
ranged during transfection (RET) proto-­oncogene
(Moline and Eng 2011), suggesting a genetic
component to this condition. In addition, a recent
experimental study in mice revealed that the
deletion of the Pten gene on chromosome 10 dis-
rupts the development of the enteric nerve sys-
tem, resulting in a phenotype similar to human
intestinal GNM (Friedmacher and Puri 2013).
70.3.3 Clinical Presentation
The vast majority of patients with intestinal
GNM present with severe chronic constipation
and abdominal distension due to intestinal
obstruction (O’Riordain et al. 1995). Constipation
may also fluctuate with episodes of diarrhea (de
Krijger et al. 1998). The similarity of the gastro-
intestinal symptoms between patients with HD,
IND and MEN 2B-associated intestinal GNM
suggests that these three conditions could be the
result of mutations actually affecting the same
domain of the RET proto-oncogene (Friedmacher
and Puri 2013). Despite the fact that gastrointes-
tinal dysmotility is a common initial presentation
of patients with MEN 2B, the rarity of this syn-
drome often delays the diagnosis. Further find-
ings are mucosal neuromas of the lips and tongue,
as well as medullated corneal nerve fibers, dis-
tinctive facies with enlarged lips and an asthenic
“marfanoid” body habitus (Moline and Eng
2011).
70.3.4 Diagnosis
Intestinal GNM is mainly diagnosed on the basis
of clinical presentation and histological analysis
of full-thickness rectal biopsies, showing mas-
sive proliferation of submucosal and myenteric
plexuses comprising thick nerve trunks with scat-
tered mature neurons, giant ganglia with often
15–40 nerve cells and a high AChE activity (Torre
954
et  al. 2002). Unlike neurofibromatosis, which
occurs more commonly in the small intestine,
intestinal GNM appears to be largely confined to
the colon and rectum. Although AChE immuno-
histochemistry has often been used to show the
typical submucosal and myenteric changes in
intestinal GNM (i.e., increased thickness of nerve
fibers), it can easily be appreciated in standard
hematoxylin and eosin-stained paraffin sections
(Yin et al. 2006). NSE, synaptophysin and S-100
protein immunostaining has also been applied to
evaluate and diagnose intestinal GNM (D’Amore
et  al. 1991). It has been demonstrated that the
submucosal hyperplasia can be extensive, but not
as prominent as that seen in the myenteric plexus
(Yin et al. 2006). As intestinal GNM is frequently
associated with MEN 2B, the diagnosis should
prompt additional molecular, endocrinological
and oncological investigations (Feichter et  al.
2009). In general, a mutational analysis of the
RET proto-oncogene is strongly recommended in
all patients with intestinal GNM and MEN 2B, as
well as their first-degree relatives (de Krijger
et al. 1998).
70.3.5 Management
In patients with MEN 2B-associated intestinal
GNM, surgical resection of the affected bowel
segment is not always necessary. In fact, it has
been shown that in most cases, the gastrointesti-
nal symptoms can be managed with daily laxa-
tives and enemas (Cohen et al. 2002; Smith et al.
1999). However, some patients eventually require
surgery for severe intestinal obstruction or stric-
ture formation. Furthermore, all patients with
intestinal GNM who carry MEN 2B mutations
should undergo a prophylactic total thyroidec-
tomy to prevent the development of medullary
thyroid carcinoma (O’Riordain et al. 1995; Torre
et al. 2002).
70.3.6 Outcome
Early diagnosis and treatment of patients with
MEN 2B-associated intestinal GNM is essential
F. Friedmacher and P. Puri
for long-term survival. Moreover, a yearly fol-
low-­up with monitoring of basal plasma calcito-
nin and carcinoembryonic antigen levels for
possible tumor recurrence is strongly recom-
mended (Torre et al. 2002). Additionally, the con-
tinued surveillance of the adrenal glands with
abdominal ultrasonography and urine analysis of
catecholamine metabolites including metaneph-
rine, normetanephrine, dopamine and vanillyl-
mandelic acid is required as patients with MEN
2B have at least a 50% risk of developing pheo-
chromocytoma (Smith et al. 1999).
70.4 Isolated Hypoganglionosis
Isolated hypoganglionosis (HG) is a rare entity
that has been classified as a hypogenetic type
of intestinal innervation disorder. The clinical
presentation of patients with isolated HG is
similar to those with classical HD, with non-
specific symptoms of severe constipation or
bowel obstruction (Puri and Gosemann 2012).
It has been demonstrated that congenital and
acquired HG are two separate entities with dif-
ferent clinical features and histological find-
ings (Taguchi et al. 2006). At present, there are
only a few cases in the published literature as
isolated HG is one of the rarest subtypes of
intestinal innervation disorders and there
remains controversy regarding its existence as
a distinct isolated histopathological entity
(Martucciello et al. 2005).
70.4.1 Epidemiology
Reports of finding isolated HG in rectal biopsies
are rare, ranging between 0.3% and 6.4%
(Friedmacher and Puri 2013; Montedonico et al.
2011). Since 1978, there have been a total of 92
cases published in the English literature and 32%
of them were diagnosed in the newborn period
(Dingemann and Puri 2010). However, the
median age at diagnosis was 4.8  years, which
was most likely due to the fact that in several
patients, the diagnosis was not made until they
were adolescents.
70  Variant Hirschsprung’s Disease 955

70.4.2 Pathogenesis tal biopsy is required for the definitive diagnosis

of isolated HG (Bruder and Meier-Ruge 2007b;
The pathogenesis and genetic basis of isolated Schäppi et al. 2013). The vast majority of reported
HG is still largely unclear. Although various cases have been analyzed by immunohistochemi-
mutational analyses of the RET gene have been cal staining showing sparse and small myenteric
performed, neither causative missense mutations ganglia, absent or low AChE activity in the lam-
nor neutral substitutions were found (Friedmacher ina propria as well as hypertrophy of muscularis
and Puri 2013). Some cases of isolated HG were mucosae and circular muscle (Friedmacher and
reported to exhibit a deficient expression of Puri 2013). Further histopathological differences
c-Kit-positive ICCs within the myenteric plexus between resected bowel specimens from patients
and the smooth muscle layer (Rolle et al. 2007), with isolated HG and normal bowel tissue have
which may contribute to the observed motility been discovered using AChE immunohistochem-
dysfunction in the hypoganglionic bowel seg- istry (Puri and Gosemann 2012). For instance, a
ment. A lack or reduced expression of NCAM-­ 40% reduction in the number of nerve cells has
positive nerve fibers within the lamina propria been shown, accompanied by a doubled distance
and muscularis mucosae, as well as the circular between ganglia and a three times smaller plexus
and longitudinal muscle layers of patients with area in the hypoganglionic bowel segment. These
isolated HG, has also been described (Friedmacher observations currently form the basis for the his-
and Puri 2013). topathological diagnosis of isolated HG.  It has
also been suggested that the size of the myenteric
plexus may be an indicator of clinical severity.
70.4.3 Clinical Presentation Therefore, various neuronal markers have been
introduced to facilitate the diagnosis of isolated
The most common presenting symptoms of iso- HG. NADPH-d staining has been used to deter-
lated HG are severe chronic constipation com- mine the muscular nitrergic innervation and dif-
bined with intestinal obstruction and enterocolitis, ferentiation of mature from immature ganglia in
thus resembling the clinical features of HD. The patients with isolated HG, demonstrating a
median age at diagnosis is considerably higher in reduced number of positive nerve fibers in the
patients with isolated HG compared to patients muscularis mucosae with absent or sparse sub-
with HD, which is predominately diagnosed dur- mucosal and myenteric ganglion cells
ing the newborn period. Enterocolitis of the new- (Fig.  70.2a–d) (Friedmacher and Puri 2013).
born has been reported to be the most serious and Additionally, c-Kit staining has been employed
potentially life-threatening complication of iso- to investigate the expression of ICCs and thus
lated HG, similarly to HD (Dingemann and Puri intestinal pacemaker activity, which is markedly
2010). decreased or even absent in patients with isolated
HG (Rolle et al. 2007).

70.4.4 Diagnosis
There is an ongoing debate whether isolated HG
represents an extreme form of intestinal dysgan-
glionosis or solely a developmental abnormality
of the enteric nervous system that leads to severe
constipation (Martucciello et  al. 2005). Hence,
the diagnosis of isolated HG remains difficult
and a consensus in diagnostic criteria still needs
to be established. In general, a full-thickness rec-
70.4.5 Management
The treatment of isolated HG is similar to that of
HD.  According to the literature, most patients
undergo resection of the affected bowel segment
with subsequent pull-through procedure
(Dingemann and Puri 2010). However, the man-
agement of isolated HG should always be tai-
lored to the individual patient’s findings.
956 F. Friedmacher and P. Puri

a b

c d

Fig. 70.2  NADPH-d staining in whole mount preparation of a normal myenteric plexus (a, c). Myenteric plexus of a
patient with isolated HG showing markedly reduced number of ganglion cells (b, d)

70.4.6 Outcome also affect the urinary system, recurrent urinary

The postoperative outcome after resection of the
hypoganglionic segment is usually good. Typical
complications of isolated HG are enterocolitis,
chronic constipation, overflow encopresis and the
need for redo pull-through operation due to resid-
ual hypoganglionosis. An overall mortality rate
of 8% has recently been reported with the major-
ity of patients who died being newborns suffering
from severe enterocolitis (Puri and Gosemann
2012). As the dysmotility in isolated HG may
tract infections are a common problem and there-
fore should be monitored actively.
70.5 Immature Ganglia
Immature ganglia (IG) are normally found in rec-
tal biopsies from premature infants presenting
with functional bowel obstruction. Not surpris-
ingly, delayed maturation of ganglion cells in the
submucosal and myenteric plexuses has been
70  Variant Hirschsprung’s Disease
reported to be the most common cause of chronic
constipation during the first year of life (Feichter
et al. 2009).
70.5.1 Epidemiology
Strong epidemiological data on the incidence of
IG is unfortunately lacking. In 1997, 4 (2.8%)
cases of immature ganglion cells were reported in
a cohort of 141 patients with intestinal neuronal
malformations. More recently, 10 (5.6%) cases of
IG were discovered in bowel specimens of 178
patients with variants of HD (Friedmacher and
Puri 2013).
70.5.2 Pathogenesis
A combination of large (i.e., fully mature) and
small (i.e., immature) ganglion cells can be found
at birth (Burki et al. 2011). In the early postnatal
period, ganglion cells in the submucosal plexus
are generally less developed than the ones in the
myenteric plexus (Smith 1968). It has further
been demonstrated that this immaturity is a phys-
iological, age-dependent phenomenon and the
maturation of IG strongly correlates with the age
of the patient (Holschneider et  al. 1994; Smith
a b
Fig. 70.3  Hematoxylin and eosin staining (a) and AChE immunohistochemistry (b) in a patient with IG, showing
immature ganglion cells
957
1968). Strong evidence supporting this theory
has arisen from several animal studies showing
postnatal maturation of the submucosal and
myenteric plexuses (Friedmacher and Puri 2013).
Hence, the finding of IG on rectal biopsy may be
a reliable indicator of transient functional imma-
turity of the bowel (Burki et al. 2011).
70.5.3 Clinical Presentation
Patients with IG usually present with a history of
chronic constipation or functional bowel obstruc-
tion resembling HD.  Further clinical features
may include slow transit peristalsis and insuffi-
cient defecation.
70.5.4 Diagnosis
In general, the diagnosis of IG can be made from
full-thickness rectal biopsies. The ganglion cells
appear very small and have a less significant
nucleus with an inconspicuous nucleolus
(Fig.  70.3a, b) (Smith 1968). However, with
AChE immunohistochemistry, it is often not pos-
sible to distinguish between these small ganglion
cells and their supporting enteric glial cells.
Therefore, NADPH-d and NCAM staining have
958
been suggested as neuronal markers to show the
small ganglion cells more clearly (Puri 1997).
Enzyme histochemistry for succinate and lactate
dehydrogenase is also commonly used to deter-
mine IG, demonstrating an absent or rather weak
positive reaction (Holschneider et al. 1994; Meier-
Ruge and Bruder 2005). In addition, cathepsin D
has been recommended to assess the maturation
of immature ganglion cells in more detail
(Friedmacher and Puri 2013). Another helpful
biomarker to detect IG is the apoptosis regulator
B-cell lymphoma 2 (Friedmacher and Puri 2013),
which clearly differentiates immature small neu-
rons from enteric glial cells and satellite cells.
70.5.5 Management
The management of patients with IG is conserva-
tive with the use of laxatives and enemas
(Friedmacher and Puri 2013).
70.5.6 Outcome
The vast majority of patients with IG can be suc-
cessfully managed with conservative treatment
measures until their ganglion cells are fully
mature (Holschneider et al. 1994). Nevertheless,
it is recommended to repeat the biopsy 12 to
18 months after initial investigation.
70.6 Absence of the Argyrophil
Plexus
The lack of argyrophil cells in the myenteric
plexus, which is also known as the absence of the
argyrophil plexus (AP), is a rare cause of consti-
pation and functional bowel obstruction in infants
and children.
70.6.1 Epidemiology
There is a paucity of published data on the inci-
dence of this very rare condition. In 1997, Prem
Puri found three cases with an absence of the AP
in their series of patients with functional bowel
F. Friedmacher and P. Puri
disorders (Puri 1997). Familial cases in the off-
spring of consanguineous parents and recurrence
in siblings suggest that the absence of the AP
may be inherited in an autosomal-recessive man-
ner (Auricchio et al. 1996; Tanner et al. 1976).
70.6.2 Pathogenesis
There are two distinct subtypes of nerve cells in
the myenteric plexus, which can be distinguished
by their affinity for silver stains: (1) argyrophil
cells and (2) argentaffin cells. Argyrophil cells
normally comprise between 5% and 20% of the
total number of myenteric neurons (Tanner et al.
1976). The processes of these cells along with
extrinsic and parasympathetic fibers form a com-
plex neuronal network within the myenteric
plexus, which is involved in the regulation of gas-
trointestinal peristalsis and transit time. Argyrophil
cells coordinate the activation of argentaffin cells,
which in turn secrete specific neurotransmitters
and ultimately cause contraction or relaxation of
muscle fibers within the bowel wall (Tanner et al.
1976). A distinct time lag has been demonstrated
between the appearance of both cell types, with
argyrophil cells appearing earlier than argentaffin
cells (Singh 1963). It has further been shown that
there is a caudocranial gradient in the differentia-
tion of these neuron cells in the human bowel
(Singh 1963). Therefore, it is suspected that the
disruption of this differentiation process may lead
to an absence of the AP.
70.6.3 Clinical Presentation
The clinical symptoms of patients with the
absence of the AP are highly similar to HD, thus
presenting with severe constipation, moderate
abdominal distension and the lack of peristalsis
(Auricchio et al. 1996).
70.6.4 Diagnosis
The absence of argyrophil cells and their neuronal
processes can only be demonstrated by using sil-
ver impregnation of full-thickness rectal b­ iopsies
70  Variant Hirschsprung’s Disease
a b
Fig. 70.4  Silver staining showing normal AP (a) and absence of argyrophil cells (b) in a patient with absence of the
AP
(Fig. 70.4a, b), whereas conventional hematoxy-
lin and eosin staining, AChE immunohistochem-
istry and histochemistry with other neuronal
markers fail to show this abnormality (Puri 1997).
70.6.5 Management
Conservative management with laxatives and
enemas is sufficient in most patients with absence
of the AP.  However, in some cases, internal
sphincter myectomy or the formation of a colos-
tomy may be required due to persistent constipa-
tion (Puri 1997).
70.6.6 Outcome
Conservative and surgical treatment of patients
with absence of the AP usually results in a satis-
factory outcome (Puri 1997).
70.7 Internal Anal Sphincter
Achalasia
Internal anal sphincter achalasia (IASA) has a
similar clinical presentation to HD, but with the
presence of ganglion cells in rectal biopsies.
959
Previously, IASA was referred to as ultrashort-­
segment HD, which is characterized by an agan-
glionic segment of 1 to 3 cm above the pectinate
line, normal AChE activity in the lamina propria
and increased AChE activity in the muscularis
mucosae (Friedmacher and Puri 2012). Thus, it
has been suggested that IASA is a more accurate
term for this pathological entity, as many patients
with the absence of the rectosphincteric reflex on
anorectal manometry actually showed the pres-
ence of ganglion cells combined with normal
AChE activity in rectal biopsies (Doodnath and
Puri 2009).
70.7.1 Epidemiology
A total number of 395 cases with IASA have
been reported in the literature since 1973
(Friedmacher and Puri 2012). However, the exact
incidence of IASA is unknown.
70.7.2 Pathogenesis
Despite attempts of numerous investigators to
determine the pathophysiological mechanisms of
IASA in more detail, the exact pathogenesis
remains unclear. Age-related changes in the
960 F. Friedmacher and P. Puri

developing intramuscular innervation of the
internal anal sphincter (IAS) most likely form the
basis for the observed motility dysfunction
(Doodnath and Puri 2009). By analyzing
NADPH-d activity (Fig.  70.5a, b), it has been
shown that absent to marked reduction of nitrer-
gic innervation within the IAS of patients with
IASA may be the underlying pathomechanism
leading to spasm or increased tone (Friedmacher
and Puri 2013). Additionally, a defective innerva-
tion of the neuromuscular junction of the IAS
with decreased expression of PGP9.5 and synap-
sin-­1 has been identified (Friedmacher and Puri
2013). Absent to markedly reduced NADPH-d
and NCAM activity in the IAS of patients with
IASA has also been demonstrated (Friedmacher
and Puri 2013). More recently, a reduced number
of c-Kit-positive ICCs have been found in the
IAS of patients with IASA (Rolle et  al. 2007).
The deficiency in nitrergic innervation and ICCs
may explain the impaired IAS relaxation in
patients with IASA.
70.7.3 Clinical Presentation
In most cases, the clinical presentation of IASA
is similar to that of HD. Patients with IASA gen-
erally suffer from chronic and severe constipa-
tion with or without soiling. Approximately one
third of these patients have a history of abdomi-
nal distension and failure of laxative therapy
(Doodnath and Puri 2009).

a Air Inflation Air Inflation

Rectum

Internal Sphincter

External Sphincter

Rectosphincteric Reflex Rectosphincteric Reflex

b
Air Inflation Air Inflation Air Inflation

Rectum

Internal Sphincter

External Sphincter

Fig. 70.5  NADPH-d staining of normal IAS (a). Reduced NADPH-d-positive innervations (b) in a patient with IASA
70  Variant Hirschsprung’s Disease 961

a b

Fig. 70.6  Anorectal manometry showing evidence of the with marked increased rhythmic activity of the IAS (b) in
rectosphincteric reflex in a normal IAS (a). The absence a patient with IASA
of the rectosphincteric reflex on rectal balloon inflation

70.7.4 Diagnosis analysis has indicated that posterior IAS myec-

The diagnosis of IASA is based on clinical symp-
toms in combination with anorectal manometry
findings, showing the absence of the rectosphinc-
teric reflex along with marked increased rhyth-
mic activity (Fig.  70.6a, b). Furthermore, rectal
biopsies demonstrate the presence of ganglion
cells and normal AChE activity, as well as deple-
tion of nitrergic nerves within the IAS (Puri and
Gosemann 2012).
70.7.5 Management
tomy appears to be a more effective treatment
option resulting in better functional outcome
compared to intrasphincteric botulinum toxin
injection (Friedmacher and Puri 2012). The rate
of transient fecal incontinence, non-response and
subsequent surgical procedures was significantly
higher after injection of botulinum toxin, whereas
long-term improvements were significantly more
frequent following IAS myectomy. Interestingly,
no differences were found in the postoperative use
of laxatives or enemas, postoperative soiling and
constipation between both procedures.

Traditionally, posterior IAS myectomy has been 70.8 Megacystis-Microcolon-­

recommended for the treatment of IASA.  More Intestinal Hypoperistalsis
recently, intrasphincteric injection of botulinum Syndrome
toxin has been introduced as a less invasive thera-
peutic alternative (Friedmacher and Puri 2013).Megacystis-microcolon-intestinal hypoperistal-
sis syndrome (MMIHS) is an extremely rare con-
dition and the most severe form of functional
70.7.6 Outcome bowel obstruction in the newborn, characterized
by massive abdominal distension caused by a
The vast majority of patients with IASA have large-dilated, non-obstructed bladder, microco-
regular bowel movements after treatment irre- lon with malrotation and decreased or absent
spective of the therapeutic approach (Puri and intestinal peristalsis (Fig.  70.7a, b) (Puri and
Gosemann 2012). However, a recent meta-­ Gosemann 2012).
962
a b
Fig. 70.7  Voiding cystourethrogram (a) and barium contrast enema (b) showing massively enlarged bladder and
microcolon in a patient with MMIHS
70.8.1 Epidemiology
In the published literature, there has been a total
of 450 reported MMIHS patients (Nakamura
et al. 2019). A male-to-female ratio of 1:2.3 has
been observed, suggesting a distinct female pre-
dominance. In comparison to affected females,
male patients with MMIHS seem to have a
shorter life expectancy. This is most likely due to
increased severity of this condition in males
compared to females (Köhler et  al. 2004).
Furthermore, the occurrence of familial cases in
the offspring of consanguineous parents and
recurrence in siblings indicates that MMIHS
may be inherited in an autosomal-recessive man-
ner (Nakamura et al. 2019).
70.8.2 Pathogenesis
MMIHS was first observed in 1976 in five new-
born girls (Berdon et  al. 1976), who presented
F. Friedmacher and P. Puri
with the above characteristic features. Since
then, various hypotheses have been proposed to
explain the pathogenesis of MMIHS.  Genetic,
myogenic, neurogenic and hormonal etiologies
have been discussed (Puri and Gosemann 2012).
However, most of these theories have been
derived from case reports due to the rarity of this
condition. Thus, the etiology remains poorly
understood. At present, several candidate genes
for MMIHS have been identified. The gene
knockout of the nicotinic acetylcholine receptor
(nAChR) in transgenic mice resulted in a pheno-
type similar to that of human MMIHS
(Friedmacher and Puri 2013). Furthermore, it
has been shown that a lack of expression of the
α3, β2 and β4 subunits of nAChR in small bowel
tissue from patients with MMIHS contributes to
the pathogenesis of this rare syndrome (Puri and
Gosemann 2012). CHRNA3 and CHRNB4 genes,
which both code for the β4 subunit of nAChR,
are additional strong candidates (Friedmacher
and Puri 2013). Published data also points to a
70  Variant Hirschsprung’s Disease
dysfunction within the smooth muscle layer.
Vacuolar changes in the smooth muscle cells
within the bowel and bladder wall have been
reported in patients with MMIHS (Fig. 70.8a, b)
(Puri et  al. 1983). Therefore, it has been sug-
gested that the observed smooth muscle myopa-
thy may be the underlying cause of MMIHS
(Puri et al. 1983; Rolle et al. 2002). The expres-
sion of several contractile and cytoskeleton pro-
teins, such as αSMA, calponin, caldesmin and
desmin, has also been found to be absent or
decreased in the colonic smooth muscle tissue of
patients with MMIHS (Puri and Gosemann
2012). Further support of this theory is derived
from the findings of abnormal synaptophysin
distribution in the circular muscle layer of bowel
and bladder specimens, as well as increased con-
nective tissue and atrophic smooth muscle fibers
(Friedmacher and Puri 2013). Moreover, a
decreased expression of ICCs in the bladder has
been observed (Rolle et al. 2007). In addition to
the smooth muscle dysfunction, other studies
have focused on neuronal differences in
MMIHS.  Previously, immaturity and malfunc-
tion of autonomic nerve endings in the whole
gastrointestinal tract have been demonstrated
(Puri and Gosemann 2012). Axonal dystrophy
and additional defects of the autonomic innerva-
a b
Fig. 70.8  Electron microscopy of smooth muscle cells in
normal bowel (a). Central core degeneration resulting in
empty vacuoles (asterisks) and increased connective tis-
963
tion in the intestine of patients with MMIHS
have also been reported (Friedmacher and Puri
2013). In addition, an intramural inflammatory
process that affects the gastrointestinal and uri-
nary tract has been proposed in the pathogenesis
of MMIHS (Puri and Gosemann 2012).
70.8.3 Clinical Presentation
On prenatal ultrasound or MRI scans, an enlarged
bladder with hydronephrosis, microcolon and
intestinal malrotation can often be demonstrated.
The clinical presentation of MMIHS is similar to
that of other severe neonatal intestinal obstruc-
tion, typically presenting with massive abdomi-
nal distension, which is a consequence of the
largely dilated, non-obstructed bladder with or
without upper urinary tract dilatation. The major-
ity of patients with MMIHS require either vesi-
costomy or catheterization as they are unable to
void spontaneously. A strikingly abnormal detru-
sor muscle has been identified in these patients,
which is most likely the cause for the voiding
dysfunction (Friedmacher and Puri 2012).
Additional common findings are bile-stained
vomiting, absent or decreased bowel sounds and
failure to pass meconium.
sue (triangles) in smooth muscle cells (b) of a patient with
MMIHS
964
70.8.4 Diagnosis
There should be an emphasis on the prenatal
diagnosis of MMIHS to allow for adequate pre-
natal counseling to enable future parents to
make an informed decision regarding the con-
tinuation of the pregnancy, given the poor prog-
nosis of this severe condition. The diagnosis of
MMIHS is usually made on the basis of prenatal
ultrasound findings and characteristic clinical
presentation in the immediate neonatal period.
An enlarged bladder and hydronephrosis can
already be found on fetal ultrasound studies. An
analysis of enzymatic changes in amniotic fluid
in combination with MRI scans can further con-
tribute to the prenatal diagnosis of MMIHS
(Garel et al. 2006). A histological evaluation of
myenteric and submucosal plexuses revealed
normal ganglion cells in 77% of the investigated
bowel specimens from patients with
MMIHS.  The remaining 23% were shown to
have various neuronal abnormalities including
hyper-/hypoganglionosis and immature ganglia
(Friedmacher and Puri 2013). In addition, some
authors found significant anomalies in smooth
muscle cells from bowel and bladder specimens,
such as vacuolar degeneration and thinning of
the longitudinal muscle (Puri et al. 1983; Rolle
et al. 2002).
70.8.5 Management
The management of patients with MMIHS is lim-
ited. A number of prokinetic drugs and gastroin-
testinal hormones have been tried without any
success. Surgical treatment usually has to be per-
formed for malrotation, bowel obstruction and
megacystis to achieve decompression of bowel
and bladder. However, in most cases, these inter-
ventions do not result in any improvement in
enteral food intake, functional bowel obstruction
or bladder function. Consequently, the majority
of patients with MMIHS are maintained on total
parenteral nutrition (TPN), which leads to further
comorbidities, such as catheter sepsis, dyslipid-
emia, TPN-associated liver disease and eventu-
F. Friedmacher and P. Puri
ally chronic liver failure (Gosemann and Puri
2011). Intestinal and multivisceral transplanta-
tion has recently been introduced as a valuable
therapeutic alternative in patients with irrevers-
ible intestinal pathology and failure of TPN. To
date, 12 multivisceral transplantations have been
reported in MMIHS patients (Puri and Gosemann
2012).
70.8.6 Outcome
The survival rate of MMIHS has improved con-
siderably in recent years from 12.6% initially to
55.6%. Between 1977 and 2011, an overall sur-
vival rate of 19.7% was reported with the oldest
survivor being 24 years old. Overwhelming sep-
sis, multiple organ failure and malnutrition have
been shown to be the most frequent causes of
death in MMIHS patients (Gosemann and Puri
2011). A 3-year survival rate of 50% with all sur-
vivors tolerating enteral feeding and showing
adequate gastric emptying has recently been
reported (Loinaz et  al. 2005). Improvements in
outcome have mainly been due to more special-
ized care, innovations in parenteral nutrition and
the introduction of multivisceral transplantation.
70.9 Conclusion
Conditions that clinically resemble HD despite
the presence of ganglion cells on rectal biopsies,
can be diagnosed by providing an adequate
biopsy specimen and employing a variety of his-
tological staining techniques. The two most com-
mon disorders in variant HD are IND and
IASA. The majority of patients with IND can be
successfully managed with conservative treat-
ment measures or IAS myectomy if required.
Thus, pull-through operation is rarely indicated
in the management of IND. IASA, which is char-
acterized by a reduction of nitrergic innervation
within the IAS, can be diagnosed by anorectal
manometry and successfully treated by either
posterior IAS myectomy or alternatively, an
intrasphincteric injection of botulinum toxin.
70  Variant Hirschsprung’s Disease 965

References Garel C, Dreux S, Philippe-Chomette P, Vuillard E, Oury

J-F, Muller F (2006) Contribution of fetal magnetic res-
onance imaging and amniotic fluid digestive enzyme
Auricchio A, Brancolini V, Casari G, Milla PJ, Smith VV,
assays to the evaluation of gastrointestinal tract abnor-
Devoto M et al (1996) The locus for a novel syndromic
malities. Ultrasound Obstet Gynecol 28:282–291
form of neuronal intestinal pseudoobstruction maps to
Gosemann JH, Puri P (2011) Megacystis microcolon
Xq28. Am J Hum Genet 58:743–748
intestinal hypoperistalsis syndrome: systematic review
Berdon WE, Baker DH, Blanc WA, Gay B, Santulli TV,
of outcome. Pediatr Surg Int 27:1041–1046
Donovan C (1976) Megacystis-microcolon-intestinal
Granero Cendón R, Millán López A, Moya Jiménez MJ,
hypoperistalsis syndrome–a new cause of intestinal
López Alonso M, De Agustín Asensio JC (2007)
obstruction in the newborn: report of radiologic find-
Intestinal neuronal dysplasia: association with diges-
ings in five newborn girls. AJR Am J Roentgenol
tive malformations. Cir Pediatr 20:166–168
126:957–964
Holschneider AM, Meier-Ruge W, Ure BM (1994)
Bruder E, Meier-Ruge WA (2007a) Intestinal neuronal
Hirschsprung’s disease and allied disorders–a review.
dysplasia type B: how do we understand it today?
Eur J Pediatr Surg 4:260–266
Pathologe 28:137–142
Kobayashi H, Hirakawa H, Puri P (1995) What are the
Bruder E, Meier-Ruge WA (2007b) Hypoganglionosis as
diagnostic criteria for intestinal neuronal dysplasia?
a cause of chronic constipation. Pathologe 28:131–136
Pediatr Surg Int 10:459–464
Burki T, Kiho L, Scheimberg I, Phelps S, Misra D, Ward
Kobayashi H, Hirakawa H, Surana R, O’Briain DS, Puri
H et al (2011) Neonatal functional intestinal obstruc-
P (1995) Intestinal neuronal dysplasia is a possible
tion and the presence of severely immature ganglion
cause of persistent bowel symptoms after pull-through
cells on rectal biopsy: 6 year experience. Pediatr Surg
operation for Hirschsprung’s disease. J Pediatr Surg
Int 27:487–490
30:253–257
Coerdt W, Michel JS, Rippin G, Kletzki S, Gerein V,
Köhler M, Pease PW, Upadhyay V (2004) Megacystis-­
Müntefering H et al (2004) Quantitative morphomet-
microcolon-­ intestinal hypoperistalsis syndrome
ric analysis of the submucous plexus in age-related
(MMIHS) in siblings: case report and review of the
control groups. Virchows Arch 444:239–246
literature. Eur J Pediatr Surg 14:362–367
Cohen MS, Phay JE, Albinson C, DeBenedetti MK,
Koletzko S, Jesch I, Faus-Kebetaler T, Briner J, Meier-­
Skinner MA, Lairmore TC et al (2002) Gastrointestinal
Ruge W, Müntefering H et al (1999) Rectal biopsy for
manifestations of multiple endocrine neoplasia type 2.
diagnosis of intestinal neuronal dysplasia in children:
Ann Surg 235:648–654
a prospective multicentre study on interobserver varia-
D’Amore ESG, Manivel JC, Pettinato G, Niehans GA,
tion and clinical outcome. Gut 44:853–861
Snover DC (1991) Intestinal ganglioneuromatosis:
Loinaz C, Rodriguez MM, Kato T, Mittal N, Romaguera
mucosal and transmural types. A clinicopathologi-
RL, Bruce JH et al (2005) Intestinal and multivisceral
cal and immunohistochemical study of 6 cases. Hum
transplantation in children with severe gastrointestinal
Pathol 22:276–286
dysmotility. J Pediatr Surg 40:1598–1604
de Krijger RR, Brooks A, van der Harst E, Hofstra RM,
Martucciello G, Pini Prato A, Puri P, Holschneider AM,
Bruining HA, Molenaar JC et al (1998) Constipation
Meier-Ruge W, Jasonni V et al (2005) Controversies
as the presenting symptom in de novo multiple endo-
concerning diagnostic guidelines for anomalies of
crine neoplasia type 2B. Pediatrics 102:405–408
the enteric nervous system: a report from the fourth
Dingemann J, Puri P (2010) Isolated hypogangliono-
international symposium on Hirschsprung’s dis-
sis: systematic review of a rare intestinal innervation
ease and related neurocristopathies. J Pediatr Surg
defect. Pediatr Surg Int 26:1111–1115
40:1527–1531
Doodnath R, Puri P (2009) Internal anal sphincter achala-
Martucciello G, Torre M, Pini Prato A, Lerone M, Campus
sia. Semin Pediatr Surg 18:246–248
R, Leggio S et al (2002) Associated anomalies in intes-
Fadda B, Maier WA, Meier-Ruge W, Schärli A, Daum
tinal neuronal dysplasia. J Pediatr Surg 37:219–223
R (1983) Neuronal intestinal dysplasia. Critical
Meier-Ruge W (1971) Casuistic of colon disorder with
10-years’ analysis of clinical and biopsy diagnosis. Z
symptoms of Hirschsprung’s disease. Verh Dtsch Ges
Kinderchir 38:305–311
Pathol 55:506–510
Feichter S, Meier-Ruge WA, Bruder E (2009) The histo-
Meier-Ruge WA, Ammann K, Bruder E, Holschneider
pathology of gastrointestinal motility disorders in chil-
AM, Schärli AF, Schmittenbecher PP et  al (2004)
dren. Semin Pediatr Surg 18:206–211
Updated results on intestinal neuronal dysplasia (IND
Friedmacher F, Puri P (2012) Comparison of posterior
B). Eur J Pediatr Surg 14:384–391
internal anal sphincter myectomy and intrasphincteric
Meier-Ruge WA, Brönnimann PB, Gambazzi F, Schmid
botulinum toxin injection for treatment of internal anal
PC, Schmidt CP, Stoss F (1995) Histopathological cri-
sphincter achalasia: a meta-analysis. Pediatr Surg Int
teria for intestinal neuronal dysplasia of the submuco-
28:765–771
sal plexus (type B). Virchows Arch 426:549–556
Friedmacher F, Puri P (2013) Classification and diag-
Meier-Ruge WA, Bruder E (2005) Pathology of chronic
nostic criteria of variants of Hirschsprung’s disease.
constipation in pediatric and adult coloproctology.
Pediatr Surg Int 29:855–872
Pathobiology 72:1–102
966
Moline J, Eng C (2011) Multiple endocrine neoplasia type
2: an overview. Genet Med 13:755–764
Montedonico S, Acevedo S, Fadda B (2002) Clinical
aspects of intestinal neuronal dysplasia. J Pediatr Surg
37:1772–1774
Montedonico S, Cáceres P, Muñoz N, Yáñez H, Ramírez
R, Fadda B (2011) Histochemical staining for intes-
tinal dysganglionosis: over 30 years experience with
more than 1,500 biopsies. Pediatr Surg Int 27:479–486
Moore SW, Kaschula ROC, Cywes S (1993) Familial and
genetic-aspects of neuronal intestinal dysplasia and
Hirschsprung’s disease. Pediatr Surg Int 8:406–409
Nakamura H, O’Donnell AM, Puri P (2019)
Consanguinity and its relevance for the incidence of
megacystis microcolon intestinal hypoperistalsis syn-
drome (MMIHS): systematic review. Pediatr Surg Int
35:175–180
O’Riordain DS, O’Brien T, Crotty TB, Gharib H, Grant
CS, van Heerden JA (1995) Multiple endocrine neo-
plasia type 2B: more than an endocrine disorder.
Surgery 118:936–942
Puri P (1997) Variant Hirschsprung’s disease. J Pediatr
Surg 32:149–157
Puri P, Gosemann JH (2012) Variants of Hirschsprung dis-
ease. Semin Pediatr Surg 21:310–318
Puri P, Lake BD, Gorman F, O’Donnell B, Nixon HH
(1983) Megacystis-microcolon-intestinal hypoperi-
stalsis syndrome: a visceral myopathy. J Pediatr Surg
18:64–69
Puri P, Lake BD, Nixon HH, Mishalany H, Claireaux
AE (1977) Neuronal colonic dysplasia: an unusual
association of Hirschsprung’s disease. J Pediatr Surg
12:681–685
Rolle U, O’Briain S, Pearl RH, Puri P (2002) Megacystis-­
microcolon-­intestinal hypoperistalsis syndrome: evi-
dence of intestinal myopathy. Pediatr Surg Int 18:2–5
F. Friedmacher and P. Puri
Rolle U, Piaseczna-Piotrowska A, Puri P (2007)
Interstitial cells of Cajal in the normal gut and in intes-
tinal motility disorders of childhood. Pediatr Surg Int
23:1139–1152
Schäppi MG, Staiano A, Milla PJ, Smith VV, Dias JA,
Heuschkel R et  al (2013) A practical guide for the
diagnosis of primary enteric nervous system disorders.
J Pediatr Gastroenterol Nutr 57:677–686
Singh I (1963) The prenatal development of enterochro-
maffin cells in the human gastro-intestinal tract. J Anat
97:377–387
Smith B (1968) Pre- and postnatal development of the
ganglion cells of the rectum and its surgical implica-
tions. J Pediatr Surg 3:386–391
Smith VV, Eng C, Milla PJ (1999) Intestinal ganglioneu-
romatosis and multiple endocrine neoplasia type 2B:
implications for treatment. Gut 45:143–146
Taguchi T, Masumoto K, Ieiri S, Nakatsuji T, Akiyoshi J
(2006) New classification of hypoganglionosis: con-
genital and acquired hypoganglionosis. J Pediatr Surg
41:2046–2051
Tanner MS, Smith B, Lloyd JK (1976) Functional
intestinal obstruction due to deficiency of argyro-
phil neurones in the myenteric plexus: familial
syndrome presenting with short small bowel, mal-
rotation, and pyloric hypertrophy. Arch Dis Child
51:837–841
Torre M, Martucciello G, Ceccherini I, Lerone M,
Aicardi M, Gambini C et  al (2002) Diagnostic and
therapeutic approach to multiple endocrine neopla-
sia type 2B in pediatric patients. Pediatr Surg Int
18:378–383
Yin M, King SK, Hutson JM, Chow CW (2006) Multiple
endocrine neoplasia type 2B diagnosed on suction rec-
tal biopsy in infancy: a report of 2 cases. Pediatr Dev
Pathol 9:56–60
 Anorectal Anomalies
Alberto Peña, Andrea Bischoff,
and Luis De la Torre
71.1 Introduction
Advances in Science and Technology are chang-
ing the way we diagnose and treat our patients.
This is also true for patients with anorectal mal-
formations. The changes are mostly beneficial
and allow us to have an optimistic opinion about
the future of the treatment of congenital
malformations.
In this chapter, we will be discussing the pre-
natal diagnosis, neonatal approach, importance
of the early determination of the future functional
prognosis, colostomy or no colostomy, diagnosis,
associated defects, incidence, medical and surgi-
cal management, complications, long-term fol-
low-­up, bowel management for the treatment of
fecal incontinence and surgical treatment of rec-
tal prolapse, importance of the multidisciplinary
centers and the importance of “transition of care.”
The quality of life of patients born with anorectal
malformations depends on the specific type of mal-
formation, the characteristics of the sacrum (sacral
A. Peña · A. Bischoff
International Center for Colorectal and Urogenital
Care, Children’s Hospital Colorado, University of
Colorado School of Medicine, Aurora, CO, USA
e-mail: Alberto.pena@childrenscolorado.org;
Andrea.bischoff@childrenscolorado.org
L. De la Torre (*)
International Center for Colorectal and Urogenital
Care, Children’s Hospital Colorado,
Aurora, CO, USA
e-mail: Luis.delatorre@childrenscolorado.org
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_71
71
ratio), the presence or absence of tethered cord, and
(very importantly) the quality of the surgical tech-
nique employed to repair the malformation.
71.2 Historical Overview
The history of the surgical treatment of anorectal
malformations is a representative sample of the
history and evolution of medicine in general.
Anorectal malformations have been known
through centuries, simply because cannot go
unnoticed. Since the “early times,” the mothers
have been aware of this defect. Sorcerers, witches,
and barbers offered their “services,” to try to cure
this condition (Pena and Bischoff 2015). The
results were dismal, with occasional successes in
cases that we now recognize as “very benign.”
Several important landmarks can be
mentioned:
–– 1465: Cerrahiyei Ilhaniye—perhaps the first
illustration describing an anorectal procedure
(Pena and Bischoff 2015).
–– 1606: Guilhelmus Fabricius Hildanus—
described a case of recto-bladder neck fistula.
Patient died (Pena and Bischoff 2015).
–– 1710: Littre proposed, but did not perform a
colostomy (Pena and Bischoff 2015).
–– 1683–1731: Frederik Ruysch described the
spontaneous rupture of the anal membrane.
The baby died (Pena and Bischoff 2015).
967
968
–– The practice of a perineal incision followed by
dilatations was a method of choice until the
later part of the nineteenth century.
–– 1753: M. Louis, from Paris, described the case
of a female patient, who suffered from what
we now call “Posterior cloaca” (Louis 1754).
–– 1787: Benjamin Bell, from Edinburgh, devel-
oped a good experience in the treatment of
these defects, including two successful
­operations finding the rectum inside the abdo-
men (Bell 1787).
–– 1783: Antoine Dubois performed the first
inguinal colostomy in a patient with anorectal
malformation. The patient died (Allan 1797).
–– 1835: Amussat, in Paris, is considered the first
surgeon who sutured the rectal wall to the skin
(anoplasty). The baby survived. He also cre-
ated a classification (Amussat 1835).
–– 1860: Bodenhamer proposed a classification
(Bodenhamer 1860).
–– 1866: Chassaignac proposed the opening of a
colostomy prior to the repair (Chassaignac
1856).
–– 1897: Rudolph Matas presented 22 conclu-
sions related with the management of these
defects (Matas 1897).
–– 1915: Brenner presented 61 cases (Brenner
1915).
–– 1930: Rice and Wangesteen described the
“invertogram” to determine the location of the
blind end of the location (Wangesteen and
Rice 1930).
–– 1934: Ladd and Gross published a very com-
prehensive series of cases, with a mortality of
26% (Ladd and Gross 1934).
–– 1948: Rhoads presented the first neonatal sur-
vivor of an abdominal-perineal pull-through
for imperforate anus (Rhoads et al. 1948).
–– 1953: Douglas Stephens, from Melbourne,
Australia, published his experience in the dis-
section of cadavers with anorectal malforma-
tions (Stephens 1953). He concluded that the
key part of the sphincter mechanism to
achieve bowel control was the “puborectalis
sling.” That concept started a whole era, dur-
ing which several surgical procedures were
proposed, all of them with the specific pur-
pose of preserving the “puborectalis sling”
(Kiesewetter 1967; Rehbein 1967). There
A. Peña et al.
was, however, a problem, nobody (except for
Dr. Stephens in autopsies) had seen the
famous puborectalis sling. A very extensive
and sophisticated classification was also cre-
ated by Dr. Stephens, known as the
“Melbourne Classification.” However,
nobody actually used that classification.
In August 10th 1980, in Mexico City, out of
curiosity, the senior author of this chapter per-
formed the first PSARP (Posterior Sagittal Ano
Recto Plasty), looking for the famous “puborec-
talis sling.” There was no such structure. Instead,
he found striated muscles distributed in a com-
pletely different fashion. In addition, fortunately,
we were looking for the first time at the “intrinsic
anatomy” of the malformation, particularly at the
rather complex way in which the rectum joins the
urogenital tract. By learning this, we become
aware of the potential damage that these patients
may suffer when the operations were done blindly
(DeVries and Pena 1982).
In 1982, a cloaca malformation was com-
pletely repaired using the posterior sagittal
approach for the first time (Pena 1989).
In 1988, Doctor Douglas Stephens, and Dr.
Durham Smith, his close collaborator, organized
a special meeting inviting prominent surgeons,
particularly those interested in the management
of these defects. The meeting was in Wingspread,
Wisconsin, USA. Some concessions were made
concerning the classification and the new ana-
tomic concepts (Stephens and Smith 1988). In
2005, another meeting was organized by Dr.
Holschneider in Krickenbeck, Germany
(Holschneider and Hutson 2006). At that meet-
ing, a new, rational, practical classification was
adopted.
In 2000, Dr. Georgeson repaired an anorectal
malformation via laparoscopy (Georgeson et al.
2000). With that operation, Dr. Georgeson inau-
gurated the era of the minimally invasive
approach for the treatment of anorectal malfor-
mations. That approach had a very significant
benign impact in the treatment of those malfor-
mations that traditionally required a laparotomy.
Many more surgeons are currently trying the
laparoscopic approach for the repair of recto-­
urethral-­prostatic fistula. More enthusiastic sur-
71  Anorectal Anomalies
geons are also trying the laparoscopic approach
for recto-urethral-bulbar fistula and even in cases
of recto-vestibular fistula, with very controversial
results.
71.3 Incidence
Anorectal malformations are reported to occur in
2.0–2.5 per 10,000 live births. However, there are
several reports indicating that there are differ-
ences between different countries and races, as
well as social and economic levels (Chavez et al.
1988; Ameh and Chirdan 2000; Garne et  al.
2002; Leck and Lancashire 1995). These malfor-
mations seem to be more common in developing
countries; however, we do not know if that
reflects poor statistical methodology, poor cen-
sus, excessive population growth, or poor prena-
tal care.
71.4 Etiopathogenesis
Significant progress has occurred in the knowl-
edge of the clinical genetics, cytogenetics, and
molecular genetics of anorectal malformations.
However, our knowledge is still limited, since
these malformations belong to a category known
as “multifactorial,” which means they have a
great phenotypic variability and a multigenic ori-
gin. The so-called Currarino Triad (anorectal
malformation, pre-sacral mass, and hemi sacrum)
is the only association for which a single gene
has been identified in humans (Currarino et  al.
1981). In addition, this association has a well
demonstrated tendency to run in families.
71.5 Pathophysiology
and Pathology
Anorectal malformations present in the form of a
spectrum, including anatomic defects of different
degrees, which result in functional deficiencies
with different degrees of severity. In addition, it
is very commonly associated with defects in
other parts of the body, such as the urogenital
tract (50%). However, if one considers the fact
969
that the majority of cases have a communication
between the rectum and the urogenital tract, and
also the bladder innervation deficits consecutive
to the sacrum abnormalities, and frequently the
nerve damage provoked during the surgical repair
of it results that the majority of cases of anorectal
malformations have urologic concerns of differ-
ent degrees. Other important associations involve
the spinal cord (25%), the esophagus (8%), the
duodenum (4%), and the heart (30%) (although
only one third of the cardiac defects require
treatment).
There is a clear correlation between the loca-
tion of the blind rectum and (or) the location of
its connection with the urogenital tract and the
severity of the anatomic abnormalities. When the
blind end of the rectum is located very high in the
pelvis, or else connected to the bladder neck in
males, most of those cases have poor sphincters,
sacral deformities and an association with other
malformations reaching up to 90%. Whereas, in
cases of perineal fistula, the sphincters are almost
normal, the sacrum (with a few exceptions) is
normal, and less than 10% of those cases have
associated malformations.
All cases of anorectal malformations will have
functional disturbances of different degrees, the
most common being CONSTIPATION, followed
by fecal incontinence, both in different degrees of
severity. Interestingly, the more benign the mal-
formation (lower location of the rectum) the bet-
ter the continence, but the worse the
constipation.
It is not difficult to understand why these
patients may suffer from fecal, and sometimes
urinary incontinence. Under normal circum-
stances, the innervation of the rectum and the
sphincters originates in the sacral vertebrae.
There is a clear correlation between the degree of
sacral deficiency and the severity of the fecal and
urinary incontinence. That is the reason why it is
not acceptable to evaluate bowel and urinary con-
trol and compare results of different techniques,
without describing the characteristics of the
sacrum. In addition, these patients are subjected
to a surgical procedure that includes the dissec-
tion and mobilization of the rectum. When the
rectum is mobilized, some nerves are divided
with obvious potential, functional implications.
970
Fig. 71.1  Illustration of the Sacral ratio
In addition, sometimes the surgical repair is done
in a technically incorrect manner and the damage
to the nerves is even worse. All this contributes to
explain the functional deficits of these patients.
Patients born without a sacrum will suffer
from total fecal and urinary incontinence. Patients
born with a normal sacrum have a “good chance”
of having bowel and urinary control, although it
is not guaranteed. Traditionally, the sacrum has
been evaluated by the number of vertebrae.
However, sometimes the patients have five verte-
brae and yet the sacrum is very short, with obvi-
ous functional consequences. This highlighted
the need to design a more objective way to evalu-
ate the sacrum, which resulted in the creation of
the sacral ratio (Fig. 71.1). In the experience of
the authors, a sacral ratio less than 0.4 always
results in a lack of bowel control.
The external appearance of the perineum pro-
vides good information that can be correlated, to
a degree, with functional prognosis. The presence
of a prominent midline groove, and a noticeable
anal dimple, is usually associated with the pres-
ence of a good sphincter mechanism. The longer
the distance between the lowest palpable portion
of the sacrum and the anal dimple, the less chance
of having bowel control.
Bladder innervation can be deficient since
birth, due to a deficient sacrum and (or) the pres-
ence of tethered cord. In addition, the bladder
innervation can be damaged during the repair of
A. Peña et al.
an anorectal malformation. The surgeons must
keep in mind that under normal circumstances,
the bladder and corpora innervation come from
the sacrum and run lateral, but very close to the
rectum. That is the reason why it is highly recom-
mended to remain as close as possible to the rec-
tal wall during the mobilization of the rectum. It
is not uncommon to hear the parents of these
children describe that, prior to the repair, their
baby had normal intermittent voiding episodes,
as well as penile erections. After the operation,
the parents observe only dribbling of urine and a
lack of erections, all of which is evidence of
intraoperative nerve damage.
71.6 Classification
The classification presented here is based on
therapeutic and prognostic implications. This
means that each malformation requires specific
technical maneuvers to be repaired and each one
has a different specific functional prognosis.
Males 
–– Recto-Bladder Neck fistula (Fig. 71.2).
–– Recto-Urethral-Prostatic fistula (Fig. 71.3).
–– Recto-Urethral-Bulbar fistula (Fig. 71.4).
–– Perineal fistula (Fig. 71.5).
–– Anorectal malformation without fistula
(Fig. 71.6).
71  Anorectal Anomalies 971

Fig. 71.4  Recto-urethral bulbar fistula

Fig. 71.2  Recto-bladder-neck fistula

Fig. 71.3  Recto-urethral prostatic fistula Fig. 71.5  Recto-perineal fistula in a male
972 A. Peña et al.

Fig. 71.8  Cloaca >3 cm. common channel

Fig. 71.6  Anorectal malformation without fistula

Fig. 71.9  Vestibular fistula

Fig. 71.7  Cloaca with <3 cm. common channel

Females 
–– Cloaca <3 cm. common channel (Fig. 71.7).
–– Cloaca >3 cm. common channel (Fig. 71.8).
–– Recto-vaginal fistula (almost non-existent)
Malformation.
–– Recto-vestibular fistula (Fig. 71.9).
–– Recto-perineal fistula (Fig. 71.10).
–– Anorectal malformation without a fistula.

Complex Malformations 
–– Cloacal exstrophy.
–– Covered cloacal exstrophy.
Fig. 71.10  Perineal fistula in a female
71  Anorectal Anomalies 973

Neonatal Management  Diagnosis and These questions must be answered in the

Differential Diagnosis.
Typically, the pediatric surgeon is called to see
a newborn baby who was born without an anus.
There are two important questions that the sur-
geon must try to answer during the first 24 h of
life:
order in which they are presented here. There
is a tendency for surgeons to concentrate
their attention on the second question; this is
dangerous, since it puts the baby at risk of
suffering serious problems due to an unde-
tected serious cardiac or renal anomaly.

(a) Does the baby have an associated malforma- An important recommendation is to not take the
tion that threatens his life? baby to the operating room during the first 24 h of
(b) Does the baby need a colostomy or a primary life. That period must be used to rule out important
repair of his malformation? associated malformations (Figs. 71.11 and 71.12).

Fig. 71.11 Newborn Newborn Male-Anorectal Malformation

male anorectal
malformation algorithm Perineal inspection
Spine Sacrum
Kidney U/S Spinal U/S
20-24 hrs Urinalysis Cardiac echo
R/O esophageal atresia
Re-evaluation and cross-table lateral film

Perineal Rectal gas below coccyx Rectal gas above coccyx

fistula No associated defects
Associated defects
Abnormal sacrum
Flat bottom
Consider PSARP
with or without
Anoplasty
colostomy
Colostomy

Fig. 71.12 Newborn Female Newborn-Anorectal Malformation

female anorectal
- Sacrum - Kidney & Abdominal U/S
malformation algorithm
R/O serious, potentially
- Esophagus - Spinal U/S (Tethered
lethal associated defects
- Lumbar spine - Cardiac Echo

PERINEAL INSPECTION

Single perineal No
orifice visible (<10%)

Cloaca Perinea Vestibular 24 hrs

fistula fistula
Urol
Evaluation Cross table
R/O lateral X-Ray
hydrocolpu
COLOSTOMY
or Primary
COLOSTOMY Rectum High
repair *
Drain below rectum
hydrocolpos COLOSTOMY coccyx
Urinary Anoplasty or Primary
diversion or repair *
COLOSTOMY
(if necessary)
974
Fig. 71.13  Flat bottom in a newborn
The baby must remain fasting, a nasogastric
tube must be inserted, and the following studies
must be ordered:
–– Kidney ultrasound.
–– Sacro lumbar ultrasound to rule out the pres-
ence of tethered cord.
–– X-ray films of the sacrum in AP and lateral
positions.
–– Echo cardiogram.
–– “Babygram”—an X-ray film of the entire baby,
which will allow to rule out the presence of
hemivertebrae, an image consistent with esoph-
ageal atresia, as well as duodenal atresia (dou-
ble bubble) and the degree of bowel dilatation.
–– Instructions must be given to the nurses to
watch carefully for the presence of meconium
coming from the female genitalia, from the
penis, or from a very small perineal orifice in
the male.
–– The inspection of the perineum with a good
index of suspicion gives extremely valuable
information for the decision-making
process.
A. Peña et al.
Fig. 71.14  Meconium in the perineum, coming from a
perineal fistula
Males 
–– A flat bottom is very frequently associated
with a very high location of the rectum and
poor functional prognosis (Fig. 71.13).
–– Meconium in the perineum, coming from a
small perineal orifice, (Fig.  71.14) allows to
make the diagnosis of a perineal fistula. This
is a malformation with excellent prognosis,
provided the sacrum is normal. A primary
repair is recommended within the first 72 h of
life without a colostomy.
–– The presence of a sub-epithelial fistula filled
with meconium or mucus, (Fig. 71.15) is asso-
ciated with a perineal fistula.
–– A prominent intergluteal cleft and anal dim-
ple, with or without meconium in the urine
(Fig. 71.16), probably means that the baby has
a recto-urethral bulbar fistula or an anorectal
malformation without a fistula. A colostomy is
recommended.
Females 
–– An orifice located in between the genitalia and
the anal dimple (Fig.  71.17) allows to make
the diagnosis of a perineal fistula, with the
same implications described for males.
–– An orifice located inside the vestibule of the
female (Fig. 71.18) makes the diagnosis of
“vestibular fistula,” the most common mal-
formation in females. Some surgeons rec-
ommend a colostomy, others perform a
primary repair in the first 72  h of life and
71  Anorectal Anomalies 975

Fig. 71.17  Recto perineal fistula in a female

Fig. 71.15  Perineal fistula with subepithelial tract along

scrotal raphe

Fig. 71.18  Recto vestibular fistula in female

Fig. 71.16  Perineal groove and anal dimple
finally, others offer a delayed primary repair,
provided the baby can empty the colon
spontaneously.
–– A single perineal orifice makes the diagnosis
of a cloaca (Fig.  71.19). The surgeon must
know that the patient has a high chance of suf-
fering from serious urologic problems. It is
mandatory, at that stage, to order a pelvic
ultrasound to rule out the presence of a hydro-
colpos, which must be drained while the
colostomy opening.
976 A. Peña et al.

Fig. 71.20  Ideal colostomy

Fig. 71.19  A single perineal orifice in a patient with
Cloaca
Twenty-four hours later, the results of these
studies and observations will provide enough
information to the surgeon to make the right
decision concerning the management of the
baby. In addition, he will have enough informa-
tion to predict the future functional prognosis of
the baby in an accurate manner, which is some-
thing that the parents will appreciate very much.
through. An ideal colostomy should be done at
the descending colon. No sigmoid colostomies.
–– The proximal stoma must be opened at the
first mobile portion of the descending colon,
to avoid prolapse.
–– The distal stoma (mucous fistula) must be
tapered, creating a small orifice, to avoid
prolapse.
Figure 71.20 Ideal Colostomy.

71.7 Colostomy
71.8 Primary Repair without
Colostomy in newborn babies with anorectal mal- a Colostomy
formations is a delicate operation with significant
morbidity (Pena et  al. 2006). The technique is Primary repair without a colostomy is recom-
radically different to the colostomies performed in mended in patients with perineal fistula. Ideally,
adult patients. The following principles must be the procedure must be performed during the first
followed when opening colostomies in newborn 72 h, before the baby is fed, to take advantage of
babies with anorectal malformations: the fact that the meconium is sterile. The babies
can be fed immediately after surgery.
–– The colostomy must be completely diverting (no Primary delayed repair without a colostomy is
loop colostomies). The reason being that more also indicated in cases of perineal fistula.
than 80% of the patients have a communication However, since the patient has been fed and is
of the colon with the urogenital tract. A partially passing real stool, we recommend cleaning the
diverting colostomy (loop-type) represents a colon preoperatively with Golytely*, inserting a
potential risk of fecal contamination of the uro- central line and keeping the patient fasting for
genital tract. The stomas must be separated 1 week, receiving parenteral nutrition.
enough to allow the placement of a bag covering In cases of vestibular fistula, the decision of
only the proximal, functional stoma. colostomy or primary repair must take into con-
–– The distal colon must be irrigated and left sideration the experience of the surgeon, the gen-
completely clean at the time of the operation. eral condition of the baby and the environmental,
–– Enough length of distal colon must be guaran- and technical support. The opening of a colos-
teed, to avoid interference with the future pull-­ tomy is always the safest way to prevent cata-
71  Anorectal Anomalies
strophic complications, which may change the
final functional prognosis of the baby.
The primary repair of all other malformations
is contraindicated, since the location of the distal
blind end of the rectum or the location of the fis-
tula is unknown, and therefore the operation,
under those circumstances, becomes an “explora-
tion,” frequently ending in catastrophic results.
The repair of anorectal malformations should
always be a planned procedure, with a full knowl-
edge of the anatomic characteristics of the mal-
formation to be repaired.
The repair of all other anorectal malforma-
tions must be done with a previous protective
colostomy.
In cases of cloacas, the surgeon must rule out
the presence of a hydrocolpos, prior to the open-
ing of a colostomy. If present, the hydrocolpos
must be drained on a permanent basis, during the
same anesthesia. Not draining the hydrocolpos
exposes the patient to suffer infection (pyocol-
pos), perforation, and urinary tract obstruction
due to the extrinsic compression of the uretero-­
vesical junction, by a very distended vagina.
71.9 High Pressure Distal
Colostogram (Males)
One or two weeks after the colostomy has been
opened, the baby is ready to undergo the most
important and valuable diagnostic study, which is
the “High Pressure Distal Colostogram” (Gross
et  al. 1991). There is no other diagnostic study
(MRI and CT Scan included) which provides the
precise information of this fluoroscopic proce-
dure. The term “High Pressure” has been inten-
tionally added to emphasize the need to apply
significant hydrostatic pressure, in order to over-
come the muscle tone of the striated muscle fun-
nel surrounding the most distal portion of the
rectum, proximal to the fistula.
The study is performed under fluoroscopy and
consists of an injection of water-soluble contrast
material through the mucous fistula of the colos-
tomy, to reach the most distal portion of the bowel
and demonstrate the location of the fistula.
977
The study starts with the baby in supine
position; the first piece of information that the
surgeon needs is to determine how much colon
is available distal to the colostomy, to perform
a pull-through. One of the most common errors
in the making of a colostomy is to open it too
distal, leaving a very short, distal piece of
colon, not long enough to do the pull-through.
Without this information, the surgeon may be
unhappily surprised during the main repair,
having to abort the operation, or to perform a
series of complex maneuvers or procedures to
complete the repair.
The baby is then turned into a perfect lateral
position to continue the injection of the contrast.
Unexperienced radiologists inject the contrast
material by gravity and do not exert hydrostatic
pressure; therefore, the contrast material will stop
running at the level of the pubo-coccygeal line,
which represents the upper limit of the funnel-­
like striated muscle cuff. The radiologists then
have the false impression that they are looking at
the lowest part of the rectum and consequently,
make a wrong diagnosis of “High Imperforate
Anus without fistula,” which is almost a non-­
existent malformation. Applying hydrostatic
pressure by hand (with a syringe) will allow one
to overcome the muscle tone of the muscle funnel
and demonstrate the most distal portion of the
rectum. The pressure must be maintained until
the fistula is demonstrated and the contrast
reaches the urinary tract. The injection must con-
tinue filling the bladder and provoking a micturi-
tion. At that point, pictures are taken showing the
precise location of the fistula, the urethra, and the
bladder. Figures  71.21, 71.22, and 71.23 show
colostogram images of different malformations.
It is important to notice that in cases of recto-­
bladder neck fistulae, the contrast material flows
easily into the bladder, with minimal hydrostatic
pressure. The reason to explain this is that in that
malformation, the rectum is located above the
funnel-like muscle structure. Using the old termi-
nology, this malformation is the only real “supra-
levator” malformation and fortunately, it only
happens in about 10% of all anorectal malforma-
tions in males.
978 A. Peña et al.

Fig. 71.23  Colostogram showing a recto urethral-bulbar

fistula

In order to be able to exert enough hydrostatic

Fig. 71.21  Colostogram showing a recto-bladder neck
fistula
pressure, it is essential to use a Foley catheter for
the injection, the balloon must be inflated and
pulled back, in order for the balloon to act as a
plug to avoid the leakage of contrast during the
injection under pressure.

71.10 High Pressure Distal

Colostogram (Females)

This study has limited value in females because,

in cases of perineal and vestibular fistulae, the
precise diagnosis is made by simple perineal
inspection. The study is useful in cases of anorec-
tal malformation without a fistula.
In cloacas, the study is done in conjunction
with the injection of contrast material through the
single perineal orifice and injection through the
catheter draining the hydrocolpos when present.
More recently, a rotational CT scan, done simul-
taneously with the colostogram, demonstrated its
usefulness to clarify the anatomy of complex clo-
acas (Fig. 71.24).

Fig. 71.22  Colostogram showing a recto-urethral-pros-

tatic fistula
71  Anorectal Anomalies 979

sagittal approach. This operation consists of plac-

ing the patient in prone position, with the pelvis
elevated. An incision is made exactly in the mid-
line, in between both buttocks. The entire sphinc-
ter mechanism is identified and divided in the
midline. The posterior rectal wall is identified
and opened in the midline. Under direct vision,
the fistula is identified, and the rectum is meticu-
lously separated from the urethra. This part of the
operation represents the most delicate part of the
procedure, this is the time when more serious
complications may occur. It is important to keep
in mind that between the anterior rectal wall and
the posterior wall of the urethra, there is no plane
of dissection. These two structures share a com-
Fig. 71.24  3-D CT scan cloacogram mon wall. A meticulous, delicate technique
allows the surgeon to make two walls out of one,
71.11 Repair of the Malformation without damaging important structures of the
male anatomy, such as the vas deferens, seminal
After the colostomy has been opened, it is expected vesicles, and nerve erigentes.
for the baby to grow and develop normally. This Once the rectum has been successfully sepa-
can be demonstrated observing the baby for a rated from the urethra, it is necessary to perform
period of 1 or 2 months, after which the baby can a circumferential dissection of the rectum, in
be subjected to the main repair of the malforma- order to gain enough length to create a non-tense
tion. If the surgeon is not familiar with performing anastomosis with the skin.
these procedures at such an early stage, he can An electric stimulator helps to find the best
wait until he feels comfortable to do it. possible location of the new anus, at the center of
Most malformations can be repaired through a the sphincter.
limited or full posterior sagittal approach (PSARP). In cases of recto-urethral-bulbar fistula, the
Ten percent of the malformations in males suf- rectum needs minimal mobilization to reach the
fer from a bladder neck fistula, which cannot be perineum, yet, the common wall between the ure-
reached from below and therefore, they require thra and rectum is longer and wider. On the other
an abdominal approach. Patients suffering from hand, the common wall between the rectum and
this defect are also ideal candidates for a laparo- urethra is much shorter in cases of recto-urethral-­
scopic approach, combined with a limited peri- prostatic fistula, but the rectum requires much
neal incision. more mobilization.
Laparoscopy enthusiasts with experience can The same approach can be used in cases of
also successfully repair cases of recto-urethral-­ recto-vestibular fistula. However, the incision is
prostatic fistulae. Overenthusiastic laparoscopists much shorter. On the other hand, the separation
insist that they can repair cases of recto-urethral-­ of the rectum from the vagina also requires a
bulbar fistula laparoscopically, which has resulted meticulous, delicate technique, since the rectum
in an increased number of urethral complications, and the vagina also share a long and wide com-
including posterior urethral diverticulums (piece mon wall.
of rectum left attached to the urethra). The diver- Patients with perineal fistulae can be repaired
ticulum provokes serious symptoms and requires using what could be called a “minimal” posterior
a secondary complex operation. sagittal incision, long enough to be sure that the
Cases of recto-urethral fistula (bulbar and rectum has been separated from the urethra in
prostatic) can be repaired safely with a posterior males and from the vagina in females.
980 A. Peña et al.

71.12 Cloaca Repair Size of Dilators According to Patient’s Age

Newborn 12 Hegar
The technique used to repair this malformation 4 months 13 Hegar
is called PSARVUP (Posterior Sagittal Ano 8 months 14 Hegar
12 months 15 Hegar
Recto Vagino Urethro Plasty) and includes a
Older 16–17 Hegar
posterior sagittal incision, followed by the sep-
aration and mobilization of the rectum from the
urogenital sinus. Once the rectum has been sep- Dilatations must not produce pain. Painful,
arated and mobilized, a maneuver called “total forceful dilatations reflect an evolving anal stric-
urogenital mobilization” is done. This consists ture, consecutive to ischemia and (or) tension at
of diving the suspensory ligaments that anchor the anal anastomosis.
the urethra and vagina to the pubis. By doing There are surgeons who believe that anal dila-
that, in more than 50% of the cases, the urethra tations are not necessary; that is true when the
and vagina reach a normal position. This is pos- anoplasties are done in a technically deficient
sible when the length of the common channel is manner, creating huge anal openings, which look
shorter than 3 cm. With the help of an electrical like colostomies. A technically correct anoplasty,
stimulator, the rectum is then placed in the best placing the anus at the center of the sphincter, in
possible location. patients with sphincter mechanism, allows one to
In cases with a common channel longer than see a closed anus at the end of the procedure,
3 cm, the surgeon can anticipate that it will not because of the sphincter. Not doing dilatations in
be possible to complete the repair via posterior those kinds of cases may result in a superficial,
sagittal and that it will be necessary to open the ring-like, anal stricture.
abdomen. The trans-abdominal separation of
Results (after the age of three)
the two structures that constitute the urogenital
Voluntary Bowel Movements: (patients with normal
sinus (urethra and vagina) represents one of the sacrum)
most serious technical challenges in the repair Bladder neck fistula 20%
of anorectal and urogenital malformations. Recto-prostatic fistula 66%
Both structures (vagina and urethra) share an Recto-urethral-bulbar fistula 78%
extensive common wall. In addition, the ure- No fistula (no down syndrome) 85%
ters pass through that common wall. The sepa- Vestibular fistula 90%
Perineal fistula (both sexes) 100
ration of these structures is easier in younger
Cloacas 70%
babies and that is the reason it should not be
postponed to be done later in life. The repair of
cloacas, with common channels longer than Need for Urinary Catheterization in Cloacas (normal
3 cm, must be done in specialized centers with sacrum)
a high volume of cases and by experienced Common channel <3 cm 20%
surgeons. Common channel >3 cm 80%
The colostomy closure is usually done about
2 months after the repair and after the anal dilata-
tion protocol has been completed. 71.13 Complications
Anal dilatations are started 2 weeks after the
operation. In the clinic, the surgeon must instruct Anal stricture —due to ischemia of the distal rec-
the parents how to do it, beginning with a number tum or lack of post-op anal dilatations.
6–8 Hegar dilator (6–8 mm diameter). The par- Dehiscence—due to ischemia and (or) exces-
ents learn to do the dilatations twice a day, before sive tension at the anal anastomosis.
meals and move up to the next the sized dilator Recurrent, acquired, or persistent fistula—due
every week, until reaching the size that is ade- to technical deficient technique.
quate for the patient’s age.
71  Anorectal Anomalies 981

71.14 Prolapse layer. The dissection continues up to 8  cm

This is not considered a complication, but rather
a common consequence. It can produce bleeding,
pain, excessive mucus production, ulceration,
and interference with daily activities.
(Fig. 71.27). The next step is a maneuver aimed to
achieve tension-free anastomosis after resection
of the dissected mucosa. This maneuver involves

Surgical Repair of Prolapse (Belizon et  al.

2005; Brisighelli et  al. 2014)  The “transanal
rectal mucosectomy and muscular plication for
rectal prolapse” (TRMMP) (De La Torre et  al.
2020) is a novel operation to repair rectal pro-
lapse. The patient is placed in the prone position,
with the pelvis elevated. The hooks of the Lone-­
Star® retractor are positioned on the perianal
skin, 5 mm peripheral to the mucocutaneous junc-
tion (previous anoplasty). Multiple traction 5–0
silk sutures are placed at the rectal mucosa cir-
cumferentially, 5 mm proximal to the mucocuta-
neous junction. A circumferential incision is made
with a fine needle tip cautery, distal to the traction
sutures, including only the mucosa (Figs.  71.25
Fig. 71.26  Rectal mucosectomy identifying two planes:
and 71.26), creating a submucosal plane of dis- (A) The highly vascular mucosa and (B) the circular fibers
section, separating the mucosa from the muscular of the muscular cuff. (C) Mucocutaneous line

Fig. 71.27  Coronal view at the end of the dissection of

Fig. 71.25 Partial thickness circumferential incision
using fine electrocautery. (A) Traction sutures are placed
in the mucosa. (B) The mucocutaneous line from the pre-
vious anoplasty. (C) Skin. Note the lack of a pectinate line
in the distal rectum
the rectal mucosa. The length of the dissection is 8  cm.
(MC) mucocutaneous line. (A) Proximal mucosa. (B)
Distal mucosa located where the circumferential incision
began. The distance between the distal mucosa (B) and the
mucocutaneous line (MC) is 5 mm
982 A. Peña et al.

plication of the muscular wall; 8–12 symmetric

longitudinal non-absorbable stitches are used
(Fig. 71.28). When these sutures are tied, the mus-
cular wall is plicated (Fig. 71.29). As a result, the
proximal mucosa and distal mucosa are placed
very close to each other (Fig.  71.30). The dis-
sected mucosa is removed, and a mucosa–mucosa
anastomosis is performed, with 5–0 interrupted
long-term absorbable sutures (Fig.  71.31). The
bulk of the plicated muscular wall pulls the anas-
tomosis and the mucocutaneous line, producing a
unique positive esthetic effect, and the anus
appears closed (Fig. 71.32). Patients with a colos-
tomy begin a regular diet on the same day and are
discharged the next day.

71.15 Conclusion Fig. 71.29 The longitudinal imbricating sutures are

placed and partially tied. Note the partial plication of the
Anorectal malformations are represented by many muscular cuff; the proximal mucosa (A) and the distal
congenital defects, shown in the form of a spec- mucosa (B) are getting closer
trum. The management of these defects require
different surgical maneuvers, which must be
implemented by dedicated, experienced surgeons.

Fig. 71.30 The longitudinal imbricating sutures are

Fig. 71.28 Coronal view showing the longitudinal
stitches in the muscular cuff. (MC) mucocutaneous line.
(A) Proximal mucosa. (B) Distal mucosa
completely tied. Plication is complete. The proximal
mucosa (A) reaches the distal mucosa (B). The proximal
(A) and distal mucosa (B) are together. (MC) mucocuta-
neous line. The dotted line indicates the site of the muco-
sectomy and anastomosis
71  Anorectal Anomalies
Fig. 71.31  The mucosa was resected on the dotted line
illustrated in Fig. 71.30. The anastomosis (mucosa–
mucosa) is completed. Note the mucocutaneous line
(MC) from the original anoplasty remains intact
Fig. 71.32  Clinical picture. Before and after transanal
rectal mucosectomy with muscular plication. The bulk of
the plicated muscular wall pulls the anastomosis and the
References
Allan R (1797) Rapport sur les observations et reflexions
de Dumas, relative aux imperforations de l’anus. In:
Recueil Periodique de la Societe de Medicine de parís,
No. XIII, pp 123–131
Ameh EA, Chirdan LB (2000) Neonatal intestinal obstruc-
tion in Zaria, Nigeria. East Afr Med J 77:510–513
983
mucocutaneous line, producing a unique esthetic effect,
and the anus appears closed
Amussat JZ (1835) Histoire d’ume operation d’anus prac-
tique avec success par un nouvean procede. Gaz Ed
Paris 3:753–758
Belizon A, Levitt M, Shoshany G, Rodriguez G, Peña
A (2005) Rectal prolapse following posterior sagit-
tal anorectoplasty for anorectal malformations. J
Pediatr Surg 40:192–196. https://doi.org/10.1016/j.
jpedsurg.2004.09.035
Bell B (1787) A system of surgery, 3rd edn. Bell and
Bradfute, Edinburgh, p 375
984
Bodenhamer WH (1860) A practical treatise on the aetiol-
ogy, pathology and treatment of the congenital mal-
formations of the rectum and anus. Wood, New York
Brenner EC (1915) Congenital defects of the anus and
rectum. Surg Gynecol Obstet 20:579–588
Brisighelli G, Di Cesare A, Morandi A, Paraboschi I,
Canazza L, Consonni D et al (2014) Classification and
management of rectal prolapse after anorectoplasty for
anorectal malformations. Pediatr Surg Int 30:783–789.
https://doi.org/10.1007/s00383-­014-­3533-­7
Chassaignac M (1856) Presentation de Malades. Bulletin
de la societe de chirurgie de Paris, p 410
Chavez GE et al (1988) Leading major congenital mal-
formations among minority groups in the United
States, 1981-1986. MMWR CDC Surveil Summ
37:17–24
Currarino G, Coln D, Votteler T (1981) Triad of anorectal,
sacral, and presacral anomalies. AJR Am J Roentgenol
137:395–398
De La Torre L, Zornoza M, Peña A, Bischoff A, Ketzer
J, Ruiz A (2020) Transanal rectal mucosectomy and
muscular plication: a new technique for rectal prolapse
in patients with an anorectal malformation. J Pediatr
Surg 55(11):2531–2535. https://doi.org/10.1016/j.
jpedsurg.2020.05.025
DeVries PA, Pena A (1982) Posterior sagittal anorecto-
plasty. J Pediatr Surg 17(5):638–643
Garne E et  al (2002) Gastrointestinal malformations in
Funen county, Denmark- epidemiology, associated
malformations, surgery and mortality. Eur J Pediatr
Surg 12:101–106
Georgeson KE, Inge TH, Albanese CT (2000)
Laparoscopically assisted anorectal pull-through for
high imperforate anus- a new technique. J Pediatr Surg
35(6):927–930; discussion 930–931
Gross GW, Wolfson PJ, Pena A (1991) Augmented pres-
sure colostogram in imperforate anus with fistula.
Pediatr Radiol 21(18):560–562
Holschneider AM, Hutson JM (2006) Anorectal malfor-
mations in children. Embryology, diagnosis, surgical
treatment, follow-up. Springer, Berlin, Heidelberg,
New York
A. Peña et al.
Kiesewetter W (1967) Imperforate anus the rationale and
technic of the sacroabdominal operation. J Pediatr
Surg 2(2):106–110
Ladd WE, Gross RE (1934) Congenital malformations
of anus and rectum: report of 162 cases. Am J Surg
23(1):167–183
Leck I, Lancashire RJ (1995) Birth prevalence of mal-
formations in members of different ethnic groups
and in the offspring of mating between them, in
Birmingham, England. J Epidemiol Community
Health 49:171–179
Louis (1754) Thesis Anatomicae Chirurgicae. Paris, zit.
bei Bushe (57)
Matas RD (1897) The surgical treatment of congenital
ano-rectal imperforation considered in the light of
modern operative procedures. Trans Am Surg Ass
15:87–93
Pena A (1989) The surgical management of persistent
cloaca: results in 54 patients treated with a posterior
sagittal approach. J Pediatr Surg 24(6):590–598
Pena A, Bischoff A (2015) Surgical treatment of colorec-
tal problems in children. Chapter 1. Springer, pp 1–14
Pena A, Migotto-Krieger M, Levitt MA (2006) Colostomy
in anorectal malformations: a procedure with seri-
ous but preventable complications. J Pediatr Surg
41(4):748–756
Rehbein F (1967) Imperforate anus: experiences with
abdominal-perineal and abdominal-sacro-perineal
pull-through procedures. J Pediatr Surg 2(2):99–105
Rhoads JE, Pipes RL, Randall JP (1948) A simultaneous
abdominal and perineal approach in operations for
imperforate anus with atresia of the rectum and recto-
sigmoid. Ann Surg 127(3):552–556
Stephens FD (1953) Congenital imperforate rectum,
recto-urethral and recto-vaginal fistulae. Aust N Z J
Surg 22(3):161–172
Stephens D, Smith D (1988) Anorectal malformations in
children: update 1988. Birth Defects Orig Artic Ser
24(4):1–604
Wangesteen OH, Rice CO (1930) Imperforate anus: a
method of determining the surgical approach. Ann
Surg 92(1):77–81
 Appendicitis
Markus Almström and Tomas Wester
72.1 Introduction
Acute appendicitis is the most common surgical
emergency in children over 2 years of age. A con-
siderable effort has been made to develop and
improve diagnostic workup, treatment, and out-
come; nevertheless, acute appendicitis still con-
stitutes a significant challenge among pediatric
surgical emergencies.
72.2 Historical Overview
Most probably, the ancient Egyptians were
already aware of the vermiform appendix, as sep-
arate jars for the “worm of the bowel” were found
in graves. Otherwise, the appendix is not men-
tioned in early history. The first written descrip-
tions did not appear until the renaissance, when
both Leonardo da Vinci (1492) and Andreas
Vesalius (1543) noted the presence of the organ.
The first famous appendectomy was per-
formed by Claudius Amyand at St George’s
Hospital in 1735, in a case of appendicitis occur-
ring in an inguinal hernia in an 11-year-old boy—
thus, the first appendectomy was performed
scrotally.
M. Almström · T. Wester (*)
Department of Pediatric Surgery, Astrid Lindgren
Children’s Hospital, Karolinska University Hospital,
Stockholm, Sweden
e-mail: tomas.wester@regionstockholm.se
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_72
72
Still, during the eighteenth and nineteenth
centuries, surgery was not widely available, and
surgical management of lower right quadrant
emergencies was mainly restricted to incising
and draining abscesses.
The modern era of surgical management of
acute appendicitis did not start until the first
appendectomy for the treatment of typhlitis by
Robert Lawson Tait in Edinburgh 1880. In 1886,
Reginald Fitz coined the anatomically more cor-
rect term appendicitis, and in 1889, Charles
McBurney presented a series of appendectomies,
including the first appendectomy for non-­
perforated appendicitis.
The related morbidity was significantly
reduced after the introduction of sulfonamide
(1935) and penicillin (1943) for the treatment of
infectious complications of appendicitis and
appendectomy.
Minimally invasive surgery for appendicitis
was introduced after Kurt Semm’s first in 1980
(Semm 1983) and was repeated by Benno Ure in
children in 1982 (Ure et al. 1992).
The first randomized controlled trial on non-
operative treatment with antibiotics for the treat-
ment of acute appendicitis was performed in
adults in 1995 (Eriksson and Granström 1995)
and in children in 2014 (Svensson et al. 2015).
985
986
72.3 Incidence
The lifetime risk of acute appendicitis has been
estimated to be 7–8%. Reduced incidence rates
were reported both from Europe and North
America, in children and adults, over a long
period of time. Contemporary incidence rates
of appendicitis in children reaches 100 cases
per 100,000 person years. Rates strongly relate
to age, with a peak incidence in adolescents,
and higher rates noted in boys compared to
girls. Furthermore, in children, the incidence
rates and trends differ between non-perforated
and p­erforated appendicitis, with a more
marked reduction for non-perforated disease
indicating a possible difference in etiology
(Livingston et al. 2007; Kang et al. 2003).
For unknown reasons, incidence rates also
vary by geographic region, ethnicity, sex, and
season.
In children, both perforation (20–25%) and
subsequent postoperative complications are more
commonly seen. Negative appendectomies, pre-
viously being common and accepted, rarely
exceed 1–4% in modern series. In children,
appendectomy-related mortality is rare, with case
fatality rates close to 1/10,000 appendectomies
(Almström et al. 2018).
72.4 Etiopathogenesis
The appendix is a reservoir of normal intesti-
nal flora, which has been suggested to con-
tribute to re-colonization of the bowel after a
gastrointestinal tract infection. The appendix
has also been shown to produce mesenchymal
stem cells, potentially involved in bowel
repair throughout life. Furthermore, the
appendix contains abundant gut-associated
lymph tissue (Spencer et al. 1985).
Traditionally, it was assumed that the inflam-
mation process in acute appendicitis started
with an obstruction of the appendix lumen.
However, the appendix is often not obstruc-
tive, and other factors certainly contribute to
the etiology.
M. Almström and T. Wester
72.5 Pathology
Suppurative or phlegmonous appendicitis is char-
acterized by transmural inflammation with neutro-
philic infiltration of the mucosa, submucosa, and
muscularis propria, along with an acutely inflamed
and often ulcerated mucosa. Edema, fibrinopuru-
lent serositis, and micro-abscesses of the appendix
wall may be seen. In gangrenous appendicitis,
transmural inflammation and infiltration of neutro-
phils is accompanied by necrosis of the appendix
wall and extensive mucosal ulceration (Carr 2000).
The classification of appendicitis in the clinical
setting is not strictly defined, and there is not a gen-
eral consensus among surgeons or between surgeons
and pathologists, on the definitions of complex or
noncomplex appendicitis (Bliss et al. 2010). A cor-
rect diagnosis of complex appendicitis is highly cor-
related to increased rates of postoperative
complications and prolonged hospital stays. In
attempts to improve concordance, stricter definitions
of perforated appendicitis have been introduced to
better identify patients at risk of postoperative com-
plications, emphasizing the need for consensus defi-
nitions of appendicitis grades (Ponsky et al. 2009).
Administrative healthcare registers use diagnosis
according to the International Classification of
Disease (ICD) coding system. Several versions of
ICD have been released, and appendicitis classifica-
tion has differed somewhat over the years, introduc-
ing possible bias in register-­based studies.
72.6 Diagnosis
Despite evolving diagnostic support including
novel laboratory markers, modern imaging tech-
nology, and scoring systems appendicitis, diagno-
sis may still be challenging. The traditional
concept of appendicitis as an inevitably progres-
sive disease towards gangrene and perforation
does not hold its ground an all cases, as indicated
by the evolving acceptance of evidence for sponta-
neously resolving appendicitis (Andersson 2007).
Originally, the differentiation of appendicitis
from other causes of acute abdominal pain was
based on patient history and clinical examination
72 Appendicitis 987

alone. Diagnostic accuracy has been markedly
improved with increased accessibility of modern
diagnostic tools, reducing both delayed diagnosis
and negative appendectomies.
Early clinical signs of appendicitis are often
vague and nonspecific, with malaise and perium-
bilical pain followed by vomiting and low-grade
fever. With increased inflammation pain, tender-
ness and guarding migrates to the right lower quad-
rant, and the general condition worsens with nausea
and fatigue. Right lower quadrant tenderness and
guarding are typical for appendicitis. Gentle pres-
sure in the left lower quadrant that elicits pain in the
right lower quadrant increases the likelihood of
appendicitis (Bhangu et al. 2015). Atypical presen-
tation, without classical signs and symptoms, is
relatively common in children. In cases where the
appendix is located in the pelvis, pain and tender-
ness may be more nonspecific, but diarrhea may
occur. It is often very difficult to identify a palpable
mass in children with appendix mass or abscess.
Additional laboratory investigations and
radiologic investigations increase the diagnostic
accuracy but must be interpreted with caution.
Body temperature has no diagnostic significance
as a single diagnostic test, but in combination and
with repeated measures, the discriminating power
may increase.
Interleukine-6 levels increase early in acute
appendicitis and correlate with the degree of
inflammation, but the test has not proved to
improve the diagnostic precision in a clinical set-
ting. Recently, low plasma-sodium level upon
hospital admission was shown to strongly corre-
late to advanced disease and perforation (Giannis
et al. 2020), but results remain to be repeated
before full conclusions can be drawn.
72.8 Imaging Techniques
Computed tomography (CT) and ultrasonogra-
phy (US) have been used and refined in the diag-
nostic workup in suspected appendicitis since the
early 1980s (Puylaert 1986). The best diagnostic
discrimination in children has been achieved
using a combined protocol with ultrasonography
first, with further investigation by CT in unclear
cases (Kaiser et al. 2004; Doria et al. 2006). This
selection of cases minimizes the number of chil-
dren exposed for ionizing radiation. The use of
low-dose CT protocols may further reduce harm-
ful radiation without reducing diagnostic accu-
racy (Figs. 72.1 and 72.2) (Kim et al. 2012).

72.7 Laboratory Tests

White blood cells (WBC) are usually elevated in

acute appendicitis. However, a positive test alone
is a highly nonspecific marker of inflammation,
and the power to discriminate appendicitis from
non-appendicitis is low. Also, C-reactive protein
(CRP) is a poor discriminator of appendicitis from
its differential diagnoses, yielding better perfor-
mance discriminating perforated appendicitis.
There are several studies on the combination of
traditional and/or novel biochemical markers, with
diverting results depending on selected combina-
tions, methodology, and study settings, and there is
no established standard combination (Yu et al. 2013).
In attempts to improve diagnostic accuracy, Fig. 72.1  Acute appendicitis (A) with peri-appendicular
novel biomarkers have been proposed. abscess (B), ultrasonography
988
Fig. 72.2  Acute appendicitis (A), computed
tomography
With increased availability, magnetic reso-
nance imaging (MRI) has become an interesting
alternative to CT. In children and adults, US+MRI
protocols have displayed comparable discrimi-
nating performance to US+CT protocols, exclud-
ing ionizing radiation (Moore et al. 2016).
Previous concerns that an intensified use of
imaging would identify an increased number of
mild appendicitis cases that under other circum-
stances would not have been diagnosed and
treated have not been confirmed, as proven by the
persistent negative incidence rate trends for acute
appendicitis.
72.9 Scoring Systems
In an attempt to increase the diagnostic accuracy
in acute appendicitis, Alvarado included eight
predictive factors in a first scoring system for the
diagnosis of acute appendicitis (Alvarado 1986).
In a systematic review, a low Alvarado score had
an excellent sensitivity of ruling out appendicitis,
in children and adults (Ohle et al. 2011). However,
the specificity for “ruling in” appendicitis at
higher scores was poor. Furthermore, later stud-
ies have failed to reproduce the high sensitivity
initially presented. Samuel constructed a pediat-
ric appendicitis score (PAS) using eight variables
selected by logistic regression of clinical and
M. Almström and T. Wester
investigative parameters in acute appendicitis
(Samuel 2002). The appendicitis inflammatory
response (AIR) score (Andersson and Andersson
2008), including eight parameters, displayed
similar sensitivity and specificity compared to
the Alvarado score while reducing the number of
patients in the intermediate or equivocal group.
In comparison to a senior surgeon assessment,
the Alvarado and AIR scores performed similar
discriminating capacities for overall appendicitis,
while the AIR score outperformed the Alvarado
score and the senior surgeon in positive predic-
tive value and specificity.
In children, the AIR score outperformed the
Alvarado and PAS scores in a recent retrospec-
tive study (Macco et al. 2016).
72.10 Differential Diagnosis
Possible differential diagnoses of acute appendi-
citis in children mainly include benign causes of
lower abdominal pain but also conditions that
require urgent attention or cautious nonurgent
investigation. The most common differential
diagnoses are listed in Table 72.1.
Table 72.1 Common differential diagnoses to acute
appendicitis in children
Pneumonia, upper respiratory infections
Gastroenteritis
Constipation
Mesenterial lymphadenitis
Meckel’s diverticulitis
Intussusception
Small bowel obstruction, ileus
Incarcerated inguinal hernia
Gallstone disease and biliary tract malformations
including cholecystitis and cholestasis
Pancreatitis
Inflammatory bowel disease (IBD)
Urinary tract infection and pyelonephritis
Hydronephrosis
Testicular conditions (torsion and epididymitis)
Ovarian conditions (torsion, cyst rupture)
Salpingitis
Henoch-Schönlein purpura and other vasculitis
72 Appendicitis
72.11 Management
The first line of treatment in suspected appendici-
tis is the restoration of homeostasis, which should
already be initiated during diagnostic workup at
the emergency department. An intravenous line is
established for rehydration. For instance, Ringer’s
solution, corresponding to 5% of the child’s body
weight, is given in 4  hours. The child should
remain fasting. Individualized pain relief (acet-
aminophen, morphine) should be administered in
all cases. There is no evidence that early, bal-
anced pain relief delays or obscures the
diagnosis.
In the presence of peritonitis or suspected per-
foration at diagnosis, antibiotic treatment should
be started preoperatively. Perioperative antibiotic
prophylaxis reduces the rate of
­appendectomy-­associated infectious complica-
tions and should be administered.
72.11.1 Surgical Management
The gold standard treatment for acute appendici-
tis is the surgical removal of the inflamed vermi-
form appendix (Rentea et al. 2017). The operation
may be performed by open surgery or minimally
invasive surgery. Numerous studies and trials
have compared laparoscopic to open surgery in
appendicitis. In a meta-analysis comparing lapa-
roscopic to open appendectomy (Aziz et  al.
2006), postoperative complication rates were
comparable, except for a reduced risk of wound
infection after laparoscopic operation. A subse-
quent meta-analysis of 26 studies including
123,000 children, laparoscopic operation was
superior for all outcome measures, except for
postoperative abscess rates which were compa-
rable, and the authors strongly recommended the
use of laparoscopy over open surgery for appen-
dicitis (Esposito et al. 2012). In a recent Cochrane
meta-analysis (Jaschinski et  al. 2018), laparo-
scopic appendectomy in children was associated
with reduced rates of wound infection and shorter
hospital stay. Currently, most appendectomies
are performed laparoscopically (Fig. 72.3).
989
Fig. 72.3 Phlegmonous appendicitis, laparoscopic
operation
Laparoscopic appendectomy is preferably
performed in a three-port setting, allowing the
use of a 5 mm telescope and a pair of 5 mm lapa-
roscopic instruments. A larger umbilical port
(12 mm) may be used to accommodate additional
use of a stapling devise and a specimen retrieval
bag. As the appendix is identified, the mesoap-
pendix and the appendicular vessels are divided
by diathermy or stapling, and the base of the
appendix is closed and divided, using a stapling
device or endoloops. The use of a laparoscopic
retrieval bag to bring the specimen through the
abdominal wall reduces the risk for postoperative
wound infection. Irrigation of the abdomen has
not been shown to reduce the risk of postopera-
tive abscess formation. A single-port laparo-
scopic technique may be feasible but has not
gained widespread popularity (Zhang et  al.
2015).
In phlegmonous simple appendicitis, antibiot-
ics may be stopped, and the child can be dis-
charged when mobilized and feeding, most often
the following day, although some centers have
reported discharge the same day.
In gangrenous or complex appendicitis, antibi-
otics are continued 1–2 days postoperatively, with
oral intake and mobilization started early. In chil-
dren with perforated appendicitis, broad-­spectrum
antibiotics covering gram-negative bacteria is
usually administered for approximately 5  days
postoperatively, although the optimal time of
treatment has yet to be determined. In perforated
appendicitis, oral intake and mobilization may
also be started as early as accepted by the child,
990
but it is more common that feeding is difficult to
tolerate the first few days due to paralysis.
In perforated appendicitis, there has been con-
troversy about the value of intraoperative irriga-
tions as well as drains. There is currently no
evidence that irrigations have an impact on the
course of perforated appendicitis. There is also no
evidence to support the placement of drains dur-
ing the operation to reduce the complication rate.
72.11.2 Nonoperative Treatment
Nonoperative treatment for acute appendicitis is
not a novel concept. Coldrey treated 471 patients
conservatively without appendectomy with low
mortality and morbidity in 1959 (Coldrey 1959).
In the modern era, Eriksson and Granström per-
formed the first randomized controlled trial
(RCT) on antibiotic treatment versus surgery for
acute appendicitis, finding nonoperative treat-
ment feasible but associated with a high risk of
recurrence during the first year (Eriksson and
Granström 1995). Vons et al. randomized adults
with appendicitis to antibiotic treatment or sur-
gery, concluding that antibiotic treatment was
inferior to appendectomy in noncomplicated
appendicitis (Vons et  al. 2011). In children,
Svensson et al. performed the first pilot RCT trial
randomizing 50 children to either antibiotic treat-
ment or appendectomy, showing that antibiotic
treatment was feasible short-term and at 5 years
of follow-up (Svensson et al. 2015; Patkova et al.
2020). In a metanalysis, nonoperative treatment
has been found to be safe and efficient (Maita
et  al. 2020). Full-scale statistically powered
RCTs are ongoing, and, until otherwise stated,
appendectomy should remain the standard treat-
ment for acute appendicitis.
In recent studies, an initial 1–2 day in-hospital
intravenous antibiotic treatment is usually fol-
lowed by discharge and oral antibiotics for a total
of 10 days of treatment.
72.11.3 Appendix Mass
Appendix mass, or primary appendicitis abscess,
is best treated conservatively by antibiotics
M. Almström and T. Wester
(Fugazzola et al. 2019). Initial treatment includes
fluid therapy and restoration of homeostasis and
suitable analgesics. In children, most cases can
be treated successfully without drainage, which
should only be considered in cases of treatment
failure. Children with an abscess larger than
approximately 4 cm, who do not respond to anti-
biotics, benefit from a percutaneous or transrectal
drain depending on the location of the abscess.
The need for interval appendectomy after an
episode of appendix mass is under debate.
Despite evidence for favorable outcome with a
continued conservative and expectant strategy
(Hall et al. 2017), most pediatric surgeons prefer
to perform an interval appendectomy after the
recovery from appendicitis mass.
72.11.4 Impact of Surgical Delay
The impact of delaying the curative operation with
appendectomy in acute appendicitis has been
debated. Studies supporting both the safety and
increased morbidity following surgical delay in
children and adults have been presented, many
biased by imperfect study design. In adults, sev-
eral publications indicate that at least short surgi-
cal delay does not increase the rate of perforated
appendicitis (Cameron et al. 2018). Results were
repeated in a British multicenter cohort study sup-
plemented by a meta-analysis, indicating no
increased risk for perforation with short surgical
delay. Recent findings also support the safety of
surgical delay in children (Almström et al. 2017),
and at least limited surgical delay may be consid-
ered safe, especially considering the possibilities
for improving preoperative optimization.
72.12 Conclusions
The evaluation of a child with suspected appen-
dicitis should focus on a proper diagnosis, using
modern diagnostic tools to minimize the risk of
misdiagnosis and negative appendectomy.
Appendicitis scoring systems may aid the cor-
rect diagnosis but have limited value as a single
diagnostic tool. Early rehydration, analgesics,
and in selected cases, antibiotic treatment for
72 Appendicitis 991

suspected or diagnosed perforation should be Eriksson S, Granström L (1995) Randomized controlled

prioritized, but there is no evidence that surgery trial of appendicectomy versus antibiotic therapy for
acute appendicitis. Br J Surg 82:166–169
is emergent. Nonoperative treatment with antibi- Esposito C, Calvo AI, Castagnetti M, Alicchio F, Suarez
otics is a feasible option, but until further evi- C, Giurin I et  al (2012) Open versus laparoscopic
dence is established with respect to outcomes, appendectomy in the pediatric population: a literature
the surgical removal of the vermiform appendix, review and analysis of complications. J Laparoendosc
Adv Surg Tech A 22:834–839
preferably by minimally invasive surgery, is Fugazzola P, Coccolini F, Tomasoni M, Stella M, Ansaloni
recommended. L (2019) Early appendectomy vs. conservative man-
agement in complicated acute appendicitis in children:
a meta-analysis. J Pediatr Surg 54:2234–2241
Giannis D, Matenoglou E, Moris D (2020) Hyponatremia
References as a marker of complicated appendicitis: a systematic
review. Surgeon 18:295–304
Almström M, Svensson JF, Patkova B, Svenningsson A, Hall NJ, Eaton S, Stanton MP, Pierro A, Burge DM (2017)
Wester T (2017) In-hospital surgical delay does not Active observation versus interval appendicectomy
increase the risk for perforated appendicitis in chil- after successful non-operative treatment of an appen-
dren: a single-center retrospective cohort study. Ann dix mass in children (China study): an open-label,
Surg 26:616–621 randomised controlled trial. Lancet Gastroenterol
Almström M, Svensson JF, Svenningsson A, Hagel E, Hepatol 2:253–260
Wester T (2018) Population-based cohort study on Jaschinski T, Mosch CG, Eikermann M, Neugebauer EA,
the epidemiology of acute appendicitis in children in Sauerland S (2018) Laparoscopic versus open surgery
Sweden in 1987-2013. BJS open 2:142–150 for suspected appendicitis. Cochrane Database Syst
Alvarado A (1986) A practical score for the early diag- Rev 11:Cd001546
nosis of acute appendicitis. Ann Emerg Med 15: Kaiser S, Mesas-Burgos C, Söderman E, Frenckner B
557–564 (2004) Appendicitis in children--impact of us and ct
Andersson RE (2007) The natural history and traditional on the negative appendectomy rate. Eur J Pediatr Surg
management of appendicitis revisited: spontaneous 14:260–264
resolution and predominance of prehospital perfora- Kang JY, Hoare J, Majeed A, Williamson RC, Maxwell
tions imply that a correct diagnosis is more important JD (2003) Decline in admission rates for acute appen-
than an early diagnosis. World J Surg 31:86–92 dicitis in England. Br J Surg 90:1586–1592
Andersson M, Andersson RE (2008) The appendicitis Kim K, Kim YH, Kim SY, Kim S, Lee YJ, Kim KP et al
inflammatory response score: a tool for the diagnosis (2012) Low-dose abdominal ct for evaluating sus-
of acute appendicitis that outperforms the Alvarado pected appendicitis. N Engl J Med 366:1596–1605
score. World J Surg 32:1843–1849 Livingston EH, Woodward WA, Sarosi GA, Haley RW
Aziz O, Athanasiou T, Tekkis PP, Purkayastha S, Haddow (2007) Disconnect between incidence of nonper-
J, Malinovski V et al (2006) Laparoscopic versus open forated and perforated appendicitis: implications
appendectomy in children: a meta-analysis. Ann Surg for pathophysiology and management. Ann Surg
243:17–27 245:886–892
Bhangu A, Søreide K, Di Saverio S, Assarsson JH, Drake Macco S, Vrouenraets BC, de Castro SM (2016)
FT (2015) Acute appendicitis: modern understanding Evaluation of scoring systems in predicting acute
of pathogenesis, diagnosis, and management. Lancet appendicitis in children. Surgery 160:1599–1604
386:1278–1287 Maita S, Andersson B, Svensson JF, Wester T (2020)
Bliss D, McKee J, Cho D, Krishnaswami S, Zallen G, Nonoperative treatment for nonperforated appendici-
Harrison M et  al (2010) Discordance of the pediat- tis in children: a systematic review and meta-analysis.
ric surgeon's intraoperative assessment of pediatric Pediatr Surg Int 36:261–269
appendicitis with the pathologists report. J Pediatr Moore MM, Kulaylat AN, Hollenbeak CS, Engbrecht
Surg 45:1398–1403 BW, Dillman JR, Methratta ST (2016) Magnetic reso-
Cameron DB, Williams R, Geng Y, Gosain A, Arnold nance imaging in pediatric appendicitis: a systematic
MA, Guner YS et al (2018) Time to appendectomy for review. Pediatr Radiol 46:928–939
acute appendicitis: a systematic review. J Pediatr Surg Ohle R, O'Reilly F, O'Brien KK, Fahey T, Dimitrov BD
53:396–405 (2011) The Alvarado score for predicting acute appen-
Carr NJ (2000) The pathology of acute appendicitis. Ann dicitis: a systematic review. BMC Med 9:139
Diagn Pathol 4:46–58 Patkova B, Svenningsson A, Almström M, Eaton S,
Coldrey E (1959) Five years of conservative treatment of Wester T, Svensson JF (2020) Nonoperative treatment
acute appendicitis. J Int Coll Surg 32:255–261 versus appendectomy for acute nonperforated appen-
Doria AS, Moineddin R, Kellenberger CJ, Epelman M, dicitis in children: five-year follow up of a randomized
Beyene J, Schuh S et  al (2006) Us or ct for diagno- controlled pilot trial. Ann Surg 271:1030–1035
sis of appendicitis in children and adults? A meta-­ Ponsky TA, Hafi M, Heiss K, Dinsmore J, Newman KD,
analysis. Radiology 241:83–94 Gilbert J (2009) Interobserver variation in the assess-
992
ment of appendiceal perforation. J Laparoendosc Adv
Surg Tech A 19(Suppl 1):S15–S18
Puylaert JB (1986) Acute appendicitis: us evaluation
using graded compression. Radiology 158:355–360
Rentea RM, Peter SDS, Snyder CL (2017) Pediatric
appendicitis: state of the art review. Pediatr Surg Int
33:269–283
Samuel M (2002) Pediatric appendicitis score. J Pediatr
Surg 37:877–881
Semm K (1983) Endoscopic appendectomy. Endoscopy
15:59–64
Spencer J, Finn T, Isaacson PG (1985) Gut associated lym-
phoid tissue: a morphological and immunocytochemi-
cal study of the human appendix. Gut 26:672–679
Svensson JF, Patkova B, Almström M, Naji H, Hall NJ,
Eaton S et al (2015) Nonoperative treatment with anti-
biotics versus surgery for acute nonperforated appen-
dicitis in children: a pilot randomized controlled trial.
Ann Surg 261:67–71
Ure BM, Spangenberger W, Hebebrand D, Eypasch EP,
Troidl H (1992) Laparoscopic surgery in children and
M. Almström and T. Wester
adolescents with suspected appendicitis: results of
medical technology assessment. Eur J Pediatr Surg
2:336–340
Vons C, Barry C, Maitre S, Pautrat K, Leconte M,
Costaglioli B et  al (2011) Amoxicillin plus clavu-
lanic acid versus appendicectomy for treatment of
acute uncomplicated appendicitis: an open-label,
non-inferiority, randomised controlled trial. Lancet
377:1573–1579
Yu CW, Juan LI, Wu MH, Shen CJ, Wu JY, Lee CC (2013)
Systematic review and meta-analysis of the diagnostic
accuracy of procalcitonin, c-reactive protein and white
blood cell count for suspected acute appendicitis. Br J
Surg 100:322–329
Zhang Z, Wang Y, Liu R, Zhao L, Liu H, Zhang J et  al
(2015) Systematic review and meta-analysis of single-­
incision versus conventional laparoscopic appendec-
tomy in children. J Pediatr Surg 50:1600–1609
 Intussusception
Holger Till and Erich Sorantin
73.1 Introduction
Intussusception represents the most common
abdominal emergency in infancy and early child-
hood with an overall incidence of about 1–4%. The
term describes the phenomenon of one intestinal
segment telescoping into the adjacent distal portion.
Approximately 95% of all cases occur within the
first 2  years of life (Gluckman et  al. 2017).
Interestingly, neonates are affected in less than 1%
of all cases, but even newborns with intussusception
have been published anecdotally (Carine Sun et al.
2020). Clinically, the ileocolic region is most com-
monly affected. Intussusception represents a spec-
trum of underlying causes and challenges to treat.
The following chapter elaborates on such details.
73.2 Historical Overview
Intussusception has been recognized since the
1600s, and until the early 1900s, enemas were
the preferred treatment for reduction. Then, sur-
H. Till (*)
University Clinic of Pediatric and Adolescent Surgery,
Medical University of Graz, Graz, Austria
e-mail: holger.till@medunigraz.at
E. Sorantin
Department of Paediatric Radiology, University Clinic of
Radiology, Medical University of Graz, Graz, Austria
e-mail: erich.sorantin@medunigraz.at
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_73
73
gical reduction became increasingly popular,
despite a considerable morbidity and even mor-
tality. Thus, with the refinement of conventional
radiology and especially ultrasound, the nonsur-
gical approach became popular again. Today,
nonoperative reduction represents the first-line
treatment in pediatric intussusception (Carroll
et  al. 2017), and surgical interventions remain
second-­line options in cases of complications,
intractable recurrences, or pathological “leading
points.”
73.3 Incidence
Pediatric intussusception most commonly occurs
in infants between 15 and 19 months of age. Only
10–25% of pediatric cases occur in later child-
hood, and the incidence of a pathological “lead-
ing point,” such as Meckel diverticulum, increases
with age (Lee and Yang 2020).
73.4 Etiopathogenesis
The etiology of pediatric intussusception is usu-
ally idiopathic, with only 10% of cases having an
identifiable precipitating lesion. It is often pre-
ceded by a viral illness (respiratory or gastroin-
testinal), which may contribute to the hypertrophy
of the Peyer’s patches around the terminal ileum.
Especially in older children, distinct pathologies
993
994
of the gastrointestinal tract predispose to an
intussusception, such as Meckel diverticulum,
intestinal duplication, Peutz-Jegher polyps, or
submucosal hemorrhages in Henoch-Schonlein
purpura (HSP).
73.5 Pathophysiology
Transient intussusceptions probably happen quite
often with ongoing peristalsis and recover spon-
taneously. However, if the bowel remains tele-
scoped, progressive edema of the bowel wall
leads to mucosal bleeding, insufficient arterial
blood supply, increased venous stasis, and the
risk of bowel necrosis. “Delayed cases” must ini-
tially be evaluated for systemic, metabolic altera-
tions, due to prolonged intestinal ischemia. Even
after a successful reduction, systemic hypovole-
mia and subsequent shock may occur due to
effective reperfusion of the ischemic bowel and
“systemic flooding” of vasoactive metabolites
(Wu et  al. 2017). Mortality has also been
described (Lee and Yang 2020).
73.6 Pathology
Clinically, the classic triad includes abdominal
colic pain, red jelly stools, and a palpable abdom-
inal mass (Beasley 2017). Such episodes usually
occur in short intervals, while in between epi-
sodes the infant may be clinically unremarkable.
In prolonged disease, the child may develop sig-
nificant abdominal distension, bilious vomiting
and become increasingly lethargic. Such clinical
signs “ring the emergency bell.”
On examination, the mass can be palpated
anywhere in the abdomen. Rectal exam is posi-
tive for blood in 60–90% of cases. Rectal pro-
lapse of the apex of the intussuseptum occurs
from 8% to 40% of all cases.
In children over 2–3 years of age, the presen-
tation of intussusception can be more subtle, and
the classic triad of symptoms may be present in
only up to 50% of patients (Edwards et al. 2017).
In this group, pathological lead points contribute
to a majority of cases. Although the average inci-
H. Till and E. Sorantin
dence of intussusceptions caused by pathological
lead points is about 8%, the incidence is much
higher after the second year of life (20–40%).
There may be symptoms associated with the pri-
mary disease that causes the intussusception,
such as hemorrhagic patches in Henoch-­
Schonlein purpura (HSP) and joint swellings in
coagulation disorders.
73.7 Diagnosis
Ultrasonography of the abdomen represents the
first choice to diagnose intussusception, as it seems
readily available, induces no radiation, and pro-
vides a diagnostic sensitivity of 98.5% and speci-
ficity of 100% (Lin-Martore et  al. 2020). The
primary finding is usually a soft tissue mass
between 2.5 cm and 5.0 cm having specific charac-
teristics: The “doughnut sign” (Fig. 73.1) describes
the edematous bowel wall causing an an−/
hypoechoic external rim, whereas the hyperechoic
center represents the intussuscepted mesenteric
fat. The “target sign” (a variant of the “doughnut
sign”) relates to the central echogenic area and an
additional anechoic, which probably represents
fluid at the apex of the intussusceptum, thus result-
ing in a “target”-like appearance. If the edema is
less severe, the various layers of the mucosa and
submucosa are less stretched and thinned and form
Fig. 73.1  Abdominal ultrasound—a cross-section of an
ileocolic intussusception demonstrating a doughnut sign:
an echogenic center is surrounded by an an−/hypoechoic
rim
73 Intussusception
Fig. 73.2  Abdominal ultrasound—a cross-section of an
ileocolic intussusception. The multiple layers represent
the concentric ring sign
Fig. 73.3  Abdominal ultrasound—longitudinal section
of an ileocolic intussusception. A kidney like structure
with a hypoechoic rim and an echogenic center is depicted.
Furthermore the hypoechoic areas in the echogenic center
represent intussuscepted, mesenterial lymph nodes
multiple concentric rings (Fig.  73.2). This phe-
nomenon is referred to as the “multiple concentric
ring sign.” Finally, the “pseudo-kidney sign”: In
the longitudinal scan, the intussusception appears
like a kidney with a hypoechoic rim and a hyper-
echoic center (Fig. 73.3).
Plain abdominal X-ray films are indicated as
the second line of diagnostics. Plain films are
taken in supine and upright (or alternatively in
left lateral) positions.
995
Fig. 73.4  Upright abdominal plain film of an ileocolic
intussusception. White arrows mark an intussusception
prolapsing into the ascending colon. The surrounding
colonic air leads to the meniscus sign
Characteristic radiological signs of intussus-
ception in such plain X-ray films include the “tar-
get sign,” which depicts a sound soft tissue mass
with central lucency, due to the dragged mesen-
tery fat. “Meniscus sign” meaning crescent gas
within the colonic lumen, which outlines the
apex of the intussusception (Fig. 73.4). Additional
unspecific signs for the presence of an intussus-
ception include a soft tissue mass in the upper
abdomen (Fig. 73.5), obscuring the caudal border
of the liver, empty right lower quadrant, gas in
displaced appendix, and of course signs of small
bowel obstruction.
996
Fig. 73.5  Abdominal plain film in supine position. White
arrows mark a mass in the upper abdomen representing an
intussusception
Due to the excellent sensitivity and specificity
of ultrasound to diagnose intussusception, the
benefit of abdominal CT seems limited to extraor-
dinary cases and distinct complications.
73.8 Differential Diagnosis
The most common differential diagnosis and
comorbidity in infancy include infectious dis-
eases, most likely viral causes. In older children
inflamed appendix and appendicitis, lymphadeni-
tis or lymph node disorders, and Meckel’s diver-
ticulum must be considered.
Cystic fibrosis can commonly present with
intussusception in children between 9 and
12  years of age and is caused by the thick,
putty-­like material in the mucosal lumen acting
as a lead point. Sometimes, the presentation is
more chronic with longstanding symptoms.
Nonoperative reduction is rarely successful
and most children require operative reduction.
H. Till and E. Sorantin
73.9 Management
73.9.1 Resuscitation
Even the “simple” infant with intussusception
should receive a basic assessment of vital param-
eters and laboratory studies before transferring to
diagnostics and treatment. In delayed cases with
prolonged intestinal ischemia, the correction of
dehydration and metabolic alterations is essential
to prevent hypovolemia or shock during the treat-
ment. Antibiotics may be started, because a cer-
tain degree of bacterial translocation during the
nonsurgical reduction may be expected.
73.9.2 Nonsurgical Treatment
Nonsurgical reduction has a longstanding history.
In principle, an enema pushes back the intussus-
cepted bowel. The progress of the reduction can
be monitored by fluoroscopy or ultrasound.
Figure 73.6a–c depict a nonsurgical reduction,
using air under fluoroscopic guidance (last image
hold technique). In the beginning, the intussus-
ception is barely visible (Fig.  73.6a). With
increasing air insufflation, the intussusception
appears as a soft tissue mass (Fig. 73.6b—white
arrows), and the intussusception had been pushed
further until the right colonic flexure. Finally, in
Fig.  73.6c, air in the small bowel confirms the
reduction. The pressure of the insufflated air
should be less than 120  mm Hg, and on the
authors’ institution, only three attempts for non-
surgical reduction are performed. If there is no
success, a switch to open reduction is done.
Pain relief is a crucial point, and at our institu-
tion, we prefer general anesthesia with intubation.
It should not be forgotten that the clinical picture is
an acute abdomen. Other forms of pain relief with
spontaneous breathing exposes the patient to the
risk of aspiration, since the intraluminal pressure
of the gastrointestinal tract is increased.
When ultrasound is used as navigation aid,
saline should be administered in a similar fashion
as air. The sonographer should be skilled in bowel
73 Intussusception
a b
Fig. 73.6  Last image hold technique secondary capture
images from a fluoroscopic-guided air enema in order to
achieve a nonsurgical reduction. Air was insufflated per
rectum. (a) At the beginning of the air enema, the intus-
susception is barely visible. (b) After further air insuffla-
ultrasound especially following bowel loops. It
has to be considered that due to the intussuscep-
tion, the colon is not in the normal position and
especially the cecum is pushed cranial to the right
upper abdomen. Sonographic reduction with
997
tion, the intussusception is outlined clearly as a mass—in
this case it is already pushed back to the right colonic flex-
ure. (c) Last image hold screen shot after the successful
reduction of an ileocolic intussusception—air within the
small bowel seems obvious
saline lasts longer than the reduction by air
enema. The risk of perforation is pretty much the
same for both techniques.
Altogether, a success rate of up to 95% (Sacks
et al. 2020) has been reported for the nonsurgical
998
reduction. Thus, this procedure is considered the
first-line treatment except for very sick children
with a prolonged presentation and risk of bowel
perforation.
73.9.3 Surgical Treatment
Indications for surgery include persistent incom-
plete radiological reduction, despite optimal
technique and several attempts, the risk of perfo-
ration, suspected pathological leading point, or
seriously sick patients with peritonitis or sepsis
(Gluckman et al. 2017). Preparation of the patient
for surgery include decompression of the gastro-
intestinal tract by open nasogastric tube and
resuscitation.
On the operating table, the child is positioned
supine. In cases of incomplete reduction with the
remaining apex close to the ileocecal valve, a
minimally invasive oblique incision in the right
lower quadrant of the abdomen may be sufficient.
However, when the intussusception still reaches
the transverse colon or progressive disease may
require bowel mobilization and resection, a
­longitudinal midline incision seems preferable,
especially in older children. Pararectal incisions
have been abandoned.
By inspection and palpation, the apex of the
intussusception is located. Manual reduction has
to be performed very carefully and slowly. Taking
the apex of the intussusceptum between two fin-
gers of one hand and the adjacent part of the dis-
tal bowel into the other hand, the intussusceptum
is gently squeezed toward the ileocecal valve.
Patience is important to allow the edema to dis-
solve. Pulling on the intussuscepted bowel should
be omitted as it causes tears and ruptures.
Reduction with surgical instruments usually
induces lacerations due to the vulnerable bowel
wall.
Once the reduction has been successfully per-
formed, the intussusceptum must be inspected. It
usually looks quite edematous or even ischemic.
However, rapid decisions for any kind of bowel
resections should not be made on first sight. They
may lead to the loss of the ileocecal valve.
Instead, the bowel should be wrapped into moist
H. Till and E. Sorantin
and warm gauzes to support reperfusion. If the
bowel seems necrotic even then, limited resec-
tions must be considered versus a second-look
operation.
At the same time, a pathological leading point
must be identified. This task may be more chal-
lenging than it sounds, because the “regular
edema” can already mimic intestinal pathologies.
Careful palpation should prevent unnecessary
excisions or resection, particularly in the typical
infant age group. The most common finding is a
“dimple” at the site of the Payer’s plaque or an
edematous ileocecal valve. Meckel’s diverticu-
lum must be resected at this point.
Finally, ileopexy remains a matter of discus-
sion, but conclusive data for this strategy is not
available yet (Zhang et al. 2020; Wei et al. 2015).
73.9.4 The Role of Laparoscopy
Laparoscopy represents an alternative approach
to open surgery but deserves some specific con-
siderations (Beasley 2017; Wei et  al. 2015).
Especially in infants, the working space may be
limited due to the dilatation of the bowel loops,
and the installation of the pneumoperitoneum
using CO2 introduces additional metabolic chal-
lenges. Moreover, laparoscopy challenges the
paradigm of pushing the apex and not pulling the
intussuscepted bowel (Apelt et al. 2013). Thus, a
great expertise in laparoscopic bowel surgery
seems mandatory, and laparoscopy may be
advantageous for selected cases with the apex
close to the ileocecal valve (Houben et al. 2016).
Finally, laparoscopy lacks the tactile feedback
when checking for pathological leading points.
73.9.5 Postoperative Care
The postoperative management depends on the
extent of the disease, surgical procedures, and the
overall condition of the child. In “uncompli-
cated” cases of nonsurgical reduction, enteral
feeding should be started as soon as anesthesia
allows for it. Some authors even recommend out-
patient management based on systematic reviews
73 Intussusception
(Litz et al. 2019). The more complex cases may
require a short-term nasogastric tube and paren-
teral nutrition until the edema of the bowel wall
has disappeared and bowel function has recov-
ered. Routine follow-up ultrasound in asymptom-
atic patients may not be necessary. However,
once the infant becomes symptomatic again, it
must be re-evaluated immediately.
73.9.6 Recurrent Intussusception
Recurrent intussusception occurs in up to 13% of
all cases following nonsurgical and surgical reduc-
tion (Ye et al. 2019). Thirty percent of these recur-
rences occur within the first postoperative day
(Gray et  al. 2014). The clinical complaints are
similar to a primary onset. Diagnostics and non-
surgical reduction may be repeated especially in
infants and younger children with gastrointestinal
infections. However, with every recurrence, the
success rate for the nonsurgical treatment
decreases, and the incidence for a pathological
leading point increases. Hsu et al. (2012) analyzed
the recurrence rate in 686 children after the first,
second, third, and fourth barium enema reductions
and found 15.7%, 37.7%, 68.4%, and 100%.
Consequently, surgery becomes more important in
recurrences to identify and treat a pathological
leading point. The role of ileopexy in cases with
multiple recurrences remains a matter of discus-
sion, and conclusive data for one strategy is not
available yet (Zhang et al. 2020; Wei et al. 2015).
73.9.7 Chronic Intussusception
Finally, chronic intussusception is a rare condi-
tion. This means a non-strangulating, long-­lasting
intussusception of a minimum of 14 days, mostly
due to anatomical leading points. The clinical
picture is dominated by chronic diarrhea resistant
to any therapy. Patients are evacuating bloody or
mucous stools and suffer from colicky pain
including episodes of vomiting all the time,
finally leading to weight loss. Diagnosis can be
achieved by ultrasound, contrast studies, or even
by laparoscopy. The treatment is similar to acute
intussusception.
999
73.10 Conclusion
Pediatric intussusception represents a spectrum
of underlying causes and modalities to treat.
Management requires an interdisciplinary team
of pediatricians, pediatric radiologists, and pedi-
atric surgeons. Teamwork and timing while exe-
cuting best-practice standards guarantees an
excellent overall prognosis.
References
Apelt N, Featherstone N, Giuliani S (2013) Laparoscopic
treatment of intussusception in children: a systematic
review. J Pediatr Surg 48:1789–1793
Beasley SW (2017) The ‘ins’ and ‘outs’ of intussuscep-
tion: where best practice reduces the need for surgery.
J Paediatr Child Health 53(11):1118–1122
Carine Sun CY, Ashok K, Mughni B (2020 May) A rare
case of intrauterine intussusception causing ileal atre-
sia. Med J Malaysia 75(3):304–306
Carroll AG, Kavanagh RG, Leidhin CN, Cullinan NM,
Lavelle LP, Malone DE (2017) Comparative effective-
ness of imaging modalities for the diagnosis and treat-
ment of intussusception: a critically appraised topic.
Acad Radiol 24(5):521–529
Edwards EA, Pigg N, Courtier J, Zapala MA, MacKenzie
JD, Phelps AS (2017) Intussusception: past, present
and future. Pediatr Radiol 47(1):1101–1108
Gluckman S, Karpelowsky J, Webster AC, McGee RG
(2017) Management for intussusception in children.
Cochrane Database Syst Rev 6(6):CD006476
Gray M, Li HS, Hoffmann RG, Gorelick MH (2014)
Recurrence rates after intussusception enema reduc-
tion: a meta-analysis. Pediatrics 134(1):110–119
Houben CH, Feng XN, Tang SH, Chan EK, Lee KH
(2016) What is the role of laparoscopic surgery in
intussusception? ANZ J Surg 86(6):504–508
Hsu WL, Lee HC, Yeung CY, Chan WT, Jiang CB, Sheu
JC, Wang NL, Shih SL (2012) Recurrent intussuscep-
tion: when should surgical intervention be performed?
Pediatr Neonatol 53:300–303
Lee EH, Yang HR (2020) Nationwide population-
based epidemiological study on childhood intus-
susception in South Korea: emphasis on treatment
and outcomes. Pediatr Gastroenterol Hepatol Nutr
23:329–345
Lin-Martore M, Kornblith AE, Kohn MA, Gottlieb M
(2020) Diagnostic accuracy of point-of-care ultra-
sound for intussusception in children presenting to the
emergency department: a systematic review and meta-­
analysis. J Emerg Med 21(4):1008–1016
Litz CN, Amankwah EK, Polo RL, Sakmar KA, Danielson
PD, Chandler NM (2019) Outpatient management
of intussusception: a systematic review and meta-­
analysis. J Pediatr Surg 54(7):1316–1323
1000
Sacks RS, Anconina R, Farkas E, Zolotnik-Krupenich D,
Kravarusic D, Tsodikov V, Shelef I, Taragin B (2020)
Sedated ultrasound guided saline reduction (SUR)
of ileocolic intussusception: 20 year experience. J
Pediatr Surg 55:2009–2014. https://doi.org/10.1016/j.
jpedsurg.2020.05.049
Ye X, Tang R, Chen S, Lin Z, Zhu J (2019) Risk factors
for recurrent intussusception in children: a systematic
review and meta-analysis. Front Pediatr 7:145
Wei CH, Fu YW, Wang NL, Du YC, Sheu JC (2015)
Laparoscopy versus open surgery for idiopathic intus-
susception in children. Surg Endosc 29(3):668–672
H. Till and E. Sorantin
Wu MJ, Chen M, Sang S, Hu LL, Tian ML, Li K, Lv
FQ (2017) Protective effects of hydrogen rich water
on the intestinal ischemia/reperfusion injury due to
intestinal intussusception in a rat model. Med Gas Res
7:101–106
Zhang Y, Wang Y, Zhang Y, Hu X, Li B, Ming G (2020)
Laparoscopic Ileopexy versus laparoscopic simple
reduction in children with multiple recurrences of
Ileocolic intussusception: a single-institution retro-
spective cohort study. J Laparoendosc Adv Surg Tech
A 30(5):576–580
 Hernias
Anna Svenningsson and Tomas Wester
74.1 Inguinal Hernia
74.1.1 Introduction
Inguinal hernias are among the most common
surgical conditions in infants and children, with a
peak incidence during the first 3 months of life.
In children, inguinal hernias are almost always
indirect hernias. Direct inguinal hernias are rare,
particularly in neonates. The most important fac-
tor in the management of the neonate with a her-
nia is the high risk of incarceration. Inguinal
hernias do not to go away spontaneously and
should therefore always be repaired.
74.1.2 Historical Overview
Descriptions of inguinal hernias date back to
ancient Egypt. One of the oldest preserved medi-
cal documents, the Egyptian papyrus of Ebers
circa 1550 BC, included an observation of ingui-
nal hernia. A probable attempt of surgical treat-
A. Svenningsson
Department of Pediatric Surgery, Karolinska
University Hospital, Stockholm, Sweden
e-mail: anna.svenningsson@ki.se
T. Wester (*)
Department of Pediatric Surgery, Astrid Lindgren
Children’s Hospital, Karolinska University Hospital,
Stockholm, Sweden
e-mail: tomas.wester@karolinska.se
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_74
74
ment of a scrotal hernia was observed in the
mummification of Pharaoh Merneptah in
1200 BC. The first detailed scientific description
of surgical hernia repair is credited to Celsus in
50  AD.  Processus vaginalis was described in
200 AD by Galen, who also thought a hernia was
the result of a rupture of the peritoneum with the
stretching of the overlying fascia and muscles
(Lau 2002). In the sixteenth century, the French
surgeon, Ambroise Paré, advocated a repair of
inguinal hernia in childhood (Snyder 2010).
Numerous different surgical methods for hernia
repair were developed during the nineteenth cen-
tury, as an accurate understanding of the anatomy
of the inguinal canal evolved. However, recur-
rence and complication rates were high. In 1887,
Bassini revolutionized the surgical treatment of
inguinal hernia when he introduced the technique
with the reconstruction of the posterior wall of
the inguinal canal. The term “hernia” is derived
from the Greek word “hernios” meaning bud or
offshoot (Lau 2002).
74.1.3 Incidence
The incidence of congenital indirect inguinal
hernia in infants and children is 1–5%
(Burgmeier et al. 2014; Chang et al. 2016). The
incidence of inguinal hernia in preterm infants
is considerably higher and ranges from 9% to
11% (Rescorla and Grosfeld 1984). The inci-
1001
1002
dence approaches 60% as birth weight decreases
to 500–750  g (Nakayama and Rowe 1989).
Inguinal hernia is more common in males than
in females, with a ratio ranging from 5:1 to 10:1
(Given and Rubin 1989). Approximately 60% of
inguinal hernias occur on the right side, 30% are
left-sided, and 10% are bilateral (Ein et  al.
2006). Bilateral hernias are more common in
premature infants and are reported to occur in
44–55% of cases (Rescorla and Grosfeld 1984;
Harper et  al. 1975; Boocock and Todd 1985).
There is a high familial incidence, and inguinal
hernia has been observed with increased fre-
quency in twins and siblings of patients with
inguinal hernia (Czeizel and Gardonyi 1979).
There is no geographic or ethnic predominance
reported in the literature.
74.1.4 Etiopathogenesis
Congenital indirect inguinal hernia develops
because the processus vaginalis remains patent
after birth. The processus vaginalis is a potential
hernia, which becomes an actual hernia when it
contains some part of the abdominal viscera.
The processus vaginalis is an outpouching of
the peritoneum that extends through the ingui-
nal canal. The processus vaginalis is first seen
during the third month of intrauterine life. It fol-
lows the gubernaculum and testis through the
inguinal canal and reaches the scrotum by the
seventh month of gestation (Snyder 2010). In
the female, the etiology is less clear, although
the processus vaginalis extends along the round
ligament (Lau et  al. 2007). Obliteration of the
processus vaginalis starts soon after the descent
of the testis is completed and continues after
birth. Most infants have a patent processus vagi-
nalis several months after birth. Patency has
been reported to be 80–94% in the newborn
period, approximately 60% in the 4–12-month-
old group and 20% in adulthood (Nakayama and
Rowe 1989). There is an increased incidence of
inguinal hernia in patients with various associ-
ated conditions (Table  74.1) (Snyder 2010;
Husmann et al. 1990; Powell et al. 1986; Sarpel
et al. 2005).
A. Svenningsson and T. Wester
Table 74.1  Associated conditions showing an increased
incidence of inguinal hernia
Urogenital conditions
Undescended testis
Bladder exstrophy
Increased amounts of peritoneal fluid
Ventriculoperitoneal shunts
Peritoneal dialysis
Increased intraabdominal pressure
Repair of gastroschisis or omphalocele
Meconium peritonitis
Chylous ascites
Necrotizing enterocolitis
Chronic respiratory disorders
Cystic fibrosis
Connective tissue disorders
Ehler-Danlos syndrome
Marfan syndrome
Hurler-hunter syndrome
Disorders of sex development
Complete androgen insensitivity syndrome (CAIS)
74.1.5 Diagnosis
74.1.5.1 Clinical Features
The presenting feature is a bulge in the groin,
usually noticed by the parents, which increases in
size with crying. The bulge may disappear spon-
taneously when the patient is quiet and relaxed. If
the lump in the groin is reduced, it is usually pos-
sible to feel a thickening of the cord structures by
the hernia sac. A reliable history along with pal-
pation of a thickened cord is highly suggestive of
an inguinal hernia. Once the diagnosis of an
inguinal hernia is made, elective herniotomy
should be done as soon as possible because of the
high risk of incarceration particularly in the new-
born infant (Olesen et  al. 2019a, 2019b). It is
very important to examine the contralateral side.
74.1.5.2 Incarcerated Inguinal Hernia
Incarceration occurs when the contents of the sac
can’t be reduced easily into the abdominal cavity.
Strangulation refers to vascular compromise of
the content of the sac because of constriction at
the neck of the sac. If there is a delay in the treat-
ment, incarceration rapidly progresses to strangu-
lation. The contents of the hernia sac may be a
small bowel, the cecum, appendix, omentum or
ovary, and fallopian tube. The overall incidence of
74 Hernias
incarceration has been reported to be 3–16%
(Dreuning et  al. 2020). The incidence of incar-
ceration in neonates and young infants is consid-
erably higher and varies between 24% and 40%
(Grosfeld 1989; Puri et  al. 1984; Misra et  al.
1994). The incarceration rate is much higher in
premature infants compared with full-term infants
(Dreuning et  al. 2020). There have been several
reports of testicular infarction, in association with
an incarcerated inguinal hernia. In infants under
the age of 3 months, gonadal infarction has been
reported to occur in 30% of the cases (Schmitt
et  al. 1987). The incidence of testicular atrophy
following incarceration of inguinal hernia has
been reported to be between 10% and 15%. Most
of the reports on testicular infarction or atrophy
following incarcerated inguinal hernia, however,
refer to infants who require emergency operation
for the incarceration. One report showed that
there was a statistically ­significant difference in
testicular volume between infants with incarcer-
ated inguinal hernias, that first were reduced and
then electively operated and age-matched controls
(Fasching and Höllwarth 1989).
Neonates with incarcerated inguinal hernia
usually present with irritability, vomiting, a mod-
erately distended abdomen, and a tender groin
lump. Occasionally, the infant may pass blood
per rectum. Local examination reveals a tense,
tender lump in the groin. The upper margin is not
well-defined. The testes may be normal or swol-
len and hard, due to vascular compromise. Rectal
examination is usually not necessary, but, if done,
contents of the hernia can be palpated at the inter-
nal ring. The diagnosis of incarcerated inguinal
hernia is usually made on clinical findings.
Abdominal radiographs may occasionally show
bowel gas within the lump in the groin and con-
firm the diagnosis. If intestinal obstruction is
present, plain abdominal films will show dilated
loops of bowel with fluid levels.
74.1.6 Differential Diagnosis
Clinical diagnosis of incarcerated inguinal hernia
is usually easy, but it may sometimes be difficult
to differentiate a hydrocele, inguinal lymphade-
1003
nitis, and torsion of the testis. In a hydrocele, it is
possible to get above the swelling, which is non-­
tender. Transillumination is not a reliable sign in
infants, as bowel can be transilluminant because
of its thin wall. Hydrocele of the cord is difficult
to differentiate from incarcerated hernia. There is
no previous history of reducible groin lump in
these patients. Rectal examination may be useful
in excluding incarcerated hernia. In lymphadeni-
tis, examination of the area of drainage will
reveal the source of infection. The cord and testis
are normal. In scrotal testicular torsion, it is pos-
sible to get above the swelling. Testis is tender
and slightly higher than on the other side, while
the torsion of the testis situated in the superficial
inguinal pouch will be associated with an empty
scrotum on the same side.
74.1.7 Management
The treatment of inguinal hernia is surgery. The
ideal time for surgery is as soon as possible after
the diagnosis has been made because of the risk
of incarceration. In a survey of the members of
the Surgical Section of the American Academy
of Pediatrics, 79% responded that they repair
asymptomatic reducible hernias in full-term
boys electively. In full-term girls with reducible
ovaries and no symptoms, 49% chose to repair
electively. In former preterm babies, the major-
ity of surgeons would operate as soon as conve-
nient, regardless of age. Others operate before
discharge from the neonatal intensive care unit.
In the survey of the members of the Surgical
Section of the American Academy of Pediatrics,
there is a trend toward earlier repair (Antonoff
et  al. 2005). Early repair is justified by the
increased risk of incarceration during the wait-
ing period. It has also been suggested that wait-
ing can result in adhesions between the hernia
sac and the spermatic cord, increasing the risk
of damage during the operation (Wester 2009).
Most of the inguinal hernia operations in older
infants and children are done as day-case proce-
dures, except in children with cardiac, respira-
tory, and other conditions that increase the risk
for anesthetic complications.
1004 A. Svenningsson and T. Wester

In patients presenting with incarceration, block. Spinal anesthesia can be an alternative to

reduction is the preferred treatment as long as the
patient is stable. The policy of nonoperative
reduction is based on the following facts: the
likelihood of reducing strangulated bowel in
infants is extremely rare, and the complication
rate is higher with emergency operations for
incarcerated hernia (Rowe and Clatworthy 1970).
The infant is placed in the Trendelenburg posi-
tion. This helps to relieve the edema and allows
mild traction on the contents of the hernia.
Adequate sedation is given to the infant to relax
abdominal muscles. If the hernia is not reduced
within an hour with these measures, an attempt is
made to reduce the hernia with gentle compres-
sion, where constant pressure is applied on the
fundus of the sac in the direction of the cord. The
vast majority of incarcerated hernias reduce with
these nonoperative measures. After the hernia is
reduced, the infant is kept in the hospital and
observed. Elective operation is carried out
24–48 h after the reduction, when the edema and
swelling have subsided. Failure to reduce the her-
nia is an indication for emergency operation.
These infants have to be stabilized before sur-
gery. Nasogastric suction and correction of fluid
end electrolyte balance are undertaken.
general anesthesia in low birth weight infants
undergoing inguinal hernia repair, as it is associ-
ated with a lower incidence of postoperative
apnea (Davidson et al. 2015).
74.1.7.2 Operation
Inguinal herniotomy is the procedure of choice.
The operation consists of simple ligation of the
hernia sac without opening the external ring. In
older children, the inguinal canal is longer, and it
is advisable to open the external oblique aponeu-
rosis to make it possible to ligate the sac high
enough. The child is placed in a supine position,
on a heating blanket. A transverse inguinal skin
crease incision is placed above and lateral to the
pubic tubercle (Fig. 74.1). The subcutaneous tis-
sue and the fascia of Scarpa are opened until the

74.1.7.1 Anesthesia
General anesthesia is required in the majority of
infants and children with inguinal hernia.
Endotracheal intubation is preferred in small
infants. In older infants and children, airway
management with a mask or laryngeal mask is a
commonly used alternative. Premature infants
undergoing surgery have an increased risk of life-­
threatening postoperative apnea. This risk corre-
lates with postconceptual and gestational age. It
has been reported that the risk of postoperative
apnea is less than 1% at a postconceptual age of
54 weeks (Coté et al. 1995). Except for age, ane-
mia is also an independent risk factor for postop-
erative apnea (Ozdemir and Arikan 2013).
Regional anesthesia is commonly used as a sup-
plement to general anesthesia and to give postop-
erative pain relief. The most common forms are
caudal block or regional nerve blocks, for Fig. 74.1  A transverse inguinal skin crease incision is
instance, ilioinguinal/iliohypogastric nerve placed above and lateral to the pubic tubercle
74 Hernias 1005

Fig. 74.2  The subcutaneous fat and fascia of Scarpa are

opened
Fig. 74.4  The external spermatic fascia and cremaster
are separated along the length of the cord by blunt dissec-
tion. The hernia sac is separated from the vas and vessels

Fig. 74.3  The external inguinal ring is exposed

cord is seen emerging from the external ring

(Figs. 74.2 and 74.3). The external spermatic fas-
cia and cremaster are separated along the length
of the cord by blunt dissection. The hernia sac is
identified and gently separated from the vas and
vessels (Fig. 74.4). A hemostat is placed on the Fig. 74.5  The sac is transfixed at the level of the internal
fundus of the sac. The sac is twisted so as to inguinal ring, which is marked by an extraperitoneal pad
reduce any contents into the abdominal cavity. of fat
Some surgeons use a spoon to keep the vas and
vessels away from the neck of the sac. The sac is
transfixed with a 4–0 absorbable suture at the
level of the internal ring, which is marked by an
extraperitoneal pad of fat. The part of the sac
beyond the stitch is excised (Fig.  74.5). 4–0
absorbable sutures, using two to three interrupted
stitches, approximate subcutaneous tissues
(Fig. 74.6). The skin is closed with a 5–0 mono-
filament absorbable suture, using a continuous
subcuticular stitch (Fig. 74.7). A small dressing is
applied to the wound. At the end of the operation,
the testes must be pulled into the scrotum to avoid Fig. 74.6  The subcutaneous tissue is approximated with
iatrogenic ascent of the testis. interrupted stitches
1006 A. Svenningsson and T. Wester

through a laparotomy. In a large series with 743

incarcerated hernias, only two patients required a
bowel resection, indicating that this is quite rare.
Testes are put in the scrotum irrespective of
whether they are normal or ischemic. Only
frankly necrotic testes may be removed.

74.1.7.3 Contralateral Exploration

Whether the contralateral side should be explored
or not has been discussed by pediatric surgeons
for many years. Some surgeons perform a contra-
Fig. 74.7  The skin is closed with a continuous subcuticu- lateral exploration in premature infants, because
lar suture
of the high incidence of bilateral hernias in this
age group (Rescorla and Grosfeld 1984; Harper
Adequate postoperative analgesia is achieved et  al. 1975; Boocock and Todd 1985). Others
by regional anesthesia, caudal anesthesia, or ilio- consider contralateral exploration unnecessary as
inguinal/iliohypogastric nerve block, which is only 7–10% of children will actually develop a
either administered before or at the end of the metachronous hernia on the other side (Surana
operation. Feeding is resumed as soon as the and Puri 1993; Ron et al. 2007). Except for pre-
child is awake. Most of the infants can be dis- maturity, incarceration correlates with a higher
charged the same day. Infants with a postconcep- risk for the development of a contralateral hernia
tual age less than approximately 60  weeks are (Schwöbel et al. 1999; Tackett et al. 1999). In a
postoperatively monitored for 24 h because of the survey of the current practice of 395 members of
risk of apneic episodes. the Surgical Section of the American Academy of
If the hernia is operated on as an emergency Pediatrics, 44% reported that they routinely
procedure, it often reduces spontaneously. The explore the contralateral side in boys less than
sac is then opened and the bowel inspected. If 2  years of age. Forty-seven percent routinely
there is no evidence of intestinal ischemia, a high explore the contralateral side in girls under the
ligation of the sac is done as in elective cases. If age of 4 years (Antonoff et al. 2005). It has been
there are indications of ischemia, such as blood-­ suggested that ultrasound is a good technique to
stained fluid in the sac or if miscolored bowel is show if there is a patent processus vaginalis on
seen through the opened sac, the bowel has to be the contralateral side. The advantages are that the
examined through a laparotomy. If the bowel diagnostic accuracy is high and that ultrasound is
does not reduce spontaneously, no attempts are noninvasive (Dreuning et  al. 2019a, 2019b).
made to reduce the hernia when the patient is Laparoscopy has also been shown to be highly
anesthetized. The sac is opened and the bowel sensitive in detecting a contralateral patent pro-
examined. If the bowel is viable, it is reduced. cessus vaginalis (Muensterer and Gianocolo
The internal ring is split if this is difficult. On the 2019). Although a patent processus vaginalis is
other hand, if the viability of the bowel is ques- shown, the majority of patients do not develop a
tionable, it is delivered out, and warm saline metachronous hernia after ipsilateral repair
soaks are applied. The intestine is examined after (Kokorowski et al. 2014). Contralateral explora-
about 5–10  min. If the color of the intestine tion is therefore not considered to be justified.
returns back to normal with adequate perfusion,
peristalsis is visible, and mesenteric arterial pul- 74.1.7.4 Laparoscopic Repair
sations are seen, the intestine is returned to the of Inguinal Hernia
abdomen and a herniotomy performed. If the Laparoscopic repair of inguinal hernias has
bowel is nonviable, resection and anastomosis emerged as an alternative method to open ingui-
are performed either through the same incision or nal herniotomy. Since first described more than
74 Hernias
two decades ago, there has been an ongoing
debate over which method is superior to the other.
Recent meta-analyses of complications as testic-
ular atrophy, iatrogenic ascent of the testis, infec-
tion, and recurrence rate have shown comparable
results of the two methods (Dreuning et al. 2019a,
2019b; Olesen et al. 2019a, 2019b). The majority
of pediatric surgeons do not use laparoscopic
repair as their standard method. In a recent sur-
vey at a joint congress of the European Paediatric
Surgeons’ Association and British Association of
Paediatric Surgeons, 83% of the 187 responding
surgeons stated open herniotomy as their pre-
ferred approach (Zani et  al. 2014). The main
advantage of laparoscopic hernia repair is the
opportunity to evaluate the contralateral side for
a patent processus vaginalis. When present, it is
closed to prevent the development of a contralat-
eral metachronous hernia (Dreuning et al. 2019a,
2019b). The probability of a contralateral patent
processus vaginalis is age-dependent, highest in
infants, and then gradually decreases with age
(Hall et  al. 2012). Therefore, some surgeons
restrict the use of laparoscopic hernia repair to
infants. Repeated studies have shown shorter
operative time for bilateral hernia repair with
laparoscopic technique compared to open sur-
gery (Dreuning et al. 2019a, 2019b; Olesen et al.
2019a, 2019b). There is no clear evidence of a
favorable outcome on postoperative pain with the
laparoscopic approach (Dreuning et  al. 2019a,
2019b). Laparoscopy is an attractive option for
the repair of recurrent inguinal hernia since dis-
section in a scarred tissue is avoided that other-
wise may increase the risk of damaging the vas or
testicular vessels (Lau et  al. 2007). Also, rare
events of a missed direct hernia can be excluded
(Lee and DuBois 2000). Laparoscopy allows
screening for complete androgen insensitivity
syndrome (CAIS) in girls with inguinal hernia by
visual assessment of the uterus and gonads.
The first laparoscopic method for hernia repair
was described in late 1990s. Since then, several
different techniques for the closure of the sac
have evolved. These can be subdivided into those
using intracorporeal suturing inside the abdomi-
nal cavity or extracorporeal suturing performed
extraperitoneally. The extracorporeal method has
1007
the benefit of not requiring the surgeon having
expertise in laparoscopic suturing technique. In
addition to different approaches regarding how to
close the inner inguinal ring, handling of the sac
differs. The sac can be transected, resected,
inverted, or left intact in combination with the
closure of the inner inguinal ring. Also, tech-
niques with resection of the sac without closure
of the ring are described, relying on peritoneal
scarring to close the sac. Examples of techniques
using intracorporeal suturing are purse string
suture encircling the inner inguinal ring
(Fig. 74.9) and closure with a Z-suture alone or in
combination with a purse string suture.
Techniques using extracorporeal suturing have
gained increased popularity in recent years and
includes different variants of laparoscopic-­
assisted percutaneous internal ring suturing. A
suture is introduced circumferential of the inner
inguinal ring with a percutaneous needle under
laparoscopic visual control. The suture is tied
extracorporeally and the knot buried in the sub-
cutaneous tissue. According to the European
Paediatric Surgeons’ Association and British
Association of Paediatric Surgeons survey, purse
string suture was the most commonly used lapa-
roscopic technique for pediatric hernia repair
used by 58% of the responders. The second most
common was the laparoscopic-guided percutane-
ous closure used by 28% (Zani et al. 2014).
Laparoscopic hernia repair with intracorpo-
real purse string method:
The procedure is performed under general
anesthesia with the child placed in a Trendelenburg
position. The surgeon is positioned cranial to the
patient with the screen placed at the patient’s
feet. The assistant holding the camera sits next to
the surgeon. For the procedure, 3  mm instru-
ments and a 3 mm 30° telescope are used. With
open technique, a 3 mm umbilical camera port is
established through an infraumbilical incision.
Depending on the size of the patient, pneumo-
peritoneum of 6–10 mmHg is maintained during
the operation. Stab incisions are made in the left
and right lateral border of the rectus muscle,
slightly below the umbilical plane, to achieve tri-
angulation (Fig. 74.8). A needle holder with 4–0
nonabsorbable monofilament suture and a 3 mm
1008
Fig. 74.8  A subumbilical incision is used for the camera
port. Additional stab incisions are used for the ports on
both sides of the umbilicus
grasper are introduced through the stab incisions.
The inguinal anatomy is assessed with identifica-
tion of the deep inguinal ring, epigastric vessels,
and in boys the vas deferens and the testicular
vessels. The vas and testicular vessels extend
subperitoneally out through the deep inguinal
ring, the vas medially to the vessels. In girls, the
uterus and gonads should be identified. Any
abdominal content in the hernia sac is reduced
back into the abdomen. A purse string suture is
placed in the peritoneum encircling the deep
inguinal ring, with careful attention not to inter-
fere with the vas or vessels (Fig. 74.9). The her-
nia sac is then closed by tightening and tying of
the purse string suture intracorporeally. Before
tightening the purse string suture, gas in the her-
nia sac should be reduced back into the a­ bdominal
cavity by external compression of the scrotum.
The contralateral side is evaluated for patent pro-
cessus vaginalis and if present closed in the same
manner as described. When retracting the instru-
ments from the stab incisions, attention must be
paid that no omentum is pulled out through the
wounds which could result in incisional hernia.
The stab incisions are closed with skin glue or a
A. Svenningsson and T. Wester
Fig. 74.9  A purse string suture is placed in the perito-
neum, encircling the deep inguinal ring, with careful
attention not to interfere with the vas or vessels
suture. The fascia at the umbilical port incision is
closed with monofilament absorbable suture and
the skin with absorbable monofilament subcutic-
ular suture. Several different variants of this pro-
cedure are described, e.g., an additional Z-suture
following the purse string suture or transection of
the sac at the level of the inner inguinal ring
before closing the proximal end of the sac with a
purse string suture. The postoperative regime is
the same as for open surgery. Caudal analgesia or
local infiltration anesthesia at the incision sites is
recommended.
74.1.8 Complications
The overall complication rate with elective hernia
repair is low, about 2%, while complications are
much more frequent, 8–33%, for the incarcerated
hernias requiring emergency operations (Rescorla
and Grosfeld 1984; Rowe and Clatworthy 1970).
The most common complications are shown in
List 1. Hematomas can be avoided with meticu-
lous attention to hemostasis. Rarely, it is necessary
to evacuate a hematoma. The wound infection rate
74 Hernias
is low at approximately 1% of the cases (Grosfeld
1989; Audry et  al. 1994; Phelps and Agrawal
1997; Ein et  al. 2006). Gonadal complications
occur due to the compression of the vessels.
Although a large number of testes look nonviable
in patients with incarcerated hernia, the actual
incidence of testicular atrophy is low (Puri et al.
1984). Therefore, unless the testis is frankly
necrotic, it should not be removed. Iatrogenic
ascent of the testes is important as it may be avoid-
able. One investigator showed that at follow-up of
116 patients who had unilateral inguinal herniot-
omy performed before 6 months of age, 3 patients
subsequently required orchidopexy (Surana and
Puri 1993). Another author reported that second-
ary orchidopexy was required in 0.5% of the cases
(Ein et  al. 2006). This is probably due to the
entrapment of the testes in the scar tissue or failure
to pull the testes down in the scrotum at the end of
the operation and to maintain them there (Surana
and Puri 1994). Injuries to the vas have been
reported to be very low, 0.06% (Ein et al. 2006).
The acceptable recurrence rate for inguinal hernia
repair has been considered to be approximately
1% (Rowe and Marchildon 1981). Most of the
recurrences occur within 5 years. The factors that
predispose to recurrence are ventriculoperitoneal
shunt, sliding hernia, incarceration, and connec-
tive tissue disorders (Grosfeld et  al. 1991). The
recurrence may be indirect or direct. Indirect
recurrences are due to either failure to ligate the
sac at a high level enough, tearing of a friable sac,
slipped ligature at the neck of the sac, or a wound
infection. Direct hernia may be due to inherent
muscle weakness or injury to the posterior wall of
the inguinal canal. Today, there should be virtually
no mortality as the result of an operation for ingui-
nal hernia (Bisgaard et al. 2014; Ein et al. 2006).
List 1 Typical complications after hernia repair
Hematoma
Wound infection
Testicular atrophy
Vas injury
Iatrogenic ascent of the testis
Recurrence
Intestinal infarction
1009
74.2 Congenital Hydrocele
Hydrocele is a common finding in newborn boys.
Approximately 70% of the hydroceles are scro-
tal, 26% are hydroceles of the cord, and 4% are
combined hydroceles. In females, an equivalent
to hydrocele of the cord may occur as cysts of the
canal of Nuck. Sixty percent of the hydroceles
are right-sided, 33% are left-sided, and 7% are
bilateral (Ein et al. 2006). There is consensus that
asymptomatic scrotal hydroceles will usually
spontaneously disappear during the first 1–2 years
of life. Hydroceles that persist after 2 years of age
are unlikely to resolve and require operation (Lau
et al. 2007). In a survey of North American pedi-
atric surgeons’ practice, 42% repair the hydro-
cele if it persists at 1 year of age (Antonoff et al.
2005). Some surgeons chose to operate commu-
nicating hydroceles earlier.
74.2.1 Management
A hydrocele should be operated with a high liga-
tion of the processus vaginalis and emptying of
the fluid. The procedure is identical to the opera-
tion for an inguinal hernia (Figs. 74.1, 74.2, 74.3,
74.4, 74.5, 74.6 and 74.7).
74.3 Femoral Hernia
Femoral hernias are rarely seen in children. The
etiology is not clear. The diagnosis is based on
the observation of a groin swelling located under-
neath the external inguinal orifice, although this
location is easily missed because, unless the
bulge is visible upon examination, parents and
doctors will interpret its appearance as the
expression of an inguinal hernia. This explains
why 50% of these patients are mistakenly oper-
ated on for inguinal hernia and why, only when
the sac is not found, exploration of the femoral
area allows diagnosis and repair (Al-Shanafey
and Giacomantonio 1999; Lee and DuBois
2000).
1010 A. Svenningsson and T. Wester

74.3.1 Management

The operative approach for femoral hernia is ini-

tially identical to the more commonly used
approach for inguinal hernia. An inguinal skin
crease incision is made, and the subcutaneous
layers and fascia of Scarpa are opened in order to
expose the external oblique aponeurosis at the
level of the external inguinal ring. The aponeuro-
sis is incised longitudinally taking care to pre-
serve the ilioinguinal nerve (Fig.  74.10). The
inguinal canal is opened dorsally sectioning with
cautery the conjoined tendon and the fascia
­transversalis. The spermatic cord is retracted in
order to obtain access to the femoral region. The
sac is identified and delivered into the wound
avoiding damage to the femoral vein, which is in
Inferior epigastric
close contact with the sac laterally. It may be con- artery and vein
Femoral artery, vein
Femoral sac
venient to ligate and divide the inferior epigastric
vessels in order to expose the femoral area from Fig. 74.11  The femoral hernia sac is identified medially
behind (Fig. 74.11). The sac is opened to ensure to the femoral vein
that it has no contents and it is subsequently
suture-ligated with a fine stitch flush to the peri-
toneum. The femoral defect is then narrowed by
approximating the internal insertions of the
Cooper ligament and the inguinal ligament with
two or three fine nonabsorbable stitches taking
care of not compressing the femoral vessels. The
inguinal canal is reconstructed, and the superfi-

External oblique
Fig. 74.12  The sac is suture-ligated close to the perito-
aponeurosis neum, and the inguinal canal is reconstructed

cial layers and the skin are closed as in inguinal

hernias (Fig.  74.12). Femoral hernia repair can
also be accomplished by an infra-inguinal
approach.

74.4 Umbilical Hernia

Illoinguinal n.

External inguinal ring

Umbilical hernia is the result of the failure of clo-
Fig. 74.10  The external oblique aponeurosis is opened sure of the umbilical ring. The hernia sac pro-
laterally from the external ring trudes through the defect. Symptoms related to
74 Hernias 1011

the hernia are rare. Most umbilical hernias have a umbilical scar, tractioning it downward and fix-
tendency to resolve spontaneously. However, ing it to the subcutaneous layer in the midline
hernias with a fascial defect greater than 1.5–2 cm (Fig.  74.15). The wound is closed with several
have been considered less likely to close. Surgical
intervention is limited to umbilical hernias per-
sisting after 4 years of age. Occasionally, incar-
ceration of the omentum occurs, which requires
exploration (Zendejas et al. 2011).

74.4.1 Management

Umbilical hernia repair is carried out under gen-

eral anesthesia. A semicircular incision is made
in the skin crease immediately below the umbili-
cus (Fig. 74.13). The subcutaneous layers are dis-
sected in order to expose the hernia sac. By blunt
dissection, a plane is developed on both sides of
the sac, and the sac is encircled with a hemostat
and is divided (Fig. 74.14). A clamp is placed on
Fig. 74.14  The hernia sac is identified, mobilized, and
either side of the umbilicus defect for traction. divided
The defect is closed by interrupted 2–0 absorb-
able sutures. A stitch is used to invaginate the

Fig. 74.13  A semicircular incision is made in the skin Fig. 74.15  The defect is closed and a stich fixes the
crease immediately below the umbilicus umbilicus in the midline
1012 A. Svenningsson and T. Wester

interrupted sutures placed in the subcuticular

plane. A slightly compressive dressing is main-
tained for 24 h.

74.5 Epigastric Hernia

Epigastric hernia usually occurs in the midline of

the anterior abdominal wall. It is usually small
defects through which preperitoneal fat protrudes
and may cause pain. Epigastric hernias do not
resolve spontaneously (Coats et  al. 2000;
Bowling et al. 2017).

74.5.1 Management

Epigastric hernias are repaired when they are

prominent or when they are symptomatic. It is
important to mark the location of the defect
before anesthesia, because in the recumbent posi-
tion they are often impossible to palpate along
Fig. 74.17  The protruding fat is excised after a transfix-
the widened linea alba. A transverse incision is ation stich
made directly over the previously marked loca-
tion of the hernia (Fig. 74.16). The fat that pro-
trudes through the linea alba defect is excised
after a transfixation stitch (Fig. 74.17). The defect
in the linea alba is closed with interrupted 3–0

Fig. 74.18  The fascia defect is closed with interrupted

sutures. The skin is closed with a continuous subcuticular
suture

absorbable sutures. The skin is approximated

using subcuticular sutures (Fig. 74.18).

74.6 Conclusions

Inguinal hernia repair is one of the most com-

mon procedures in pediatric surgery. Usually,
Fig. 74.16  A transverse incision is made over the marked infants and children present with a typical bulge
hernia in the groin. Premature infants have a particu-
74 Hernias
larly high risk for incarceration, which makes it
important not to delay the repair. Laparoscopic
repair has been established as an alternative to
traditional open repair. Also, umbilical hernias
are common. In children without symptoms, the
repair can be delayed at least until 4 years of age.
A large proportion of umbilical hernias closes
spontaneously.
References
Al-Shanafey S, Giacomantonio M (1999) Femoral hernia
in children. J Pediatr Surg 34(7):1104–1106
Antonoff MB, Kreyskes NS, Saltzman DA (2005)
American Academy of Pediatrics section on sur-
gery hernia survey revisited. J Pediatr Surg
40(6):1009–1014
Audry G, Johanet S, Achrafi H, Lupold M, Gruner M
(1994) The risk of wound infection after inguinal inci-
sion in pediatric outpatient surgery. Eur J Pediatr Surg
4(2):87–89
Bisgaard T, Kehlet H, Oehlenschlager J, Rosenberg J
(2014) Acceptable nationwide outcome after paediat-
ric inguinal hernia repair. Hernia 18(3):325–331
Boocock GR, Todd PJ (1985) Inguinal hernias are com-
mon in preterm infants. Arch Dis Child 60(7):669–670
Bowling K, Hart N, Cox P, Srinivas G (2017) Management
of pediatric hernia. BMJ 359:j4484
Burgmeier C, Dreyhaupt J, Schier F (2014) Comparison
of inguinal hernia and asymptomatic patent processus
vaginalis in term and preterm infants. J Pediatr Surg
49(9):1416–1418
Chang SJ, Chen JYC, Hsu CK, Chuang FC, Yang SSD
(2016) The incidence of inguinal hernia and associ-
ated risk factors of incarceration in pediatric inguinal
hernia: a nation-wide longitudinal population-based
study. Hernia 20(4):559–563
Coats RD, Helikson MA, Burd RS (2000) Presentation
and management of epigastric hernias in children. J
Ped Surg 35(12):1754–1756
Coté CJ, Zaslavsky A, Downes JJ, Kurth CD, Welborn
LG, Warner LO, Malviya SV (1995) Postoperative
apnea in former preterm infants after inguinal her-
niorrhaphy. A combined analysis. Anesthesiology
82(4):809–822
Czeizel A, Gardonyi J (1979) A family study of congenital
inguinal hernia. Am J Med Genet 4(3):247–254
Davidson AJ, Morton NS, Arnup SJ, de Graaff JC, Disma
N, Withington DE, Frawley G, Hunt RW, Hardy P,
Khotcholava M, von Ungern Sternberg BS, Wilton N,
Tuo P, Salvo I, Ormond G, Stargatt R, Locatelli BG,
McCann ME (2015 Jul) General Anesthesia compared
to Spinal anesthesia (GAS) Consortium. Apnea after
awake regional and general anesthesia in infants: the
general anesthesia compared to spinal anesthesia
study--comparing apnea and neurodevelopmental out-
1013
comes, a randomized controlled trial. Anesthesiology
123(1):38–54
Dreuning K, Maat S, Twisk J, van Heurn E, Derikx J
(2019a) Laparoscopic versus open pediatric inguinal
hernia repair: state-of-the-art comparison and future
perspectives from a meta-analysis. Surg Endosc
33(10):3177–3191
Dreuning KMA, Ten Broeke CEM, Twisk JWR, Robben
SGF, van Rijn RR, Verbeke JIML, van Heurn LWE,
Derikx JPM (2019b) Diagnostic accuracy of preop-
erative ultrasonography in predicting contralateral
inguinal hernia in children: a systematic review and
meta-analysis. Eur Radiol 29(2):866–876
Dreuning KM, Barendsen RW, van Trotsenburg AP, Twisk
JW, Sleeboom C, van Heurn LE, Derikx JP (2020)
Inguinal hernia in girls: a retrospective analysis over
1000 patients. J Pediatr Surg 55(9):1908–1913
Ein SH, Njere I, Ein A (2006) Six thousand three hundred
sixty-one pediatric inguinal hernias: a 35-year review.
J Pediatr Surg 41(5):980–986
Fasching G, Höllwarth ME (1989) Risk of testicular
lesions following incarcerated inguinal hernia in
infants. Pediatr Surg Int 4:265–268
Given JP, Rubin SZ (1989) Occurrence of contralateral
inguinal hernia following unilateral repair in a pediat-
ric hospital. J Pediatr Surg 24(10):963–965
Grosfeld JL (1989) Current concepts in inguinal hernia
in infants and children. World J Surg 13(5):506–515
Grosfeld JL, Minnick K, Shedd F, West KW, Rescorla
FJ, Vane DW (1991) Inguinal hernia in children: fac-
tors affecting recurrence in 62 cases. J Pediart Surg
26(3):283–287
Hall NJ, Choi W, Pierro A, Eaton S (2012) Age-related
probability of contralateral processus vaginalis
patency in children with unilateral inguinal hernia.
Pediatr Surg Int 28:1085–1088
Harper RG, Garcia A, Sia C (1975) Inguinal hernia: a
common problem of premature infants weighing 1000
grams or less at birth. Pediatrics 56(1):112–115
Husmann DA, McLorie GA, Churchill BM et  al (1990)
Inguinal pathology and its association with classical
bladder exstrophy. J Pediatr Surg 25(3):332–334
Kokorowski PJ, Wang HH, Routh JC, Hubert KC, Nelson
CP (2014) Evaluation of the contralateral inguinal ring
in clinically unilateral inguinal hernia: a systematic
review and meta-analysis. Hernia 18(3):311–324
Lau WY (2002) History of treatment of groin hernia.
World J Surg 26(6):748–759
Lau ST, Lee Y-H, Caty MG (2007) Current manage-
ment of hernias and hydroceles. Sem Pediatr Surg
16(1):50–57
Lee SL, DuBois JJ (2000) Laparoscopic diagnosis and
repair of pediatric femoral hernia: initial experience of
four cases. Surg Endosc 14(12):1110–1113
Misra D, Hewitt G, Potts SR, Brown S, Boston VE (1994)
Inguinal herniotomy in young infants, with emphasis
on premature infants. J Pediatr Surg 29(11):1496–1498
Muensterer OJ, Gianocolo E (2019 Aug) Contralateral
processus closure to prevent metachronous inguinal
hernia: a systematic review. Int J Surg 68:11–19
1014
Nakayama DK, Rowe MI (1989) Inguinal hernia and the
acute scrotum in infants and children. Pediatr Rev
11(3):87–93
Olesen CS, Andresen K, Öberg S, Rosenberg J (2019a)
Laparoscopic versus open repair of groin hernias in
children: a systematic review and meta-analysis. Surg
Endosc 33(7):2050–2060
Olesen CS, Mortensen LQ, Öberg S, Rosenberg J (2019b)
Risk of incarceration in children with inguinal hernia:
a systematic review. Hernia 23(2):245–254
Ozdemir T, Arikan A (2013) Postoperative apnea after
inguinal hernia repair in formerly premature infants:
impact of gestational age, postconceptional age and
comorbidities. Peditr Surg Int 29(8):801–804
Phelps S, Agrawal M (1997) Morbidity after neonatal
inguinal herniotomy. J Pediatr Surg 32(3):445–447
Powell TG, Hallows JA, Cooke RW, Pharoah PO (1986)
Why do so small infants develop an inguinal hernia?
Arch Dis Child 61(10):991–995
Puri P, Guiney EJ, O’Donnell B. (1984) Inguinal hernia
in infants: the fate of testis following incarceration. J
Pediatr Surg 19:44–46
Rescorla FJ, Grosfeld JL (1984) Inguinal hernia repair in
the perinatal period and early infancy: clinical consid-
erations. J Pediatr Surg 19(6):832–837
Ron O, Eaton S, Pierro A (2007 Jul) Systematic review
of the risk of developing a metachronous contralateral
inguinal hernia in children. Br J Surg 94(7):804–811
Rowe MI, Clatworthy HW (1970) Incarcerated and stran-
gulated hernias in children. A statistical study of high-­
risk factors. Arch Surg 101(2):136–139
Rowe MI, Marchildon MB (1981) Inguinal hernia and
hydrocele in infants and children. Surg Clin North Am
61(5):1137–1145
Sarpel U, Palmer SK, Dolgin SE (2005) The incidence of
complete androgen insensitivity in girls with inguinal
hernias and assessment of screening by vaginal length
measurement. J Pediatr Surg 2005(40):133–137
A. Svenningsson and T. Wester
Schmitt M, Peiffert B, de Miscault G, de Miscault G,
Barthelme H, Poussot D, Andre M (1987) Les compli-
cations de la hernie inguinale chez le garçon [compli-
cations of inguinal hernia in children]. Complications
des hernies inguinales chez lénfant. Chir Pediatr
28(4–5):193–196
Schwöbel MG, Schramm H, Gitzelmann CA (1999) The
infantile inguinal hernia – a bilateral disease? Pediatr
Surg Int 15:115–118
Snyder CL (2010) Inguinal hernias and hydroceles. In:
GW AKWH III, Murphy JP (eds) Ashcraft’s pediat-
ric surgery, 5th edn. Elsevier Saunders, Philadelphia,
pp 669–675
Surana R, Puri P (1993) Is contralateral exploration nec-
essary in infants with unilateral inguinal hernia? J
Pediatr Surg 28(8):1026–1027
Surana R, Puri P (1994) Iatrogenic ascent of the testis:
an under-recognized complication of iguinal hernia
operation in children. Br J Urol 73(5):580–581
Tackett LD, Breuer CK, Luks FI, Caldamone AA,
Breuer JG, DeLuca FG, Caesar RE, Efthemiou E,
Wesselhoeft CW Jr (1999) Incidence of contralateral
inguinal hernia: a prospective analysis. J Pediatr Surg
34(5):687–688
Wester T (2009) Hernias. In: Puri P, Höllwarth ME
(eds) Pediatric surgery. Springer, Berlin, Heidelberg,
pp 497–506
Zani A, Eaton S, Hoellwarth M, Puri P, Tovar J, Fasching
G, Bagolan P, Lukac M, Wijnen R, Kuebler J,
Cecchetto G, Rintala R, Pierro A (2014) Management
of pediatric inguinal hernias in the era of laparoscopy:
results of an international survey. Eur J Pediatr Surg
24(1):9–13
Zendejas B, Kuchena A, Okendi EO, Lohse CM, Moir CR,
Ishitani MB, Potter DD, Farley DR, Zarroug AE (2011)
Fifty-three-year experience with pediatric umbilical
hernia repairs. J Pediatr Surg 46(11):2151–2156
 Short Bowel Syndrome
75
Michael E. Höllwarth

75.1 Introduction and Walker Smith (1983), it equals 30% of normal

small bowel length in full-term babies. In prema-
The term short bowel has been defined by ture babies, the term also corresponds to 30% of
Rickham in 1967 as a small intestinal remnant of the total calculated length for a given gestational
75  cm in the newborn. According to Toulukian age (Fig. 75.1). According to a more recent publi-

Fig. 75.1 Normal GESTATIONAL AGE (weeks)

intestinal length and the 400 GROUP A GROUP B GROUP C
ranges in premature
babies post mortem
(From Touloukian and
Smith. J Pediatr Surg
1983, With permission)
TOTAL INTESTINAL LENGTH (cm)

MEAN MEAN
300 ± 1 SD ± 2 SD

200

100

19 20 23 26 29 32 35 38 41

M. E. Höllwarth (*)
University Clinic of Paediatric and Adolescent
Surgery, Medical University of Graz, Graz, Austria
e-mail: Michael.hoellwarth@medunigraz.at

© Springer Nature Switzerland AG 2023 1015

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_75
1016
cation measuring intestinal length in  vivo, the
small bowel equals 100 cm at GA of 27–29 weeks,
157.4 cm at 39–40 weeks, and 239.2 cm between
0 and 6 months (Struijs et al. 2009). However, the
antimesenteric measurement of intestinal length
during surgical interventions yields highly vari-
able results, due to the enormous contractibility of
the bowel in length and diameter (Hoellwarth
2014). Therefore, a more functional description of
the term short bowel syndrome (SBS) is preferred
by most authors defining a state of significant
maldigestion and malabsorption due to an exten-
sive loss of functional absorptive intestinal sur-
face area as intestinal failure (IF) (D’Antiga and
Goulet 2013). Not included in this chapter are
mucosal enteropathy cases of IF with an other-
wise normal intestinal length.
75.2 History
In the past, an extensive loss of small bowel in
newborns and babies used to be a catastrophic
event, which was nearly always followed by
malnutrition and death. According to Potts in
1955, long-term survival was possible for new-
borns if 40 cm of small bowel remains, but mor-
tality reached more than 50% in babies with
shorter small bowel and the loss of the ileocecal
valve (ICV) (Potts). Significant progress was
achieved by introducing sophisticated paren-
teral nutrition in the early 1970s. Wilmore found
that survival was possible with 15  cm jejuno-
ileum with ICV or with 38  cm jejuno-ileum
without ICV. It must be realized that little prog-
ress has been achieved since then: today, long-
term survival on enteral nutrition is possible in
infants with as little as 10 cm jejuno-ileum with
ileocecal valve (5% of total) or with 25  cm
jejuno-ileum without ileocecal valve (10% of
total). However, home parenteral nutrition
(HPN) is one option with good long-term results
(Colomb et al. 2007). The other option today is
intestinal or multi-organ transplantation; how-
ever, 10-year survival rates of grafts (25%) and
patient survival (50–60%) are significantly
lower when compared with a 10-year HPN
(Jeejeebhoy 2008).
M. E. Höllwarth
75.3 Incidence and Etiology
The prevalence of SBS has been increasing over
the past two decades because of the enormous
progress in intensive care medicine. Today, pri-
mary survival rates are significantly higher in
babies with short gut caused either due to severe
acquired intestinal diseases, such as necrotizing
enterocolitis and volvulus, or resulting after con-
genital malformations, such as multiple intestinal
atresias or intrauterine volvulus with gastroschisis.
A single population-based Canadian study found
that the incidence of neonatal SBS was
24.5/100,000 live birth. The incidence in babies
born before 37  week gestation was much higher
when compared to term newborns (353.7 vs 3.5
per 100,000) (Wales et al. 2004). The incidence of
SBS in very low birth weight babies was 0.7% and
1.1% in extremely low birth weight babies (Cole
et al. 2008). A multicenter study from Italy showed
an incidence of 0.1% in all live birth and 0.5%
among pediatric ICU admissions (Salvia et  al.
2008). It has been estimated that severe SBS cases
remaining dependent on long-term home paren-
teral nutritional support amount to 2 new patients
per one million of population/year (including all
age groups) (Mughal and Irving 1986).
In infancy, most cases of SBS occur in the
neonatal age group. There are many different
causes, which can be divided into three major
groups. The first of these consists of neonates
with prenatally acquired anomalies, character-
ized by a vascular injury to the intestinal tract in
utero, e.g., multiple intestinal atresias or gastros-
chisis with intrauterine volvulus of the prolapsed
bowel. The second group comprises postnatally
acquired diseases necessitating extensive intesti-
nal resection, e.g., necrotizing enterocolitis or
volvulus. A third group is defined by a geneti-
cally determined deficiency, e.g., in the embryo-
logical small bowel anlage causing a “true”
congenital short bowel, or in its innervation, such
as total intestinal aganglionosis (Table  75.1).
Extensive intestinal resection leading to SBS is
rarely required in older children. Indications may
be severe Crohn’s disease, traumatic avulsion of
the intestinal tract, and/or traumatic or iatrogenic
mesenteric artery lesions.
75  Short Bowel Syndrome
Table 75.1  Causes of SBS in over 100 newborn babies
collected from the literature
Necrotizing enterocolitis 36%
Volvulus 19%
Intestinal atresia 21%
Gastroschisis and atresia 10%
Hirschsprung’s disease 7%
Trauma 1%
Others 6%
75.4 Physiology
For an overview, see “Physiology and
Pathophysiology of Digestion” (Granger et  al.
2018). The digestive process already starts in the
oral cavity during chewing of the foodstuff. Up to
75% of the starch can be digested to disaccha-
rides by salivary amylase.
The digestion continues in the stomach. The
pH of the gastric juice is usually between 1 and 2
but is increased and most pronounced in mainly
milk-fed infants for 1 to 2 hours. Lipase, which is
secreted by salivary glands in the oral cavity, is
active at low pH. Thus, short-chain fatty acids are
separated from triglycerides and already absorbed
by gastric mucosa. Protein digestion is started by
pepsin in the stomach, too. Liquids pass easily
the pylorus, while solids are mixed and dispersed
in the stomach to fine emulsions at the end.
Finally, the small intestine receives a gastric
chyme consisting of a mixture of small solutes
and emulsion particles. In normal human indi-
viduals, the three-dimensional anatomy of intes-
tinal mucosa into the plicae, villi with microvilli,
and crypts results in a surface area in adults up to
400 m2. Water secretion or absorption in the duo-
denum and upper jejunum results in a rapid isoto-
nicity of the ingested food. As mentioned above,
carbohydrate and protein digestion begins to a
small extent in the oral cavity and in the stomach
and proceeds in the duodenum and upper jeju-
num. Finally, 80% of carbohydrates and most of
the proteins are absorbed in the small bowel.
More complex carbohydrates and protein-rich
secretions, mucus and epithelia reach the colon
and are fermented by bacteria to absorbable
short-chain fatty acids (SFCA) (Nordgaard et al.
1994; Jeppesen and Mortensen 1999). In con-
1017
trast, lipid absorption occurs along the entire
small bowel. Finally, most of the lipids, including
the fat-soluble vitamins, are digested and
absorbed by pancreatic lipase and co-lipase, bile,
and liver-derived bile salts. The latter are capable
of forming the so-called micelles with the hydro-
phobic long-chain fatty acids in the inner com-
partment. They are absorbed by passive diffusion
similarly as the fat-soluble vitamins (A, D, E, K).
In contrast medium- and short-chain fatty acids
are soluble in the aqueous phase and are absorbed
directly by the enterocytes. The colon not only
absorbs water and electrolytes but most of the
SCFA which are generated by bacteria while fer-
menting undigested carbohydrates and proteins.
SCFA, such as acetic, butyric, and propionic
acids, are a significant source of energy espe-
cially for SBS patients (Norgaard et  al. 1996;
Goulet et al. 2009).
75.5 Pathophysiology
Resection of the jejunum only induces a transient
reduction of absorption of all nutrients due to an
enormous adaptive capacity of the ileum, the
intact enterohepatic circulation of bile salts, and
the preserved absorption of vitamin B12. A mini-
mum of 10–15 cm of ileum and the presence of
the ileocecal valve are required so that passage
time allows at least partial absorption of bile
salts, vitamins, and nutrients. Therefore, despite
different opinions in the literature, recent publi-
cations clearly show that the presence of an ICV
has a beneficial effect on the outcome (Goulet
et al. 2005; Demehri et al. 2015).
More serious consequences result from resec-
tion of the ileum. The ileum is responsible for
absorbing all the water that has been secreted in
the upper intestinal tract following a hypertonic
meal. If the ileum is resected, this water spills
over into the colon. Although the colon can
increase water and solute absorption up to 400%
of normal, there is a limit to the excess that may
be reabsorbed (Debongnie and Phillips 1978).
Another consequence of ileal resection is the life-­
long malabsorption of vitamin B12, due to the loss
of site-specific receptors. Furthermore, the ileum
1018
is the major site of bile acid absorption. Non-­
absorbed intestinal contents including bile acids
spill over into the colon and may cause signifi-
cant diarrhea. Furthermore, the loss of the ileum
results in a depletion of the bile salt pool, leading
to a disturbed micelle formation and malabsorp-
tion of fat and fat-soluble vitamins (A, E, D, K).
As a consequence, hydrolyzed fatty acids and
bile salts reach the colon. The former bind to cal-
cium to form calcium stearate, preventing cal-
cium absorption. Subsequently, oxalate—which
is normally bound to intraluminal calcium and
then excreted with the feces—gets absorbed in
increased amounts, leading to oxaluria and
­eventually kidney stone formation. Similarly, due
to the consumption of enteric calcium by undi-
gested fatty acids, the deconjugated enteric bili-
rubin remains in solution, resulting in a significant
enterohepatic circulation of unconjugated biliru-
bin and increased secretion rates of bilirubin in
the bile, responsible for biliary sludge formation
and gallstone diseases in SBS patients (Brink
et al. 1996). Furthermore, resection of the ileum
results in a loss of the “ileal brake.” Consequently,
nutrients spend less time in the stomach and in
the upper intestinal segments. Due to the shorter
contact, digestion and absorption are reduced,
and an unusual large fluid load imposed onto the
distal remnants.
Resection of the colon causes an increased
number of fluidly stools and loss of water and
electrolytes. Although parts of the colon can
increase water and solute absorption up to 400%
of normal, short remnants of the colon cannot
compensate for that (Debongnie and Phillips
1978). Resection of most of the colon also
reduces the absorption of SCFA significantly.
75.6 Citrulline
Clinical experience has shown that the length of
intestinal remnants is a weak parameter to esti-
mate whether a patient can be finally weaned
from parenteral nutrition. Recently, it has been
shown that the amino acid, citrulline, is a useful
parameter to estimate the number of functional
M. E. Höllwarth
enterocytes. Citrulline is produced in the entero-
cytes from glutamine or ornithine and glutamate.
It is released from the enterocytes and converted
unmetabolized to arginine in the kidney. Citrulline
correlates with a number of functional entero-
cytes (Peterson and Kerner 2012). Thus, citrul-
line is a powerful marker for the active intestinal
length in patients with SBS. In adults, the thresh-
old of 20  mmol/L distinguishes patients with a
transient IF from those with a permanent IF
(Crenn et al. 2008). In children, a cutoff value of
12–15  mmol/L or less shows those who cannot
be weaned from PN (Chiba et  al. 2017). Our
experience with newborns with SBS has shown
that even babies with primarily <10 mmol/L can
be weaned of PN with final values of citrulline
around 30 mmol/L, while other babies with pri-
marily 17 mmol/L showed a decline over 3 years
to 8  mmol/L and remained on home parenteral
nutrition (HPN).
75.7 Intestinal Adaptation
Intestinal adaptation characterizes the patho-
physiology which follows intensive intestinal
resection, and by which 90% of babies with SBS
do finally reach a normal life on entirely oral
nutrition. Adaptation is characterized by an early
increase of blood flow to the intestinal remnants
and by long-term stimulation of intestinal growth,
which enormously enlarges the absorptive sur-
face area. The latter includes an increase in blood
flow, villous height, crypt depth, intestinal length,
thickness, and diameter (Hoellwarth et al. 1988b).
The latter response is more pronounced in the
ileum when compared to the jejunum, which may
result from the fact that the jejunum faces the
usual type of foodstuff, while the ileum mucosa
is confronted with a different composition of
ingested food, thereby enhancing mucosal adap-
tation. As a consequence, morphological changes
in regard to villus height and crypt depth are sig-
nificantly higher in the ileum after resection of
the jejunum and vice versa. Recent experimental
studies show that the angiotensin-converting
enzyme is associated with the morphological
75  Short Bowel Syndrome
changes after a significant loss of small bowel
(Haxhija et  al. 2008). Experimental evidence
suggests that the colon undergoes a significant
adaptation process in SBS (Goulet et  al. 2019;
Diamanti and Basso 2012). Additionally, water
and solute absorption is enhanced in the colon;
colonic bacteria ferment undigested carbohy-
drates and proteins into SCFA.  As mentioned
above, they act as important energy providers and
apparently, as additional promoters of adaptation
(Joly et al. 2009).
The precise mechanisms of adaptation are not
clear, but intraluminal nutrients and endogenous
intestinal secretions stimulate growth. In general,
the higher the workload required for digestion
and absorption, the more potent the stimulus for
adaptation is.
In response to the nutrients and secretions, a
large number of trophic polypeptides and other
mediators are secreted. Over the years, some of
them have attracted attention regarding their pos-
sible clinical value in promoting adaptation in
SBS patients. First, gastrin was demonstrated to
exhibit trophic effects on the small bowel
(Johnson 1976). Later, enteroglucagon has been
shown to stimulate the adaptive response on the
intestinal tract in animal experiments and humans
(Bloom and Polak 1982). Since monoclonal anti-
bodies failed to block this trophic effect, recently
precursors of enteroglucagon, such as glucagon-­
like peptide-2 (GLP-2), are considered to be
responsible for the intestinal effects. GLP-2 is
secreted by L-cells in the distal ileum and proxi-
mal colon in response to nutrients. GLP-2
improves digestion and absorption and the barrier
function of the intestinal mucosa (Brubaker
2018). Experimental and clinical studies have
shown a significant role of GLP-2 in the adapta-
tion process (Sigalet et al. 2017; Lai et al. 2017).
Limited studies have shown an increase in weight
gain and energy absorption with GLP-2 adminis-
tration after ileocecal resection in humans.
Recent clinical trials with the long-acting analogs
of GLP-2, Teduglutide and Glepaglutide have
shown beneficial effects on patients with SBS
(Naimi et al. 2019; Jeppesen et al. 2018). Insulin-­
like growth factor-1 (IGF-1) and epidermal
1019
growth factor (EGF) are produced by the intesti-
nal subepithelial myofibroblasts and liver and are
found in platelets, macrophages, and saliva. Both
stimulate after exogeneous administration intes-
tinal mucosal growth in experiments in close
connection with GLP-2.
Lately, human growth hormone (GH), in
combination with the enterotrophic hormones
epidermal growth factor (EGF) and with insulin-
like growth factor-1 (IGF-1), has been shown to
regulate small intestinal growth and adaptation.
IGF-1 receptors have been identified in all seg-
ments of the gastrointestinal tract, and IGF-1
stimulates DNA and RNA synthesis and cellular
amino acid uptake. The endogenous GH-IGF-1
system is supposed to be an important regulator
of small intestinal growth and adaptation
(McMellen et al. 2010). Among the amino acids,
glutamine (GL) plays a key role in the mainte-
nance of intestinal structure and function by pro-
viding the energy required by cells with a rapid
turnover, such as macrophages and enterocytes.
After major trauma or in chronic catabolic states,
patients benefit from GL supplementation. It has
been shown that human growth hormone
increases glutamine uptake after intestinal resec-
tion, supporting the evidence that glutamine
exerts trophic effects in the small intestine and
colon of patients with SBS.  However, more
research is needed, since other studies could not
confirm the role for GL or GH as trophic agents
for the intestinal tract. Thus, it remains contro-
versial, whether or not GL alone or supplemen-
tation of GL and GH, or other trophic factors or
hormones, should be recommended for the treat-
ment of SBS (Kim and Kim 2017). There is a
need for more controlled clinical trials in order
to elaborate which agents have a lasting impact
on intestinal adaptation and prove beneficial for
the long term.
Prostaglandin (Pg) E2 and polyamines have
also been shown to stimulate cell proliferation in
animal experiments by increasing blood flow and
DNA synthesis (Hoellwarth et  al. 1988a).
Experimental evidence exists suggesting that tes-
tosterone enhances adaptation after small bowel
resection in cats (Pul et al. 1991).
1020
75.8 Nutritional Therapy
The clinical course of SBS can be divided into
three stages which require individual manage-
ment: the acute phase, the adaptation phase, and
the maintenance phase. The acute phase—the
duration of which depends on the underlying dis-
ease—is characterized by insufficient absorption,
dysmotility, diarrhea, and gastric hypersecretion
and hypergastremia. Therapeutic measures are
guided by the underlying disease and the severity
of the illness of the patient. They are primarily
aimed at restoring and maintaining fluid, electro-
lyte, and acid-base equilibrium. The following
adaptation phase is slower in humans compared
to animals and often takes more than a year to
reach its peak—unrelated to the absolute length
of intestinal remnants. Treatment strategies dur-
ing this phase include parenteral nutrition and
carefully balanced, stepwise increasing enteral
feeding. Diligence and patience are necessary
prerequisites to reach the maximum of what can
be achieved in a given patient. Finally, the main-
tenance phase follows in patients with a constant
malabsorption rate of 30% or more, during which
a surplus of enteral calories has to be consumed
daily, supplemented by vitamins, trace elements,
and minerals, adapted to the individual demands.
Balanced fluid and electrolyte solutions must
cover the basic demands and replace losses via a
nasogastric tube, from an enterostomy, or due to
excessive diarrhea. Existing nutritional deficien-
cies should be restored by an adequate supple-
mentation of carbohydrates, protein, and fat.
Calorie intake in children needs to be continu-
ously advanced to accommodate the increasing
demands of the growing organism. Growth in
weight, height, and head circumference are the
basic parameters to ensure adequacy of paren-
teral nutritional solutions and enteral feeding. In
newborns, an infusion running continuously over
24 h is most appropriate. In contrast, when infants
take at least 20% of their total calorie require-
ments by the enteral route, intermittent parenteral
feeding can be attempted, and the intervals
increased as long as serum glucose levels are
maintained. In the older age groups and in adults,
a 12-h infusion time is usually well tolerated.
M. E. Höllwarth
Finally, parenteral nutrition can be restricted to
the nocturnal period. The possibility of HPN con-
siderably improves the quality of life for the
patients and their families allowing a more nor-
mal lifestyle and substantially reduces hospital
costs. However, long-term laboratory monitoring
is essential (Auber Smith et al. 2018).
The institution of multi-expert intestinal reha-
bilitation programs was one of the important
achievements resulting in significantly better
results in all patients with SBS and IF (Merritt
et  al. 2017; Duggan and Jaksic 2017; Goulet
et al. 2019; Javid et al. 2018).
Enteral feeding is usually introduced by con-
tinuous infusion via a gastric or jejunal tube as
soon as the intestinal remnants resume normal
motility. This technique avoids gastric distension
and offers a constant load of carrier proteins to
the microvilli. Elemental diets are started in a low
concentration and are slowly increased to
0.67  kcal/ml in infants and 1.0  kcal/ml in chil-
dren and adults. High carbohydrate content rep-
resents a substantial osmotic load and may cause
diarrhea. A stool volume, which increases by
>50% compared to baseline, is an indication to
reduce the amount and/or concentration of enteral
feedings. Stool samples positive for reducing
substances also suggest that enteral feeding
should not be advanced and carbohydrate uptake
should be reduced. In infants with normal colon
length, a decrease of stool pH below 5.5 signals
carbohydrate malabsorption. Protein hydrolysate
diets are beneficial due to their higher content of
di- and tripeptides. In infants up to 6 months old,
protein-containing diets might cause a protein-­
sensitive enteritis/allergy. Therefore, amino acid
formulas are preferred. Medium-chain triglycer-
ides are water soluble and can be absorbed with-
out the need for bile acid micelle formation, but
their efficiency in enhancing the adaptation pro-
cess is far lower in comparison to long-chain
fatty acids. Therefore, elemental diets in pediat-
rics contain both medium- and long-chain fatty
acids, the mixture probably being the most effica-
cious for stimulation of adaptation. As soon as
the neonate’s condition improves, oral feeding of
small amounts of breast milk three or four times
daily should be attempted so that the child learns
75  Short Bowel Syndrome
to suck, taste, chew, and swallow properly with
the aim of avoiding the problem of total refusal of
any oral intake, not infrequently encountered
after long-term total parenteral nutrition (TPN)
and enteral tube feeding.
Blood levels of vitamins, trace elements, and
minerals must be checked regularly, and any defi-
cient substance has to be supplemented either
orally or parenterally as appropriate. Dietary
modifications are recommended depending on
the individual situations. Patients suffering from
SBS with the colon intact benefit from a high
fiber intake because the colon is capable of fer-
menting carbohydrates. On the contrary, patients
who have had the large bowel removed or
excluded may fare better with a diet rich in fat
providing a high energy concentration and a rela-
tively low osmotic load. In animal experiments,
triglycerides with highly unsaturated long-chain
fatty acids such as menhaden oil have exerted
more trophic effects on the intestinal remnants
after resection than other long-chain fatty acid-­
containing oils. In addition, recent studies in rats
have shown that SCFA (w-3 fatty acids) supple-
mentation of TPN enhances morphological and
functional aspects of adaptation. Fish oil- and
olive oil-based lipid emulsions have nutritional
advantages over soybean oil emulsions (Goulet
and Lambe 2017; Dai et al. 2016).
Nutritional strategy
Enteral
First: Protein hydrolysates such as Pregestimil and
Pregumin 15% (later SOM)
Later: Human milk (immunoglobulin A, nucleotides,
EGF, GH, GL, and acid-resistant lipase)
 Carbohydrates: Rice gruel 1%, 2%
 Fat: Mixture of MCT (fish/olive oil 1%) and LCT
 Prebiotics: Soluble fiber
 Probiotics: Acidophilus, Bifidobacterium
Parenteral
 Standard solutions: Cycling over night as soon as
possible
 Vitamins and trace elements
 50% ω-3 fatty acids
Overnight, one-bottle nutrition avoids pro-
longed hospitalization and allows the children to
live at home with their family (HPN). It is indi-
1021
cated as soon as the patient is in a stable situation
receiving a mixture of oral and parenteral ther-
apy. Again, a specialized team is responsible for
the controls and the needed adjustments in ther-
apy. HPN is confirmed as the primary treatment
for IF.  Patients with preserved colon have a
higher chance to be weaned from HPN (Pironi
et al. 2012).
75.9 Pharmacologic Supplements
Gastric hyperchlorhydria during the early phase,
after extensive loss of intestine, can be sup-
pressed successfully either with H2 receptor
antagonists or with proton pump inhibitors,
thereby improving absorption and reducing high-­
output diarrhea. Gastric hypersecretion may
inactivate pancreatic secretions. In these cases,
addition of oral pancreatic enzymes, such as
Creon, may be useful.
Surplus of calcium is needed in patients with-
out ileum because the bile acid pool will be
depleted. Undigested fatty acids and bile salt
reach the colon, binding calcium. Thus, oxalate
and bilirubin, which are normally bound to cal-
cium and excreted, will be absorbed, leading ulti-
mately to urinary and biliary stones. In patients
with ileal resection but an intact colon, urinary
oxalate levels should be monitored. Dietary oxa-
late restriction is to be recommended in patients
with high levels of oxaluria.
Rapid intestinal transit can be slowed by opi-
oid medication. In this group, loperamide has
proved effective and safe to use in the pediatric
age group together with dietary fibers even over a
long period of time. Octreotide acetate, a soma-
tostatin analogue, which essentially inhibits all
exocrine and endocrine gastrointestinal secre-
tions, is apt to improve the quality of life in
patients with predominantly secretory losses and
has not been uniformly recommended (Sukhotnik
et al. 2002).
Cholestyramine binds bile acids and prevents
choleretic diarrhea induced by an excess of bile
salts in the colon. Thus, while reducing diarrhea,
it may increase steatorrhea.
1022 M. E. Höllwarth

75.9.1 Hormonal Treatments transport was achieved and increased absorption

documented (Panis et  al. 1997). A more recent
Today, the focus in regard to clinical hormonal report on 38 adult patients showed a complete
therapy is mainly directed at GLP-2. As discussed weaning from PN in 17 individuals and a signifi-
above, GLP-2 has been shown in experiments to cant reduction in PN dependency in 7 patients
stimulate adaptive growth after intestinal resec- (Beyer-Berjot et  al. 2012). Intestinal valves and
tion. Therefore, it has been used successfully in sphincters have been created in a few humans,
children with IF (Sigalet et  al. 2017). A long-­ mostly children, with doubtful results. Colon
acting GLP-2analogue, Teduglutide, for daily interposition in the small bowel, more proximally
subcutaneous injections, has been used success- in the isoperistaltic direction, or in an antiperistal-
fully in pediatric and adult patients (O’Keefe tic manner distally (Fig. 75.3). Experimental and
et  al. 2013; Carter et  al. 2017; Jeppesen et  al. clinical evidence show that the isoperistaltic inter-
2018). Recently, Glepaglutide, a new long-acting posed colon delays the delivery of foodstuff into
GLP-2 analogue, has been shown to improve the jejunum and have shown absorptive capacities
hepatic excretory function in patients with IF but (Lloyd 1978; Kono et al. 2001). Personal experi-
to increase liver stiffness at the same time (Naimi ence with three extreme short bowel newborns
et al. 2019). (<15  cm of jejunum without ileum) resulted in
weaning from PN in two of them.
Dilated intestinal loops with inefficient peri-
75.10 Surgical Therapy stalsis and stagnant chyme are a common prob-
lem in patients with SBS, either as a consequence
A variety of surgical procedures slowing intesti- of the underlying pathology, e.g., remnants after
nal transit time have been invented and applied to multiple atresias, or following initially effective
humans. Reversed (antiperistaltic) intestinal seg- adaptive growth that finally overshoots to result
ments (Fig.  75.2) have been used in over 40 in large dilated bowel segments with insufficient
patients, and in most of them, a delay of chyme and undirected motility. Longitudinal division

Fig. 75.2  Reversed intestinal segment aimed to reduce Syndrome in Pediatric Surgery Atlas Series, eds P. Puri, M
intestinal transit time. Appropriate length in newborn Höllwarth, Springer 2019)
babies about 3  cm (from Höllwarth M.  Short Bowel
75  Short Bowel Syndrome
Fig. 75.3  Colon interposition can be used in an isoperi-
staltic manner to increase absorption and in the antiperi-
staltic manner to decrease transit time. The method has
consists of constructing the dilated part into two
separate segments, each comprising one half of
the circumference (Bianchi 1980). Both seg-
ments remain viable because the mesenteric ves-
sels divide extramurally into branches supplying
either side of the bowel separately. The two
halves are then refashioned into tubes of normal
intestinal diameter, which are lined up in the iso-
peristaltic direction and anastomosed one to the
other, yielding twice the length of the original
part (Fig. 75.4). Although this technique has been
used in a larger number of patients, it has only
proved successful when performed in a later
stage of the disease, on the so-called self-selected
survivors, i.e., patients in stable general condition
free of other severe complications such as liver
insufficiency (Bianchi 1999). Serial transverse
enteroplasty (STEP) (Fig.  75.5) is a procedure
1023
been used only in a few patients (from Höllwarth M.:
Short Bowel Syndrome in Pediatric Surgery Atlas Series,
eds P. Puri, M Höllwarth, Springer 2019)
that gained significant attention as a method
refashioning dilated intestinal loops, thereby
improving peristalsis and motility. It is techni-
cally much easier when compared to the Bianchi
method. First experiences showed that the com-
plication rate of the procedure is low and early
outcome was encouraging. Long-term results
show excellent results with the STEP procedure,
in collaboration with the multidisciplinary short
bowel team (Greig et al. 2019).
Intestinal refashioning can be achieved surgi-
cally by tailoring the antimesenteric side of a
dilated loop, either by resection of abundant
wall—provided that enough absorptive area
remains available and stasis is the only prob-
lem—or by infolding the excessive part of the
intestinal circumference in a longitudinal way
(Fig. 75.6).
1024 M. E. Höllwarth

Fig. 75.4  Bianchi’s method of intestinal lengthening Syndrome in Pediatric Surgery Atlas Series, eds P. Puri, M
aimed to refashion dilated loop with insufficient peristal- Höllwarth, Springer 2019)
sis and stagnant chyme (from Höllwarth M. Short Bowel

Fig. 75.5  The effects of serial transverse enteroplasty (STEP) is aimed to improve insufficient peristalsis and digestion
(from Höllwarth M. Short Bowel Syndrome in Pediatric Surgery Atlas Series, eds P. Puri, M Höllwarth, Springer 2019)

Intestinal transplantation (TPx) is the most
effective method to increase the absorptive area.
Indication for isolated intestinal TPx is given in
infants with very little or no small bowel at all,
who are expected to be dependent on TPN for
their lives and/or failure of venous access after
multiple central venous lines. Indication for com-
bined TPx is an intestinal failure together with
progressive liver failure and recurrent sepsis. In
the past, the results of intestinal TPx have been
poor, mainly due to high rejection rate. Recently,
significant progress has been achieved by the
introduction of new immunosuppressive agents
(tacrolimus, everolimus, OKT 3) and induction
therapy (daclizumab, Thymoglobulin). A recent
clinical study from Paris shows that patient sur-
vival with liver-containing graft is 60% at
18  years and 46% with SBTPx alone. Recent
graft survival at 5/10 y was 44% and 31% for
liver-containing graft and 57% and 44% for
75  Short Bowel Syndrome 1025

Fig. 75.6 Circum-
scribed dilated intestinal
loop can be refashioned
by infolding the
excessive part of the gut
without losing the
absorptive area (from
Höllwarth M. Short
Bowel Syndrome in
Pediatric Surgery Atlas
Series, eds P. Puri, M
Höllwarth, Springer
2019)

SBTPx. The authors conclude that TPx remains a sion solution has been virtually eliminated by
difficult procedure and re-transplantation even developing detailed practice guidelines for asep-
more so (Lacaille et al. 2017). Tissue engineering tic care. However, bacteria may enter via the exit
is experimentally focused on creating small intes- side and expand along the central venous line
tine substitutes with digestive and absorptive into the circulation. Colonization of the catheter,
functions (Grant et  al. 2015; Shirafkan et  al. especially by Staphylococcus epidermidis, is
2016). highly resistant to antibiotic treatment as long as
the foreign body is not removed. The incidence
of this problem has been greatly reduced by using
75.11 Complications special ports, which are implanted subcutane-
ously for patients on home parenteral nutrition
Despite the progress in intensive care medicine and blocking the catheter either with ethanol or
and long-term parenteral nutrition, many compli- taurolidine (Merras-Salmio et al. 2018).
cations do occur in patients with SBS, some of
which are life threatening. Among the most
important ones are central venous catheter-­ 75.11.2 Bacterial Overgrowth
related problems, liver failure, and bacterial over-
growth with bacterial translocation. In contrast, endogenous infections caused by
continuous or intermittent bacteremia from a dis-
tant focus impose a significant problem and have
75.11.1 Central Venous Line gained increased attention in cases of gut-derived
Infection sepsis and central venous line infections. Children
with SBS have significant changes to their intes-
Infections related to the central venous line have tinal microbiota (Piper 2018; Neelis et al. 2018).
two major pathways, an external and an internal Intestinal dysmotility with poor propagation of
route. Today, bacterial contamination of the infu- intraluminal chyme is the primary problem caus-
1026 M. E. Höllwarth

ing bacterial overgrowth. Overgrowing bacteria thy (Bianchetti et  al. 2018). In the past, cyclic
interfere with digestive and absorptive function, 5–7-day therapy with nonabsorbable antibiotics
leading to deconjugation of bile salts, nutrient (neomycin) and reduction of oral carbohydrate
malabsorption, and generation of toxic nutrition has been the standard therapy. Recently,
metabolites. fecal microbiota transplant has been successfully
Nonsurgical options in cases of bacterial over- used in severe and recurrent cases of D-lactic aci-
growth are cyclic antibiotic therapy; prokinetic dosis (Davidovics et  al. 2017; Bulik-Sullivan
medications, such as erythromycin; prebiotic et al. 2018).
fibers; and probiotics, such as lactobacilli and
bifidobacterial.
Surgical options are refashioning of signifi- 75.11.4 Intestinal Failure-­Associated
cantly dilated intestinal segments with to-and-fro Cholestasis and Liver Failure
peristalsis, either by simple tapering or by intesti- (IFCL)
nal lengthening, according to Bianchi or by a
STEP procedure (Hoellwarth 2014). In earlier years, parenteral nutrition was associ-
Bacterial translocation (BT) is the major ated with liver failure. Despite the fact that long-­
harmful consequence of bacterial overgrowth. term parenteral nutrition is associated with
Intestinal motility, gastric acidity, the intact recurrent catheter-related bacteremia and sepsis,
mucus layer, secretory IgA, and the ileocecal over the years it became more evident that recur-
valve are supposed to be factors preventing rent bacterial translocation from the intestinal
­bacterial overgrowth. While under normal cir- tract is associated with IFLD. BT is significantly
cumstances, BT to the mesenteric lymph nodes associated with poor intestinal motility and is
takes place at a rate of 5–8%, it increases up to seen as the main causative factor of IFCL, due to
tenfold, and systemic spread occurs if the muco- the constant influx of aggressive bacteria from
sal barrier function is impaired and bacterial the intestine to the portal vein and liver.
overgrowth prevails. BT occurs primary to the Additionally, in adult patients, cholestasis per se
portal and secondary to the systemic circulation is a well-known risk factor for significant BT and
and is the most critical complication of bacterialrecurrent peritonitis (Wiest et  al. 2014). Our
overgrowth. Extensive intestinal resection in results show that newborn babies with congenital
rodents have shown that bacterial overgrowth short bowel, either due to gastroschisis or multi-
causes significant BT to the liver and systemic ple intestinal atresia, have significantly more
circulation when compared with sham-operated severe and long-lasting intestinal motility prob-
animals (Schimpl et  al. 1999). BT is closely lems and IFCL, when compared to SBS infants
related to D-lactic acidosis, gut-derived sepsis, after postnatally acquired NEC or volvulus (Mayr
and finally, chronic liver failure in SBS patients.
et al. 1999). Significant progress has been made
by the reduction of soybean lipid emulsions and
intravenous supplementation using ω-3 fatty
75.11.3 D-Lactic Acidosis acids. In clinical studies, fish oil-based lipid
emulsions rich in ω-3 fatty acids have been seen
Oral feeding with carbohydrates together with to reverse liver cholestasis (Sigalet et al. 2013).
high gastric juice secretion and a low pH in the Some damage to the liver may be caused by
proximal bowel may lead to the overgrowth of certain amino acids or lipids, whether adminis-
lactobacilli in patients with SBS, which in turn tered parenterally or enterally (glycine, phytos-
produce D- and L-lactic acids (Bongaerts and terols). On the other hand, the deficiency of
Severijnen 2006). While L-lactic acid is rapidly specific amino acids, such as taurine, serine, or
metabolized to pyruvate in the liver, this process methionine, may result in toxic consequences on
for D-lactic acid is slow and can cause the so-­ the hepatocytes. Bacterial-driven bile salt decon-
called D-lactic acidosis, leading to encephalopa- jugation results in increased amounts of second-
75  Short Bowel Syndrome
ary bile salts such as deoxycholic or lithocholic
acid, which are absorbed in the colon and impair
bile flow. Ursodeoxycholic acid (UDCA) is
known to protect the liver from damage by block-
ing potentially toxic bile acid metabolites from
the colon after bacterial deconjugation of unab-
sorbed cholic acid.
75.12 Conclusion and Prognosis
The mainstay of the treatment of a newborn or
child with SBS consists of a sophisticated enteral
stimulation, with an individually balanced nutri-
tional equilibrium between carbohydrates, pro-
teins, and fatty acids. The enteral nutrition is the
best stimulus for adaptation. An interdisciplinary
pediatric intestinal rehabilitation group is needed
to achieve the best possible results in regard to
enteral autonomy. Citrulline has been shown as
an important marker for a number of functioning
enterocytes, predicting whether enteral autonomy
can be achieved.
Among 23 newborns with ultrashort small
bowel and midgut volvulus (8 after gastroschi-
sis), 18 remained transplant free, and 7 achieved
enteral autonomy (none of the gastroschisis
babies). Two patients died, one waiting for trans-
plantation and one 4  years after multivisceral
transplantation (Hong et  al. 2019). Patient age
and adjusted length of intestinal remnant, absence
of gastroschisis, intact colon, and ICV are con-
sidered to have the up-most impact on survival
and achieving enteral autonomy (Infantino et al.
2013; Diamanti and Basso 2012; Hoellwarth
2014; Capriati et al. 2018).
Data on short-term outcome are published in
many journals, but data on long-term outcome
are scarce. Two long-term studies on short bowel
patients showed that height was significantly
shorter and the bone mineral content was lower
than the reference values (Olieman et al. 2012a).
A second study by the same authors showed a
significantly lower health-related quality of life
of the grown-up children and parents (Olieman
et al. 2012b). A newborn with 10 cm of jejunum
und 1 cm of ileum and preserved ICV received a
3  cm reversed jejunal segment in the authors’
1027
clinic in 1975. That patient is 45 years old today,
on full enteral nutrition under regular supplemen-
tation of fat-soluble vitamins and vitamin B12,
and is working in a nursery home for elderly
citizens.
References
Beyer-Berjot L et  al (2012) Segmental reversal of the
small bowel can end permanent parenteral nutrition
dependency: an experience in 38 adults with short
bowel syndrome. Ann Surg 256:739–744
Bianchetti DGA et  al (2018) D-lactic acidosis in
humans: systemic literature review. Pediatr Nephrol
33:673–681
Bianchi A (1980) Intestinal loop lengthening  – a tech-
nique for increasing small intestinal length. J Pediatr
Surg 15:145–151
Bianchi A (1999) Experience with longitudinal intes-
tinal lengthening and tailoring. Eur J Pediatr Surg
9:256–259
Bloom SR, Polak JM (1982) The hormonal pattern of
intestinal adaptation. A major role for enteroglucagon.
Scand J Gastroenterol 74:170–173
Bongaerts G, Severijnen R (2006) Arguments for lower
carbohydrate-high fat diet in patients with a short
small bowel. Medical Hypothesis 67:280–282
Brink MA et  al (1996) Bilirubin cycles enterohepati-
cally after ileal resection in the rat. Gastroenterology
110:1945–1957
Brubaker PL (2018) Glucagon-like peptide-2 and the
regulation of intestinal growth and function. Compr
Physiol 8:1185–1210
Bulik-Sullivan E et  al (2018) Intestinal microbial and
metabolic alterations following successful fecal
microbiota transplant for D-lactic acidosis. JPGN
67:483–487
Capriati T et  al (2018) Pediatric short bowel syndrome:
predicting four-year outcome after massive neonatal
resection. Eur J Pediatr Surg 28:455–463
Carter BA et al (2017) Outcomes from a 12-week, open
label, multicenter clinical trial of teduglutide in pediat-
ric short bowel syndrome. J Pediatr 181:102–111
Chiba M et  al (2017) Markers on intestinal adapta-
tion in pediatric short bowel syndrome. Pediatr Int
59:911–916
Crenn P et al (2008) Citrulline as a biomarker of intesti-
nal failure due to enterocyte mass reduction. Clin Nutr
27:328–339
Cole CR et  al (2008) Very low birth weight preterm
infants with surgical short bowel syndrome: incidence,
morbidity and mortality, and growth outcome at 18 to
22 month. Pediatrics 122:e573–e582
Colomb V et  al (2007) Long-term outcome of children
receiving home parenteral nutrition: A20-year single-­
center experience in 320 patients. JPGN 44:347–353
1028
Dai Y-J et  al (2016) Comparison of formulas based on
lipid emulsions of olive oil, soyabean oil, or several
oils for parenteral nutrition: a systematic review and
meta-analysis. Adv Nutr 7:279–286
D’Antiga L, Goulet O (2013) Intestinal failure in chil-
dren: the European view. J Pediatr Gastroenterol Nutr
56:118–126
Davidovics ZH et al (2017) Fecal transplantation success-
fully treats recurrent D-lactic acidosis in a child with
short bowel syndrome. JPEN 41:896–897
Debongnie J, Phillips A (1978) Capacity of the human
colon to absorb fluid. Gastroenterology 74:698–703
Demehri FR et  al (2015) Enteral autonomy in pediatric
short bowel syndrome: predictive factors one year
after diagnosis. J Pediatr Surg 50:131–135
Diamanti A, Basso MS (2012) Colon and intestinal adap-
tation in children with short bowel syndrome. Letter to
the editor. JPEN 36:501
Duggan CHP, Jaksic T (2017) Pediatric intestinal failure.
N Engl J Med 377:666–675
Goulet O et  al (2005) Outcome and long-term growth
after extensive small bowel resection in the neonatal
period. Eur J Pediatr Surg 15:95–101
Goulet O et al (2009) Role of the colon in short bowel syn-
drome and intestinal transplantation. JPGN 48(Suppl
2):S66–S71
Goulet O, Lambe C (2017) Intestinal lipid emulsions in
pediatric patients with intestinal failure. Curr Opin
Organ Transplant 22:142–148
Goulet O et  al (2019) Short bowel syndrome as the
leading cause of intestinal failure in early life: some
insights into the management. Pediatr Gastroenterol
Hepatol Nutr 22:303–329
Granger DN, Morris J, Kvietys PR (2018) Physiology
and pathophysiology of digestion. In: Granger DN,
Granger J (eds) Colloquium series on integrated physi-
ology. Morgan & Claypool Life Sciences, San Rafael,
CA
Grant C et al (2015) Human and mouse tissue-engineered
small intestine both demonstrate digestive and absorp-
tive function. Am J Physiol Gastrointest Liver Physiol
308(8):G664–G677
Greig CJ et al (2019) Retracing our STEPs: four decades
of progress in intestinal lengthening procedures for
short bowel. Am J Surg 217:772–782
Haxhija EQ et al (2008) Modulation of mouse intestinal
cell turnover in the absence of angiotensin convert-
ing enzyme. Am J Physiol Gastrointest Liver Physiol
295:G88–G98
Hoellwarth ME et al (1988a) Pharmacologic enhancement
of adaptive growth after extensive small bowel resec-
tion. Pediatr Surg Int 3:55–61
Hoellwarth ME et al (1988b) Blood flow in experimental
short bowel syndrome. Pediatr Surg Int 4:242–246
Hoellwarth ME (2014) Short bowel syndrome in child-
hood. In: Granger DN, Granger J (eds) Colloquium
series on integrated systems physiology. Morgan &
Claypool Life Sciences
Hong CR et al (2019) Long-term outcomes of ultrashort
bowel syndrome due to malrotation with midgut vol-
M. E. Höllwarth
vulus managed at an interdisciplinary pediatric intes-
tinal rehabilitation center. J Pediatr Surg 54:964–967
Infantino BJ et  al (2013) Successful rehabilitation in
pediatric ultrashort small bowel syndrome. J Pediatr
163:1361–1366
Javid PJ et al (2018) Organization and outcomes of mul-
tidisciplinary intestinal failure teams. Semin Pediatri
Surg 27:218–222
Jeejeebhoy NK (2008) Treatment of intestinal failure:
transplantation or home parenteral nutrition. Editorial
Gastroenterol 135:303–305
Jeppesen PB, Mortensen PB (1999) Colonic diges-
tion and absorption of energy from carbohydrates
and medium chain fat in small bowel failure. JPEN
23:S101–S105
Jeppesen PB et  al (2018) Factors associated with
response to teduglutide in patients with short-bowel
syndrome and intestinal failure. Gastroenterology
154:874–885
Johnson LR (1976) The trophic action of gastrointestinal
hormones. Gastroenterology 70:278–288
Joly F et  al (2009) Morphological adaptation with pre-
served proliferation/transporter content in the colon
of patients with short bowel syndrome. Am J Phys
297:G116–G123
Kim M-H, Kim H (2017) The role of glutamine in the
intestine and its implication in intestinal diseases.
Int J Mol Sci 18:1051. https://doi.org/10.3390/
ijms18051051
Kono K et al (2001) Interposed colon between remnants
of the small intestine exhibits small bowel features
in patients with short bowel syndrome. Dig Surg
18:237–241
Lacaille F et  al (2017) Twenty-eight years of intesti-
nal transplantation in Paris: experience of the oldest
European center. Transp Int 30:178–186
Lai SW et  al (2017) Effects of exogeneous glucagon-­
like peptide-2 and distal bowel resection on intestinal
and systemic adaptive response in rats. PLoS One
12(7):e0181453
Lloyd DA (1978) Colon interposition between jejunum
and ileum after massive small bowel resection in rats.
Prog Pediatr Surg 12:51–106
Mayr JM et  al (1999) Morbidity and mortality of short
bowel syndrome. Euro J Pediatr Surg 9:231–235
McMellen ME et al (2010) Growth factors: possible role
for clinical management of the short bowel syndrome.
Semin Pediatr Surg 19:35–43
Merras-Salmio L et al (2018) Pediatric intestinal failure:
The key outcomes for the first 100 patients treated in a
National Tertiary Referral Center during 1984–2017.
JPEN 00:1–10
Merritt RJ et al (2017) Intestinal rehabilitation programs
in the management of pediatric intestinal failure and
short bowel. JPGN 65:588–596
Mughal M, Irving M (1986) Home parenteral nutrition in
the united Kingdome and Ireland. Lancet 2:383–387
Naimi RM et  al (2019) Effects of glepaglutide, a novel
long-acting glucagon-like peptide 2 analogue, on
markers of liver status in patients with short bowel
75  Short Bowel Syndrome
syndrome: findings from a randomized phase 2 trial.
EBioMedicine 46(444):451
Neelis E et al (2018) The gut microbiome in patients with
intestinal failure: current evidence and implications
for clinical practice. JPEN 00:1–12
Nordgaard I et  al (1994) Colon as digestive organ in
patients with short bowel. Lancet 343:373–376
Norgaard I et al (1996) Importance of colonic support for
energy absorption as small-bowel failure proceeds.
Am J Nutr 64:222–231
Olieman JF et  al (2012a) Long-term impact of infan-
tile short bowel syndrome on nutritional status and
growth. Brit J Nutr 107:1479–1489
Olieman JF et  al (2012b) Impact of infantile short
bowel syndrome on long-term health-related qual-
ity of life: a cross-sectional study. J Pediatr Surg
47(1309):1316
O’Keefe SJD et  al (2013) Safety and efficacy of tedu-
glutide after 52 weeks of treatment in patients with
short bowel intestinal failure. Clin Gastroenterol and
Hepatol 11:815–823
Panis Y et al (1997) Segmental reversal of small bowel as
an alternative to intestinal transplantation in patients
with short bowel syndrome. Ann Surg 225:401–407
Peterson J, Kerner JA (2012) New advances in the man-
agement of children with intestinal failure. JPEN
36:S36–S42
Piper HG (2018) Intestinal microbiota in short bowel syn-
drome. Semin Pediatr Surg 27:223–228
Pironi L et al (2012) Outcome on home parenteral nutri-
tion for benign intestinal failure: a review of the litera-
ture and benchmarking with the European prospective
survey of ESPEN. Clin Nutrition 31:831–845
Potts WJ (1955) Pediatric surgery. JAMA 157:627–630
1029
Pul M et al (1991) Enhancement by testosterone of adap-
tive growth after small bowel resection. Isr J Med Sci
27:339–342
Rickham PP (1967) Massive intestinal resection in new-
born infants. Ann R Coll Surg 41:480–485
Salvia G et al (2008) Neonatal onset intestinal failure: an
Italian multicenter study. J Pediatr 153:674–676
Schimpl G et  al (1999) Bacterial translocation in
short bowel syndrome in rats. Eur J Paediatr Surg
9:224–227
Shirafkan A et  al (2016) New approaches to increase
intestinal le ngth: method used for intestinal regenera-
tion and bioengineering. World J Transplant 24:1–9
Sigalet D et al (2013) Nutritional support of infants with
intestinal failure: something more than fishy is going
on here. Pediatr Surg Int 29:975–981
Sigalet DL et  al (2017) Safety and dosing study on
glucagon-­like peptide 2 in children with intestinal fail-
ure. JPEN 41:844–852
Smith A et al (2018) Laboratory monitoring of children on
home parenteral nutrition: a prospective study. JPEN
42:148–155
Struijs M-C et al (2009) Establishing norms for intestinal
length in children. J Pediatr Surg 44:933–938
Sukhotnik I et  al (2002) Sandostatin impairs intestinal
adaptation in a rat model of short bowel syndrome.
Dig Dis Sci 47:2095–2102
Touloukian RJ, Walker-Smith GJ (1983) Normal intesti-
nal length in preterm infants. J Pediatr Surg 6:720–723
Wales PW et al (2004) Neonatal short bowel syndrome:
population-based estimates of incidence and mortality
rates. J Pediatr Surg 39:690–695
Wiest R et al (2014) Pathological bacterial translocation
in liver cirrhosis. J Hepatol 60:197–209
 Inflammatory Bowel Disease
Risto J. Rintala and Mikko P. Pakarinen
76.1 Ulcerative Colitis
76.1.1 Introduction
Ulcerative colitis (UC) is an inflammatory bowel
disease (IBD) of unknown aetiology confined to
rectum and colon. Males and females are equally
affected. UC is suspected in patients having diar-
rhoea that is often bloody, abdominal pain and
rectal tenesmus. In severe cases, the disease is
associated with weight loss, fatigue and general
feeling of illness. The incidence of UC ranges
between 1 and 4/100000 per year in Northern
America and Europe (Benchimol et al. 2011). The
incidence is highest in the Nordic countries and in
the British Isles. About 5% of all cases of UC
have their onset before the age of 10  years and
about 20% before the age of 20 years. The inci-
dence of paediatric cases has remained unchanged
since the 1980s in some parts of the Western
World but is still increasing in Northern countries,
especially in Finland (Lehtinen et al. 2011).
At diagnosis, paediatric UC is extensive in
60–80% of case, which is much more than in
R. J. Rintala (*)
Department of Pediatric Surgery, Children’s Hospital,
Helsinki University Central Hospital, Helsinki, Finland
e-mail:risto.rintala@saunalahti.fi
M. P. Pakarinen
Department of Pediatric Surgery, Hospital for
Children and Adolescents, University of Helsinki,
Helsinki, Finland
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_76
76
adults (Van Limbergen et al. 2008). Widespread
disease is associated with more severe disease
course. Therefore, the risk of colectomy in paedi-
atric onset UC within 10 years is reported to be
higher than in adults, 30–40% of the patients
require colectomy compared to about a 20% col-
ectomy rate in adults (Van Limbergen et al. 2008;
Gower-Rousseau et al. 2009).
Paediatric UC has unique features when com-
pared to adult-onset disease. It may affect the
development of growth and puberty in preadoles-
cent patients. Moreover, the disease onset is com-
monly at a psychosocially vulnerable age,
namely, prepuberty or puberty.
76.1.2 Aetiology
The aetiology of UC is unknown and most likely
multifactorial. Various potential contributing fac-
tors have been studied including environmental
factors, infections, psychosocial factors, immu-
nological factors and genetic factors. There is a
clear inherited predilection for UC with up to a
20% chance of UC in identical twins (Orholm
et al. 2000). Also, the genetic basis of the disease
has long been implied by the high prevalence of
disease within families. No specific gene locus
has been identified.
Most believe intestinal bacterial flora is vital
for persistence of the inflammatory process
1031
1032
(Sartor 2005). Numerous environmental factors
have been implicated; these include cigarette
smoke (protective), appendectomy (protective),
breastfeeding (protective), diet (high fat/sugar),
perinatal or early childhood infections, wild-
type measles infections, Mycobacterium paratu-
berculosis, and oral contraceptives (Loftus
2004).
76.1.3 Pathology
UC is a chronic inflammatory disease of the rec-
tum and colon affecting mucosa and submucosa
of the bowel wall. The rectum is affected in more
than 95% of the cases, and the inflammation
extends contiguously to the more proximal large
bowel. Ileal involvement strongly suggests
Crohn’s disease (CD), although low-grade unspe-
cific inflammation of the ileal mucosa, backwash
ileitis, is often seen in patients with UC.
The characteristic histological appearance of
UC consists of diffuse superficial unspecific
inflammation, neutrophilic epithelial invasion,
crypt abscesses and crypt deformity. Progression
of inflammation leads to mucosal ulceration and
epithelial regeneration with pseudopolyp forma-
tion. Long-standing disease is associated with
atrophic and dysplastic mucosa. The risk of colon
cancer associated with UC is of special concern
in paediatric patients. The risk of colorectal can-
cer in early onset UC (< 20 years of age) is clearly
increased when compared with adult-onset dis-
ease. However, the absolute risk is still limited
and is 1.6% after 25 years of disease. In matched
controls, the risk is 0.08%. Patients with associ-
ated primary sclerosing cholangitis and those
with long-standing severe active disease who
have not had a colectomy carry the highest risk of
colorectal cancer.
76.1.4 Extraintestinal Manifestations
Extraintestinal manifestations are also relatively
common in UC in the paediatric age group. The
extraintestinal manifestations may affect joints,
skin, eyes, bile ducts, and various other organs.
The most frequent joint manifestations are
R. J. Rintala and M. P. Pakarinen
peripheral arthritis and axial arthropathies. UC
patients, as well as all IBD patients, have an
increased risk of osteoporosis and osteopenia.
Prevalence rates for osteoporosis have been
reported to range between 2% and 20% and
between 40% and 50% for osteopenia (Larsen
et al. 2010).
Erythema nodosum and pyoderma gangreno-
sum are common dermatologic manifestations.
Ophthalmologic complications include episcleri-
tis, iridocyclitis, and uveitis.
Anaemia is also frequently seen. The underly-
ing mechanisms behind anaemia are anaemia due
to chronic inflammation and iron deficiency due
to chronic intestinal bleeding. UC patients are
also susceptible to thromboembolic complica-
tions such as venous thromboembolisms
(Bernstein et al. 2001).
There is a strong association between UC
and primary sclerosing cholangitis. In
Scandinavian countries, most patients with pri-
mary sclerosing cholangitis have IBD (Bergquist
et al. 2007). The association is less pronounced
in other parts of the world. On the other hand,
3% to 7% of patients with UC have sclerosing
cholangitis.
76.1.5 Diagnosis
76.1.5.1 TÄSTÄ
Diagnosis of paediatric UC should be based on a
combination of history of intermittent or con-
stant bloody diarrhoea and abdominal tenesmus,
physical examination, laboratory tests and ileo-
colonic and upper gastrointestinal tract endos-
copy with multiple biopsies (Turner et al. 2018).
In order to detect small bowel involvement to
rule out CD, MRI enterography (MRE), wireless
capsule endoscopy (WCE) (Nuutinen et  al.
2011) and double-balloon enteroscopy are rec-
ommended. Physical examination should include
the pattern of growth, body mass index, abdomi-
nal distention and tenderness, anorectal exami-
nation to confirm the absence of fistulas, fissures
and skin tags and search for extracolonic mani-
festations. Significant growth delay (height
velocity < 2SD) is uncharacteristic of UC (Turner
et al. 2012).
76  Inflammatory Bowel Disease
Laboratory Tests
Initial laboratory investigations should include
a full blood count, liver enzymes, albumin,
erythrocyte sedimentation rate, iron status and
C-reactive protein. Serological markers perinu-
clear anti-neutrophil cytoplasmic antibodies
(p -ANCA), found in 60–80% of UC patients vs
<30% in CD patients, and anti-saccharomyces
cerviciae antibodies (ASCA), found in 10–15%
of UC patients and in 50–70% of CD patients,
may be useful in reducing the likelihood of UC,
but even ‘positive’ CD serology (ASCA+/
pANCA-) does not preclude the diagnosis of
UC. Immunological testing for atypical presen-
tation and for children younger than 2  years
should include lymphocyte phenotyping (T, B,
NK, NK-T cells), immunoglobulin levels and
assessment of lymphocytic responses to anti-
gens and mitogens including chronic granulo-
matous disease and immunodeficiencies. Faecal
inflammatory parameters, calprotectin and lac-
toferrin may differentiate colitis from nonifla-
mmatory diarrhoea, and calprotectin may be
useful in assessing the outcome of medical
therapy. Stool culture to exclude infectious ori-
gin of diarrhoea with testing for Salmonella,
Shigella, Yersinia, Campylobacter and
Clostridium difficile and for pathogens likely
acquired from recent travel are mandatory.
Endoscopy
Endoscopic evaluation by ileocolonoscopy is
recommended at diagnosis, before major treat-
ment changes and when the clinical assessment
is in question. The endoscopist should be famil-
iar with the typical and atypical features of
paediatric-­onset UC and CD. In active UC, the
findings are typically diffuse and continuous,
the mucosa is oedematous, erythematous and
friable with loss of vascular pattern and surface
erosions and ulcerations, and with long-stand-
ing UC, pseudopolyps may be seen. The inflam-
mation may either end at a transition zone
anywhere in the colon or involve the whole
colon (pancolitis). It is imperative that the
ileum is reached, and at least two biopsies from
a minimum of six areas of the inspected colon
1033
and ileum should be taken. With pancolitis, the
terminal ileum and the ileocecal valve may
show non-erosive erythema and edema (back-
wash ileitis). In addition, the anus should be
inspected to exclude fistulizing CD. The biop-
sies taken at first presentation are the most
informative because post-treatment biopsies are
modified by treatment and chronicity. However,
some typical histological features only develop
later on the course of the disease. In 5–30% of
children with UC, the rectum may appear to
have no or only patchy disease, but combined
macroscopical and microscopical rectal sparing
is rare. In children aged less than ten years,
biopsies taken shortly after the onset of symp-
toms may disclose patchy disease and a lack of
typical architectural distortion. In 2% of paedi-
atric UC patients, left-sided colitis with a peri-
appendicular area of caecal inflammation
(caecal patch) may be found at diagnosis
(Feakins 2013; Turner et al. 2012).
Endoscopy of the upper GI tract should be
performed with multiple biopsies from both
affected and macroscopically unaffected loca-
tions from the duodenum, stomach and esopha-
gus. In UC, gastric erosions and non-serpiginous
small ulcers may be seen in 4–8% of patients.
Focal gastritis and duodenitis may occur in
both UC and CD, whereas the presence of deep
ulcers, erosions and aphtous lesions with his-
tology of non-cryptolytic granulomas are typi-
cal of CD (Turner et  al. 2012; Kovacs et  al.
2012).
 aediatric Ulcerative Colitis Activity Index
P
The activity of UC can be assessed by a validated
standardized Paediatric Ulcerative Colitis
Activity Index (PUCAI) (Table  76.1). PUCAI
score points of ≥65 indicate severe, 35–64 mod-
erate and 10–34 mild activity. Change of disease
activity by reduction of PUCAI score points can
be expressed as significant (≥35 points), moder-
ate (≥20) small (≥10) or none (<10). (Turner
et al. 2007) In medically treated UC, PUCAI has
been proven to be a valid non-invasive primary
outcome measure and correlates highly with
colonoscopy.
1034 R. J. Rintala and M. P. Pakarinen

Table 76.1  Paediatric Ulcerative Colitis Activity Index with the presence of normal cecum, fistulizing
(PUCAI) (Turner et al. 2007) disease, any signs of significant small bowel affi-
Item Points sion, skip lesions outside cecum and rectum in
Abdominal pain 0 untreated IBD or large inflamed perianal tags,
 No pain 5 should be diagnosed as CD.
 Pain can be ignored 10
Inflammatory disease otherwise consistent
 Pain cannot be ignored
Rectal bleeding with UC with any of the following features: com-
 None 0 bined macroscopic and microscopic rectal spar-
 Small amount only in less than 50% of stools 10 ing, significant growth delay, transmural
 Small amount only in more than 50% of 20 inflammation in the absence of severe colitis,
stools duodenal and esophageal ulcers, aphthous ulcers
 Large amount (>50% of stool content) 30
in stomach, ASCA+/ANCA- serology, reverse
Stool consistency of most stools
 Formed 0
(proximal >distal) gradient of inflammation) or
 Partially formed 5 with two to three of the following features: severe
 Completely unformed 10 scalloping of stomach and duodenum or duodeni-
Number of stools/24 h tis not explained by celiac disease or H. pylori,
 0–2 0 focal active colitis on histology in more than one
 3–5 5 biopsy of macroscopically inflamed site, non-­
 6–8 10
bloody diarrhoea and aphthous ulcerations of
 >8 15
Nocturnal stools (any episode causing
colon or upper gastrointestinal tract should be
wakening) diagnosed as IBD-U (Levine et al. 2014; Turner
 No 0 et al. 2012).
 Yes 10 Involvement of the small bowel in the pres-
Activity level ence of normal cecum is not typical for UC, and
 No limitation of activity 0 any positive finding in endoscopy or radiologic
 Occasional limitation of activity 5
imaging should raise suspicions of CD.  Small
 Severe restricted activity 10
Sum of PUCAI = 0–85
bowel wall thickening in MRE is sensitive but
not pathognomonic for CD.  In WCE, the
absence of small bowel lesions is supportive of
Differential Diagnostics UC, but frequently, false-positive lesions com-
In clinical practice, the differential diagnosis plicate the interpretation of the study, and a few
between UC and CD based on endoscopic biop- small erosions in WCE do not preclude the
sies, or even on examination of full-thickness diagnosis of UC (Levine et  al. 2014; Turner
biopsies from a resected specimen, is often prob- et  al. 2012). Despite careful assessment, how-
lematic. For example, children presenting with ever, as much as 13% of children who undergo
acute severe UC may have transmural inflamma- colectomy and ileoanal J-pouch for UC end up
tion with deep ulcers characteristic of CD, but the with the ultimate diagnosis of CD (Mortellaro
absence of lymphoid aggregates and presence of et al. 2011).
fissuring V-shaped ulcers are supportive of
UC.  The inflammatory bowel disease may pos-
sess features of both UC and CD; these, espe- 76.1.6 Medical Management
cially in young children, are often classified as
IBD- unclassified (IBD-U). The main goals of the treatment are to alleviate
In ESPGHAN criteria, an IBD endoscopically symptoms, to provide normal growth and devel-
limited to the colon with the following findings: opment and to avoid disease-related long-term
well-formed granulomas anywhere in the gastro- complications. Children often present with
intestinal tract, deep serpentine ulcerations, cob- widespread disease and pancolitis, necessitat-
blestoning or stenosis, any ileal inflammation ing aggressive medical treatment. If patients
76  Inflammatory Bowel Disease
are significantly malnourished or when there is
growth retardation, intensive nutritional sup-
port is provided.
Oral treatment with 5-ASA preparations is the
first-line medication to induction therapy for
mild to moderate active UC in children (Ford
et al. 2011; Turner et al. 2018). These drugs are
useful for maintenance therapy of remission,
regardless of initial induction treatment. Oral
5-ASA therapy may be combined with rectal
5-ASA enemas.
Oral steroids are commonly needed to control
moderate or severe disease. In addition, cases fail-
ing to achieve remission with 5-ASA therapy
should be treated with oral steroids. The steroids
are effective for inducing remission but not for
maintaining it. In patients with severe disease, the
induction therapy with steroids should be started
with intravenous infusions (Turner et al. 2018).
Up to 80% of children with UC receive ste-
roids to achieve remission within 3 months after
the diagnosis (Hyams et al. 2006). This is associ-
ated with clinical response in the majority.
However, steroid dependency, meaning inability
to achieve and maintain steroid-free clinical
remission, has been reported in up to 45% of chil-
dren with UC (Hyams et al. 2006). Steroid depen-
dency should not be tolerated as this is associated
with significant complications such as osteoporo-
sis, changes in body image, psychological altera-
tions and growth disturbances. Steroid
dependency may be avoided by optimization of
5-ASA therapy and considering immunomodu-
lating therapy with thiopurines or biological
medications such as infliximab.
Thiopurines, such as azathioprine and mercap-
topurine, can be used to maintain remission in chil-
dren who have an intolerance to 5-ASA or steroid
dependence. The role of other immunosuppressive
medications, such as cyclosporine and tacrolimus,
in severe childhood UC is unsettled but they may
be useful in steroid dependency (Ziring et al. 2007).
Biological drugs, such as infliximab, should be
considered for persistently active or steroid-depen-
dent UC that cannot be controlled by 5-ASA and
thiopurines (Turner et al. 2018).
1035
76.1.7 Surgical Management
76.1.7.1 Principles and Indications
UC can be cured by surgical removal of the dis-
eased colon and rectum in patients, who fail to
reach or maintain remission with medical ther-
apy. Children with UC are more likely to have
an extensive colonic involvement and a pro-
gressive clinical course than their adult coun-
terparts. Approximately 25% to 40% of children
will require surgery by disease duration of
10 years, and a quarter of them are operated on
for fulminant colitis (Pakarinen et  al. 2009).
The operation rate has remained stable over the
last decades. Operative treatment offers high-
quality functional outcomes in the long term,
while abolishing the severe side effects of
immunosuppressive medication. Therefore sur-
gical therapy should be considered for any
patient prior to development of complications
secondary to the disease itself or its medical
treatment. Optimally, operative treatment
should be performed before completion of
growth to enable catch up growth and to avoid
developmental delays.
Surgery is performed electively in most cases.
Elective proctocolectomy may be indicated in
children with active or steroid-dependent UC
despite maximal treatment with 5-ASA, thiopu-
rines, calcineurin inhibitors and biological ther-
apy (Turner et  al. 2018). Confirmed finding of
colonic dysplasia, although rare during child-
hood, should prompt operative treatment without
delay. Emergency colectomy is required in about
25% of cases due to fulminant disease refractory
to medical therapy, extensive rectal bleeding or
toxic megacolon. If symptoms and clinical signs
of toxic megacolon including fever, tachycardia,
dehydration, electrolyte disturbances, hypoten-
sion and altered level of consciousness worsen or
do not resolve within 24–72  hours of medical
treatment, immediate colectomy should be per-
formed (Turner et  al. 2018). Emergency colec-
tomy for UC entails removal of the colon, closure
of the remaining rectum and construction of end
ileostomy.
1036
76.1.7.2 Preoperative Assessment
In children, CD commonly presents as colitis
without small intestinal involvement, rendering
differential diagnosis between CD and UC chal-
lenging. Because CD strongly associates with
postoperative complications, ileoanal
­pull-­through failure and pouch removal, patients
should undergo full diagnostic work-up includ-
ing upper and lower gastrointestinal endoscopy,
determination of serologic markers, capsule
endoscopy and magnetic resonance imaging
enterography before proceeding to proctocolec-
tomy (Nuutinen et al. 2011; Piekkala et al. 2012).
It is very important that all aspects and options
of surgical therapy are discussed, preferentially
well in advance, with the patients and parents.
Reliable and honest information on possible
postoperative complications, pouchitis and conti-
nence expectations including stooling frequency
and nighttime bowel actions should be provided.
Although corticosteroids and other medical treat-
ment for UC can be stopped after operation, most
patients continue medical management to
improve stool consistency and frequency or for
pouchitis. In female patients, potential conse-
quences on sexual function and fertility should
also be addressed. Preoperative discussion with a
stoma nurse will accommodate the family with
postoperative enterostomy care, and evaluation
by a nutritionist helps to optimize diet and nutri-
tion before and after the operation. Malnutrition,
anaemia and hypoalbunemia should be corrected
before operation. For optimal placement, the site
of protecting loop ileostomy in the right lower
abdominal quadrant should be marked preopera-
tively while the patient is sitting. Prophylactic
antimicrobial therapy against gram-negative and
anaerobic bacteria is started an hour before sur-
gery and continued 24 hours postoperatively.
76.1.8 Operative Approach
76.1.8.1 Selection of Surgical
Approach
An ileoanal pull-through procedure is invariably
preferred instead of proctocolectomy with per-
manent end ileostomy by the patients. At present,
R. J. Rintala and M. P. Pakarinen
the latter procedure is mainly performed in a case
of pull-through failure. Colectomy with ileorec-
tal anastomosis is not recommended for operative
treatment of UC due to an exceedingly high fail-
ure rate, requiring removal of the remaining rec-
tum because of refractory proctitis, dyplasia or
cancer. Ileorectal anastomosis may be considered
in selected girls who are primarily concerned
about the reduced fecundity associated with
restorative proctocolectomy.
The ileoanal pull-through may be performed
with or without an ileal pouch in one to three
stages as an open or laparoscopic-assisted proce-
dure. Emergency colectomy should be always
performed in three stages. The three-stage sur-
gery may be also considered for malnourished
patients, those treated with high-dose corticoste-
roids or immunosuppressive medication and if
the possibility of a diagnosis of CD is appreciable
especially among the youngest patients (Turner
et al. 2018). In these patients, colectomy allows
discontinuation of systemic steroids, improve-
ment of nutritional and general status as well as
detailed histological evaluation of the resected
specimen before proceeding to ileoanal pull-­
through. Following colectomy and end ileostomy
in the first stage, removal of the remaining rec-
tum (proctectomy) and ileoanal pull-through
with a protective loop ileostomy is performed in
the second stage. The loop ileostomy is taken
down in the third stage. The two-stage operation
is the most common elective surgical approach
for UC in children and safer than the one-stage
operation, which is associated with an increased
rate of anastomotic leakage and pelvic infections
(Turner et al. 2012). However, one-stage ileoanal
pull-through without a protective ileostomy may
be safely performed in selected children without
apparent risk factors, such as high-dose cortico-
steroid or immunosuppressive treatment, malnu-
trition or high degree of disease activity, and
smoothly performed operation without a signifi-
cant anastomotic tension. In general, the J-pouch
ileoanal anastomosis is preferred over straight
pull-through, because it is associated with lower
stooling frequency and better faecal continence
(Tilney et al. 2006), although it carries a greater
risk of pouchitis (Seetharamaiah et  al. 2009).
76  Inflammatory Bowel Disease
Straight pull-through remains a viable option in
those rare cases when limited length of the ileal
mesentery prevents ileoanal pouch reconstruc-
tion. The laparoscopic-assisted approaches may
be also used with comparable complication rate
and better cosmetic results.
76.1.8.2 Restorative Proctocolectomy
The main goals of restorative proctocolectomy
are to cure the disease by surgical removal of the
entire colon and rectum and to retain as good
anorectal function as possible. These goals are
best achieved by removing of the distal rectal
mucosa by mucosectomy. Mucosectomy may be
performed either transabdominally or transa-
nally. The mucosectomy should leave about one
centimeter above the dentate line intact, thus pre-
serving transitional epithelium of the anal canal.
Preservation of the transitional epithelium is cru-
cial for undisturbed faecal continence, because it
mediates rectal sensation and the ability to dis-
criminate between gaseous, liquid and solid
bowel contents.
The purpose of an ileoanal pouch is to replace
the reservoir function of the removed rectum.
This artificial reservoir allows prolonged reten-
tion of bowel contents before defecation and
thereby reduced stooling frequency and extended
urgency period. The J-pouch is simplest to con-
struct and has the lowest long-term complication
rate, so it is the most commonly used pouch con-
figuration today. The ileal pouch-anal anastomo-
sis may be performed either as hand-sewn or with
a circular stapler (Geiger et al. 2003).
76.1.9 Postoperative Management
Prophylaxis against deep venous thromboem-
bolic complications includes pneumatic com-
pression stockings in electively operated older
children combined with low-molecular-weight
heparin in those undergoing an emergency colec-
tomy. Steroids are gradually tapered and discon-
tinued within a few weeks. The patient is
readmitted approximately 3  months after the
operation for ileostomy closure after anastomotic
1037
leakage or fistula formation has been ruled out by
a water-soluble contrast enema. The small
amount of rectal mucosa remaining proximal to
the dentate line at the ileoanal anastomosis neces-
sitates long-term endoscopic surveillance for
development of dysplastic or malignant changes.
Frequent small meals of low-residue and
lactose-­free diet and dietary salt supplementation
are advised to decrease intestinal excretion and
along with bulking fiber preparations to increase
stool consistency after ileostomy closure. For
optimization of enteric fluid absorption, oral
sodium chloride capsules (500 mg four to eight
times a day) are given to maintain urinary sodium
concentration above 20 mmol/l. Readily absorb-
able dietary energy supplements may be benefi-
cial to promote weight gain during early
postoperative period. Loperamide (2  mg two to
six times a day) is prescribed to decrease fre-
quency of bowel actions. Nutritional status and
bowel function require frequent monitoring and
adjustment during the first six months after ileos-
tomy closure.
76.1.10 Surgical Complications
Operative mortality is exceedingly rare. Overall,
surgical complications following proctocolec-
tomy are common, while over half of patients
encounter at least one surgical complication
(Nyholm et al. 2019). The most common compli-
cation is adhesive bowel obstruction requiring
surgery in 30% of the patients (Nyholm et  al.
2019). Decreased small bowel obstruction rate
has been reported following laparoscopically
assisted procedures by some (Linden et al. 2013),
but not all authors (Diamond et  al. 2010). In a
recent series of children with open proctocolec-
tomy and J-pouch ileoanal anastomosis, the inci-
dence of wound infection was 5%, pelvic sepsis
2–14%, anastomotic stenosis 10–19%, anasto-
motic leakage 5–10% and fistula formation 5–9%
(Tilney et  al. 2006; Nyholm et  al. 2019;
Seetharamaiah et al. 2009). Preoperative steroid
therapy and malnutrition are associated with
increased rates of postoperative surgical compli-
1038
cations (Turner et  al. 2012). Most anastomotic
stenosis are readily dilatable and rarely require
surgical intervention.
Around 10% of patients undergo repeat ile-
ostomy or abdominal salvage procedure mainly
for management of fistulas, anastomotic dehis-
cence or stricture. Other potential reasons for
repeat ileostomy include intractable high stool-
ing frequency or severe pouchitis. Overall inci-
dence of pouch failure is approximately 10%.
Approximately one-third of pouch removals
are attributed to ultimate diagnosis of CD,
which may manifest as recalcitrant inflamma-
tion of the rectal cuff or pouchitis, perianal
abscess, fistula or stricture. Of all children
undergoing proctocolectomy and ileoanal
anastomosis, 10–15% will be reclassified as
having CD (Alexander et  al. 2003; Nyholm
et al. 2019).
76.1.11 Outcomes of Ileoanal
Pull-Through
76.1.11.1 Stooling Frequency
and Faecal Continence
Following J-pouch ileoanal anastomosis, loose
and frequent bowel motions ranging from 5 to 15
times per 24 hours, with notable individual varia-
tion, characterize early the postoperative period
after ileostomy closure. Within the following
3–6  months, stooling frequency typically
decreases to 3–8 per 24 hours. Nighttime bowel
actions show a comparable temporal decrease
from 2–3 to one (Seetharamaiah et al. 2009). In
the long term, an average daytime stooling fre-
quency is 5–6, while most patients experience
one bowel motion during nights at least occasion-
ally. Stooling frequency remains stable after a
mean follow-up of 10 years (Nyholm et al. 2019).
Although faecal accidents occur only in few per-
cent of patients, nighttime seepage or soiling is
relatively common especially during the first few
months (Adler et  al. 2012). Nighttime soiling
necessitates use of protective pads in few patients.
A vast majority of the patients are able to with-
R. J. Rintala and M. P. Pakarinen
hold defecation to meet social needs, whereas
around half of them report some imperfections in
distinguishing flatus from stools. Over half of
patients continue to use medications, mainly lop-
eramide, for stool control.
76.1.11.2 Pouchitis
Pouchitis, which refers to idiopathic inflamma-
tion of the ileal reservoir, is the most common
complication after restorative ileoanal pull-­
through for UC (Turner et  al. 2012). At least a
single episode occurs between 30% and 75% of
patients. Although most patients experience one
to a few episodes during the first few postopera-
tive years, the cumulative incidence of pouchitis
increases with extending postoperative follow-up
period (Pakarinen et al. 2009). Pouchitis is classi-
fied as chronic when symptoms last more than
4 weeks. However, it only becomes chronic in a
proportion of patients. Chronic pouchitis may
severely impair bowel function and necessitate
pouch removal in the most refractory cases
(Alexander et  al. 2003). Clinical symptoms
include looser and frequent stools with or with-
out blood, cramping abdominal pain, malaise,
tenesmus, urgency and sometimes fever. The
aetiology of pouchitis remains unclear. Histology
of pouchitis includes acute and chronic inflam-
mation with neutrophilic infiltration, crypt
abscesses and ulcerations (Pakarinen et al. 2010;
Turner et al. 2012). The initial episode of pouchi-
tis should be confirmed by endoscopic examina-
tion and mucosal biopsies (Turner et  al. 2012).
Endoscopic findings of pouchitis include ery-
thema, granularity, friability, bleeding and ulcer-
ations. Faecal calprotectin reflects the degree of
pouch inflammation and may be used for follow-
­up purposes (Pakarinen et al. 2010). Most acute
episodes of pouchitis respond rapidly to a short
course of oral antibiotics, for example metroni-
datzole or ciprofloxacine. Combined antibiotic
therapy or oral budesonide may be used for
chronic pouchitis. Probiotics may have a role in
maintaining remission in chronic pouchitis. The
most refractory cases, especially when associated
with CD, may respond to infliximab.
76  Inflammatory Bowel Disease
76.1.11.3 Fertility and Sexual Function
According to large adult series, the risk of
female infertility increases after restorative
­proctocolectomy (Turner et al. 2018). Although
two long-­term studies of children showed no or
little difference in fertility and sexual function
in relation to healthy controls (Pakarinen et al.
2009; Potter et al. 2020), the possibility of sub-
fertility and sexual dysfunction should be taken
seriously. Following restorative proctocolec-
tomy in childhood, up to 50% of women dem-
onstrated sexual dysfunction when assessed by
validated normative questionnaire (Van Balkom
et al. 2012). Sexual dysfunction was related to
long-term complications of surgery such as pel-
vic infection, fistulas and pouchitis. In contrast,
men reported no sexual dysfunction, impotence
or retrograde ejaculation (Koivusalo et al. 2009;
Van Balkom et al. 2012). Both women and men
have similar sexual function after restorative
proctocolectomy when compared to their con-
trol peers with medically-­treated UC. The rates
of dyspareunia and dysorgasmia were similar
between women who had undergone procto-
colectomy and ileoanal anastomosis or man-
aged with medical treatment for UC (Koivusalo
et al. 2009).
76.1.11.4 Quality of Life
Several studies have indicated that the overall
quality of life among young adults who have
undergone proctocolectomy and ileoanal anasto-
mosis during childhood is comparable to the nor-
mal population (Pakarinen et  al. 2009; Zmora
et  al. 2013). However, surgical treatment and
associated complications will delay education by
causing absence from school and interfere with
social activities at least temporarily in approxi-
mately one-third of patients. Not unexpectedly,
decreased quality of life closely correlates with
impairments in bowel function, high stool fre-
quency and surgical complications underscoring
the importance of successful surgery (Pakarinen
et al. 2009; Van Balkom et al. 2012; Zmora et al.
2013). A well-preserved quality of life and the
fact that the great majority of patients are highly
satisfied with the operative outcome may be con-
1039
sidered surprising, taking into account that many
of them continue to experience recurrent pouchi-
tis episodes, nocturnal bowel motions, occasional
soiling and extraintestinal manifestations of UC,
such as biliary and joint disorders. This is likely
to reflect the remarkably poor quality of life
before ileoanal pull-through (Adler et al. 2012).
76.1.12 Conclusion and Future
Directions
UC in children is a chronic disease of the large
bowel that can be cured by surgically removing
the diseased intestine. Surgery for UC entails
today restorative proctocolectomy in most
patients. Restorative proctocolectomy is, how-
ever, associated with significant early and late
complications and a permanent change in bowel
habits. Moreover, there is little data available on
long-term outcomes of restorative proctocolec-
tomy performed during childhood or adoles-
cence. The functional outcome following
restorative proctocolectomy is generally good,
and the quality of life of operated children and
adolescents equal to that of their healthy peers.
76.2 Crohn’s Disease
76.2.1 Introduction
Like ulcerative colitis, CD is a chronic inflamma-
tion of the bowel which is caused by an interac-
tion of genetic and environmental factors. The
exact aetiology is not known; therefore, there is
no causal therapy available. CD is mainly found
in developed countries, and there is a clear South-­
North gradient, which is especially evident in
Europe (Shivananda et al. 1996). The incidence
of CD in children and adolescents ranges between
1 and 8/100,000 and has risen across Europe in
the past decades (Hildebrand et  al. 2003;
Sawczenko et al. 2001; Turunen et al. 2006). In
20–25% of patients, the disease presents before
the age of 18 years and may even occur in very
young children (age  <  2  years) (Auvin et  al.
1040
2005). In children today, CD is reported to be
more common than UC. Males and females are
equally affected. CD occurs more commonly in
the white population than in other ethnic groups.
There are some hereditary risk factors for CD:
5–20% patients with CD have a first degree rela-
tive who has inflammatory bowel disease, and
offspring of a CD patient have a 10% chance of
developing CD.
There are clear differences between adult- and
paediatric-onset CD in terms of natural history,
the impact of the disease on the patient and the
choice of treatment strategies (Van Limbergen
et al. 2008). The phenotype of CD in the young
has more extensive distribution at presentation
and extension of disease (Vernier-Massouille
et al. 2008). Other typical features of paediatric
CD include growth failure, which is present at
diagnosis in 10–40% of affected children. CD
presenting in childhood and adolescence is com-
monly associated with marked psychological and
social morbidity that may have an impact on edu-
cation, relationships, sexual development and
adherence to therapy (Mackner and Crandall
2005).
76.2.2 Aetiology
The aetiology of CD is unknown and is most
likely multifactorial. In addition to genetic fac-
tors, immunological and microbiological factors,
as well as environmental factors, are likely to
play a significant role.
Family history is a well-known risk factor for
developing inflammatory bowel diseases (CD).
As such, the risk of developing CD has long been
recognized to have a genetic contribution. This
concept has advanced considerably over the past
decade as genetic studies have identified numer-
ous loci involved in IBD susceptibility. The
genetic studies have confirmed the role of muco-
sal barrier function, T-cell subsets and cytokine
signalling in the pathogenesis of CD. These stud-
ies have uncovered new genes and pathways,
including autophagy, interleukin 23 signalling,
NOD2/CARD in innate immunity and innate
lymphoid cells.
R. J. Rintala and M. P. Pakarinen
CD-associated genes in host cells indicate
that altered responses to gut microbiota may be
a primary determinant of disease risk and a
likely mechanism for the disease (Knights
et al. 2013). The diversity and composition of
the gut microbiota are major environmental
factors influencing gut homeostasis. A severe
imbalance in the composition of the gut micro-
biome has been associated with IBD (Morgan
et al. 2012).
76.2.3 Pathology
CD may involve any part of the alimentary
tract. The terminal ileum is predominantly
involved, and in many cases, the disease may
stop at the ileocecal valve. In the colon, it
mainly affects the right side and may spare the
rectum. Patchy or segmental disease can affect
the transverse, descending, and sigmoid colon.
In some cases, the rectum also may be diseased.
The appendix frequently is involved, and peri-
anal disease is present in up to 60% of patients
(see below). When CD involves the upper gas-
trointestinal tract, it generally occurs in con-
junction with ileitis (Kleer and Appelman
2001). CD also may affect the oral cavity, phar-
ynx and esophagus. However, it may also pres-
ent in isolated parts of bowel without evidence
of active disease elsewhere.
CD involves the terminal ileum and the colon
in 60% of cases, the small bowel only in 30% and
the colon only in 10% of cases. The involved
bowel and the mesentery are thickened and fat
often migrates towards the antimesenteric border
of the bowel wall (creeping fat, fat wrapping).
Variable degree of stricturising is commonly seen
in the segment of bowel that is mostly affected.
Skip lesions are areas or patches of active disease
outside the main area of involvement. Skip
lesions are typically found in the small bowel
proximally to the most commonly affected termi-
nal ileum.
Histologically, mucosa is extensively ulcer-
ated and the inflammation is usually transmu-
ral. The inflammation is often interspersed
with almost normally appearing mucosal areas.
76  Inflammatory Bowel Disease
The transmural inflammatory changes may
develop to fistulas that erode to adjacent struc-
tures such as bowel, bladder, vagina, perineum
and abdominal wall. Histological evidence of
granulomas that are the mainstay of histologi-
cal diagnosis of CD occur in 40% of endo-
scopic biopsies taken from lesions in small
bowel, in more than 60% of biopsies from gas-
tric lesions and only 25% of biopsies from
colonic lesions.
76.2.4 Diagnosis
The diagnosis of CD is based on clinical history,
endoscopic findings, imaging and, to a limited
extent, on laboratory tests. All patients with sus-
pected CD need to have proper medical history
that include bowel symptoms, stooling pattern,
general symptoms, exact data on growth during
last years and family history of IBD.  A full
physical examination is mandatory and includes
oral and perianal inspection, measurement of
weight and height and assessment of the puber-
tal stage.
76.2.4.1 Clinical Features
In children, the onset of CD is most often after
the age of 10 years. However, CD may occur at
almost any age; about 5% of CD patients present
before the age of 10 years. There is commonly a
significant diagnostic delay between the onset of
symptoms and the definitive diagnosis. A typical
early symptom of CD is non-specific abdominal
pain. Other typical symptoms include diarrhea,
fever and weight loss. Growth failure may be
manifested by delayed onset of puberty. Patients
with CD are commonly more ill than those with
UC.
In up to 20% of CD patients, perianal disease
occurs as a first presentation. Typical findings are
chronic anal fissures and skin tags and chronic
fistula-in-ano. In a preadolescent child, any of
these should alert the clinician to suspect CD.
Many children with CD have extraintestinal
manifestations. Some of these may present years
before the onset of bowel CD.
1041
76.2.4.2 Laboratory Investigations
Laboratory findings are not diagnostic but usu-
ally helpful. Many patients are anaemic, and
most have elevated sedimentation rate. Serum
albumin, prealbumin and transferrin are low in
many CD patients reflecting their poor nutri-
tional status. Faecal culture, including
Clostridium difficile toxin assay and micros-
copy for parasites, should be performed ini-
tially to rule out infectious causes of bowel
symptoms.
Faecal markers of intestinal inflammation
such as faecal calprotectin and lactoferrin are
useful in the diagnosis and also follow-up of CD,
especially for CD with a colonic involvement. On
the other hand, the role of these surrogate mark-
ers in exclusively small intestine CD is still
unclear, which is unfortunate as CD commonly
presents primarily and only in small intestine.
They can be used in cases of suspected CD for
supporting diagnosis and in cases of already
known CD for confirming relapse. A positive
result could confirm the CD diagnosis or CD
relapse, due to the high sensitivity of the test, but
a negative result should not exclude the disease
(Kostakis et al. 2013).
76.2.4.3 Endoscopy
Endoscopic work-up for CD needs to include
both upper and lower fiberoptic endoscopy.
Granulomas that are diagnostic for CD may be
found more easily from upper gastrointestinal
biopsies than from biopsies from colon. In 10
and 30% of cases, the histological confirmation
of the diagnosis of CD would have been missed
if upper endoscopy was omitted (Castellaneta
et al. 2004). Full colonoscopy with ileal intuba-
tion is required for proper lower gastrointesti-
nal tract work-up (Cameron 1991). Multiple
biopsies should be obtained especially from
macroscopic lesions. Typical findings in CD
include segmental aphtous and linear ulcers and
stenotic areas that sometimes do not allow the
passage of an endoscope. Typical for CD is that
the severely affected segment of bowel may
rapidly change to an almost normal looking
bowel. In paediatric CD, it is not uncommon
1042
that the whole length of the colon may be
affected by the disease as in most cases of
childhood UC.  Even in cases of extensive
colonic CD, the rectum may be completely or
partially spared.
Capsule endoscopy is considered as a diag-
nostic investigation in suspected small bowel CD
if endoscopy and other imaging have not con-
firmed the diagnosis. Capsule endoscopy is a
noninvasive method of endoscopic imaging that
can be swallowed by the patient or, in young chil-
dren, delivered into the duodenum with endos-
copy. The capsules are propelled by peristalsis.
Images are transmitted by radiofrequency to a
sensor array and downloaded to a workstation to
be viewed. The main advantage of capsule endos-
copy is the ability to visualize the entire small
bowel with minimal discomfort to the patient.
The main limitations are the lack of bioptic and
therapeutic capabilities, the potential to miss sin-
gle lesions, high rate of incidental findings, and
difficulty in localizing identified lesions. Capsule
retention is possible although rare, which pre-
cludes the use of capsule endoscopy in patients
with suspected obstruction or strictures (Di
Nardo et al. 2012).
Balloon-assisted full enteroscopy is com-
monly used today in adults with suspected small
bowel lesions but has not been used widely in the
diagnostic work-up of paediatric IBD. The main
limitations of enteroscopy are the invasive nature
of the procedure with the risk of bleeding and
perforation, prolonged duration and requirement
for specially trained endoscopy team that is not
usually available in paediatric services (Di Nardo
et al. 2012).
76.2.4.4 Imaging
MRI enterography (MRE) is currently the modal-
ity of choice to evaluate small bowel involvement
in CD (Maaser et al. 2019). A recent systematic
review and meta-analysis has demonstrated that
MRE has at least as good sensitivity and specific-
ity as contrast follow-through for the detection of
small bowel pathology in paediatric CD and
without exposure ionising radiation. Intestinal
ultrasound can be used to evaluate the transmural
R. J. Rintala and M. P. Pakarinen
involvement of the bowel wall, but the interpreta-
tion requires experience and is operator
dependent.
76.2.4.5 Differential Diagnosis
The major differential diagnostic problem in CD
is UC.  Usually, it is easier to rule out UC in a
patient with CD than CD in a child with
UC. Indeterminate colitis also occurs in children
but in less than 10% of the cases. Indeterminate
colitis is a distinct paediatric subgroup of IBD
with a higher prevalence than in adults. Children
with indeterminate colitis have an early age of
onset and a disease that rapidly progresses to
pancolitis. A significant percentage of patients
with indeterminate colitis have their disease
reclassified as CD or UC.
From a surgical point of view, UC should be
definitively ruled out since today the gold stan-
dard for surgical management of UC is restor-
ative proctocolectomy that is not a suitable
surgical option for CD.  CD typically presents
with general symptoms such as fever and unspe-
cific abdominal pain more often than UC. Growth
failure and delayed sexual maturation are also
more common in CD. Acquired causes of bowel
symptoms, such as infectious and parasitic gas-
troenteritis and especially Clostridium difficile
infection as a cause of symptoms, have to be
ruled out. There are also very uncommon condi-
tions such as Behcet’s syndrome and chronic
granulomatous disease that may mimic CD.
Typical clinical symptoms in CD and UC are
summarized in Table 76.2.
76.2.5 Extraintestinal Manifestations
Extraintestinal manifestations of CD are com-
mon and affect a variety of different organs. At
least one extraintestinal manifestation affects
up to 30% of paediatric patients (Dotson et al.
2010). Some of them associate temporally with
intestinal inflammation activity, while others
have an independent clinical course. Peripheral
arthritis, erythema nodosum and episcleritis
are related to disease activity, whereas axial
76  Inflammatory Bowel Disease
arthropathy, pyoderma gangrenosum, uveitis
and primary sclerosing cholangitis have an
independent course (Van Assche et  al. 2010).
Impaired intestinal absorptive function and
bowel resections essentially contribute to
development of some non-­ inflammatory
extraintestinal disorders associated with CD
such as nephrolithiasis and metabolic bone dis-
ease. In children, the majority of extraintesti-
nal manifestations occur within the first year
after onset of CD, while the presence of any
extraintestinal manifestation, arthalgia and
erythema nodosum are increasingly common
among patients with high disease activity at
diagnosis.
Arthritis is less common than arthalgias, but
their combination constitutes the most com-
mon (20%) extraintestinal manifestation in
young CD patients. Arthritis is also the most
common presenting extraintestinal clinical
manifestation preceding development of intes-
tinal disease. Almost 10% of patients have aph-
thous stomatitis. Erythema nodosum is the
most common cutaneous lesion characterized
by raised, tender and red subcutaneous nod-
ules, particularly on the extensor surfaces
occurring in few percent of patients. Pyoderma
gangrenosum presents as erythematous pap-
ules or pustules with subsequent dermal necro-
sis and ulcerations. Ocular manifestations are
less common than in adults. Primary sclerosing
cholangitis occurs less commonly than in UC
and is often detected by raised liver enzyme
values on screening. Acute pancreatitis occurs
rarely and may be related to drug therapy (aza-
thioprine, mercaptopurine and 5-ASA), pri-
mary sclerosing cholangitis or biliary stone
disease. In ileal dysfunction caused by active
disease or surgical resection, malabsorption
predisposes to oxalate urinary tract stones.
Decreased bone mineral ­density is a frequent
finding in newly diagnosed children (Van
Assche et  al. 2010). Active intestinal inflam-
mation, impaired nutrition and corticosteroid
therapy predispose to metabolic bone disease,
which demands monitoring with densitometric
scanning (DEXA).
1043
Table 76.2  Incidence of different symptoms in Crohn’s
disease and ulcerative colitis
UC (%) CD (%)
Bloody stools 97 22
Diarrhoea 90 88
Abdominal pain 33 82
Weight loss 15 60
Fever 15 77
Growth failure 3 30
76.2.6 Medical Treatment
The full extent, location and severity of CD
should be established at diagnosis before starting
medical treatment, because the initial treatment is
guided by disease severity and distribution (van
Rheenen et  al. 2020). The medical treatment of
paediatric CD has shifted towards more aggres-
sive approach at presentation, aiming at enhanced
immunomodulation at the early stage of disease,
higher rate of mucosal healing and more effective
prevention of complications associated with dis-
ease progression in children with long life
expectancy.
Both systemic corticosteroids and exclusive
enteral nutrition are used for the induction of remis-
sion. Enteral nutrition has fewer side effects but
may be less effective in severe disease and is com-
plicated by poor compliance due to frequent
requirement of nasogastric tube feeding. Most pae-
diatric patients respond to corticosteroids, while
initial steroid resistance is encountered in 11–17%.
Oral budesonide that is released in the ileum is also
effective in mild to moderate ileocecal CD with
fewer side effects. Corticosteroids should not be
used as maintenance treatment in children with
CD, but replaced with thiopurines or anti-TNF
therapy. Thiopurines, azathioprine and mercapto-
purine, have been commonly used to maintain
remission. Alternatively, if they are not tolerated or
are ineffective, methotrexate may be used.
Infliximab is effective for induction of
remission in paediatric CD and recommended
in patients at high risk of complicated disease
course (van Rheenen et  al. 2020). It may be
introduced as the first-line treatment or when
1044
standard induction therapy is ineffective or not
tolerated. Regular infliximab infusions are
effective therapy for maintenance of remission
in paediatric CD, in fistulizing (perianal) dis-
ease and in certain extraintestinal manifesta-
tions. Experiences with anti-TNF drug
adalimubab parallel those of infliximab.
Approximately one-third of patients develop
anti-infliximab antibodies associated with
shortened response duration, acute infusion
reactions and delayed hypersensitivity.
Infliximab therapy is usually combined with
immunomodulator medication (thiopurines,
methotrexate) in order to prevent formation of
antibodies. This approach may increase rate of
toxic side effects and especially T-cell lym-
phoma. In addition to a risk of malignancy,
adverse effects of infliximab treatment include
serious or opportunistic infections, formation
of autoimmune antibodies and psoriasiform
skin lesions. Elective surgical treatment should
be considered in children with disease refrac-
tory to medical treatment especially in children
with unfinished puberty and localized disease
(Van Assche et al. 2010).
76.2.7 Perianal Disease
Children and adolescents with CD are commonly
affected by perianal disease, which is defined as
inflammation of the anus or anal region including
skin tags, fissures, fistulae, abscesses and steno-
sis. Depending on the definition, perianal CD is
observed in up to 62% of patients. Fissures are
encountered most commonly (51%) followed by
skin tags (35%), fistulae (15%) and abscesses
(13%) (Palder et al. 1991).
Skin tags can precede manifestation of intesti-
nal disease, rarely cause symptoms and should be
left alone. Fissures in patients with CD may be
large with raised edges and occur anywhere in
the anal circumference. They can be symptom-
atic but often heal without specific treatment.
Synchronous activation of intestinal disease
should be treated medically. Intersphincteric bot-
ulinum toxin injection or rarely surgical excision
may be considered for refractory symptomatic
R. J. Rintala and M. P. Pakarinen
chronic fissures only after acute inflammation
has subsided.
Presence of perianal fistula guides both medi-
cal and surgical therapy of CD.  A perianal
abscess usually occurs in conjunction with fis-
tula and presents as a tender, sometimes fluctuat-
ing, mass or discharge. Most abscesses are
perianal (60%) or (30%) ischiorectal, while a
few percent are pelvirectal above the levator ani
muscle. Most fistulas involve or traverse the
sphincteric muscles while suprasphinteric fistu-
las running cranial to the anal sphincters and
superficial subcutaneous fistulas occur more
rarely. A simple fistula involves a single, low
intersphincteric or transphincteric location with
an internal and an external opening close to the
anal verge. A complex fistula has more substan-
tial sphincter involvement; it may have multiple
tracts or branching with an internal opening
above the dentate line and an external opening
further away from the anus with or without asso-
ciated abscess. Excluding simple subcutaneous
fistulas, perianal fistulas are best assessed by
MRI prior to any surgical interventions. MRI
provides detailed information on location, pos-
sible branching and number of fistulas and delin-
eates presence of associated abscesses, guiding
subsequent medical therapy and surgical drain-
age procedures.
Examination under anesthesia is combined
with endoscopic examination. The entire anal
region and genital area is inspected carefully
and examined digitally. Abscess cavities are
incised and drained. All fistula openings and
suspicious dimples are gently probed to estab-
lish connection between the rectum and ano-
derm. Superficial fistulous tracts can be laid
open. Loose noncutting silicon seton bands are
placed in tracks of more complex fistulas to pre-
vent premature closure of the fistula and recur-
rence. Infliximab therapy can be given solely or
started at the time or shortly after seton place-
ment, which are left in place for several months.
Both treatment modalities are effective, but the
combined therapy with infliximab and setons
appears to provide higher response rates
(Hukkinen et al. 2014). Resolving of the fistula
tract takes much longer than healing of the
76  Inflammatory Bowel Disease
external orifice, and the safest approach is to
confirm healing of fistulas with MRI before
seton removal. Setons are tolerated well and do
not compromise faecal continence (Hukkinen
et  al. 2014). Some patients with complicated
extensive abscesses or proximal fistulas benefit
from additional temporary faecal diversion for
6–12  months. More complex surgery, such as
advancement flaps, remains rarely indicated and
should be performed in the absence of inflam-
mation with a covering ostomy.
Rectal stenosis or strictures occur rarely in
children with CD. They result from circumferen-
tial bowel inflammation, usually located at the
dentate line. Most patients respond to repeated
anal dilatations under anesthesia. A refractory
short distal stricture may be amenable for ran-
dom flap stricturoplasty. More extensive stric-
tures may require faecal diversion, with or
without proctectomy.
76.2.8 Surgical Treatment
76.2.8.1 Principles and Indications
Although new modalities of medical treatment
have been developed, half of the patients with
childhood onset will eventually undergo sur-
gery before the age of 30  years. In the USA
between 1997 and 2009, the incidence of intes-
tinal resections for paediatric CD exceeded
those of UC (1.0 vs 0.6/100.000 children). The
high number of resections reflects the fact that
CD cannot be cured by surgery and, because of
high the rate of recurrence, re-­operations are
common.
The indications for elective or semi-elective
surgery in 60–70% of patients are continuing dis-
ease activity and growth failure unresponsive to
optimal medical treatment, in 10–25% obstruc-
tive symptoms, whereas 20–25% undergo emer-
gency surgery for toxic megacolon, bleeding,
perforation and abscess. (Pacilli et  al. 2011;
Piekkala et al. 2013). Surgery is considered as the
primary treatment in patients presenting with
obstructive disease without inflammatory activ-
ity, or obstructive symptoms refractory to opti-
mized medical treatment. Typical surgical
1045
procedures are segmental resection of the small
bowel, ileocolic resection, segmental or subtotal
resection of the colon and total proctocolectomy.
In surgery for CD, preservation of as much of the
bowel as possible should be the guiding
principle.
76.2.8.2 Timing
Paediatric CD is associated with significant nutri-
tional and growth impairment resulting in delayed
puberty, and it is important to induce remission
before the onset of puberty. If the onset of puberty
is significantly delayed, the patient’s final height
may be compromised. Surgery is generally advo-
cated in prepubertal and pubertal children with
growth failure despite optimized medical treat-
ment (Amil-Dias et al. 2017). If surgical remis-
sion is achieved after puberty, no catch-up growth
occurs.
76.2.8.3 Preoperative Assessment
Paediatric CD often presents as colitis without
small intestinal involvement, rendering differen-
tial diagnosis between CD and UC challenging.
In order to reach the correct diagnosis, careful
assessment of both colonic and extracolonic fea-
tures, endoscopic histology, serologic markers,
capsule endoscopy and MRE is of utmost impor-
tance (Levine et al. 2014). Nutritionally compro-
mised patients or patients with recent major
weight loss, anaemia or hypoalbuminemia
should undergo a period of preoperative nutri-
tional support. Preoperative weaning from ste-
roids and TNF-α antibodies may be beneficial in
reducing surgical complications (Amil-Dias
et al. 2017). In patients who are malnourished or
cannot be weaned from high doses of steroids,
staged procedures with enterostomy should be
considered.
Preoperative discussion with the patients
and the parents should include all aspects and
options of the surgical therapy. Reliable infor-
mation should be given on the goals of the
planned surgery, postoperative complications,
expected high probability of a recurrent disease
and surgery. Consultations with stoma nurse
and dietitian should be arranged when
necessary.
1046 R. J. Rintala and M. P. Pakarinen

76.2.8.4 Operative Approach Small bowel resection is indicated in

and Technique patients with perforation, fistula formation or
extreme bowel wall thickening that precludes
Duodenum stricturoplasty, contiguous stricturing for
Primarily, duodenal CD is rare, but it may cause 30 cm or more of intestinal length and ileoce-
obstructive symptoms or ulcer-like hemorrhage. cal involvement. Endoscopic dilatation is a
Symptoms are initially managed medically with valid technique in endoscopically accessible
anti-TNF agents and proton pump inhibitors, but short strictures especially for recurrence after
eventually, the majority of affected patients may ileocolic resection (Ono et  al. 2012; Dignass
need surgical intervention. Mild recurrent et al. 2010).
obstructive symptoms can be successfully treated
with endoscopic balloon dilatation. In the case of Large Intestine
more severe symptoms, gastrojejunal bypass Isolated colonic CD is identified in 18–42% of
with or without duodenal stricturoplasty is rec- children at the time of surgery, and it is the most
ommended, whereas duodenal resections are common presentation in children aged 8 years or
associated with a high complication rate. less. Colonic involvement is often extensive and
aggressive and after segmental resections recur-
Small Intestine rent disease occurs almost invariably. In severe
The surgical approach in jejunal, ileal or ileocolic Crohn’s colitis with anorectal involvement, the
resections may be open or laparoscopic assisted. lowest recurrence rate is achieved with total
In adults, the advantages of laparoscopy: shorter proctocolectomy and permanent ileostomy.
hospital stay, reduced complication rate, better However, the majority of children and adoles-
cosmesis and reduced rate of intra-abdominal cents are reluctant to approve permanent
adhesions have emerged as definite benefits of ileostomy.
laparoscopy. In patients with previous open sur- Expert opinions are divided on whether local
gery, fistulas and abscess open approach are pre- colonic disease in children should be treated with
ferred. If long segments of small bowel are segmental resections. After segmental resections,
affected, it is advisable to deal only with the the reoperation rate exceeds 80%, and a signifi-
obstructive segments and leave diseased but non-­ cant percentage of the patients eventually undergo
obstructing lesions and skip lesions behind. colectomy, whereas the rate of permanent ileos-
Disease activity at the anastomotic margins does tomy is similar as after subtotal colectomy.
not compromise the outcome of the anastomosis. Decisions should take individual preferences of
In adults stapled side-to-side anastomosis carries a the patient and surgeon into account (Moir 2007;
somewhat lower complication rate than hand-­sewn van Assche et al. 2010).
end-to-end anastomosis, but whether the same is In the case of extensive colonic involvement,
true in children is unclear. A wide anastomosis subtotal colectomy with ileorectal anastomosis is
irrespective of the anastomotic technique is always a valid option. Despite high rate of clinical recur-
recommended. (Dolgin 2007; Dignass et al. 2010). rences, 10-year rectal sparing rate is 75–86%. In
In selected patients, strictureplasty may be the case of severe rectal inflammation, an ileos-
successfully used in short jejunoileal strictures tomy and a staged ileorectal anastomosis is per-
(<10  cm). Heineke-Mickulicz stricturoplasty is formed after the rectal inflammation is medically
suitable for single short (<10 cm) strictures with controlled. Because of a relatively high risk of
a linear diameter of 2  cm or less. In single or anastomotic leakage with the ileorectal anasto-
­ ultiple adjacent ileal or ileocecal strictures, a mosis, a temporary diverting ileostomy is advis-
m
side-­to side isoperistaltic stricturoplasty may be able (Alves et al. 2002).
used instead of resection. A widening stricturo- Anorectal involvement of CD considerably
plasty may also be used in a stenosed ileocolic increases the risk of recurrence after segmental
anastomosis. resections or ileorectostomy. However, as an
76  Inflammatory Bowel Disease
option for proctectomy and permanent ileos-
tomy, medically controlled rectal or perianal
involvement and fistulizing perianal disease are
not absolute contraindications to a staged ileo-
rectostomy. Biological medication may increase
the rate of rectal preservation and reduce the
need of permanent ileostomy (Coscia et  al.
2013).
Ileal pouch anastomosis is not generally rec-
ommended because of a high rate of fistulization
and pouch failure. In highly selected patients
with isolated colonic disease without ileal or per-
ineal involvement, restorative proctocolectomy
with ileal pouch may be considered (Amil-Dias
et al. 2017).
76.2.8.5 Fistulising Non-­perianal
Disease
Fistulae communicating between loops of intes-
tine, intestine and abdominal wall or between
intestine and viscera (urinary bladder, vagina)
may resolve with medical treatment. If surgery is
needed, the surgeon should resect the diseased
intestine from which the fistula arises and over-
sew the fistula opening in the healthy target
organ. In enterocutaneous fistulas, every effort
with medical and nutritional therapy should be
made. The surgeon should avoid re-operations
too early because of a high rate of recurrence and
complications in a malnourished patient. Low
anal-vaginal introitus fistulae may not need surgi-
cal treatment. In rectovaginal fistulae with persis-
tent and unacceptable symptoms, surgery with
diverting enterostomy and advancement flap or
gracilis plasty may be attempted (van Assche
et al. 2010).
76.2.8.6 Postoperative Management
In general, immediate postoperative management
is similar as in patients with UC undergoing
major bowel resections. Endoscopic follow-up
should be arranged, and the first ileocolonoscopy
should occur within one year of the resection.
Alternatively, MRE may be performed (van
Assche et al. 2010).
Postoperative medication aimed to prevent
recurrence should be based on endoscopic
assessment aided by monitoring of faecal bio-
1047
markers. Maintenance therapy with thiopurines
is recommended in patients with moderate risk
of recurrence. Treatment with biologics is
advised when CD patients have a high risk of
postoperative recurrence or have widespread dis-
ease. Failed thiopurine treatment also requires
treatment escalation to biologics (Amil-Dias
et al. 2017).
76.2.9 Complications
Anastomotic leak, haemorrhage, abscess,
obstructions and enterocutaneous fistulae are
the most common complications occurring in
15% of patients after strictureplasty and resec-
tion. Percutaneous radiologic-guided or open
drainage of abscesses may be adequate therapy,
whereas in complicated cases diverting enteros-
tomy may be needed (Dolgin 2007). In the lap-
aroscopic approach, the perioperative
complication rate is comparable with open
operations (Diamond et al. 2010). Postoperative
enterocutaneous fistulae of the small bowel
may initially be treated with a period of paren-
teral nutrition, antibiotics and administration of
subcutaneous somatostatin rather than early re-
resections. Recurring anastomotic strictures or
ulcers may eventually require re-resection (van
Assche et al. 2010).
Intestinal failure is a devastating complication
of CD. An active and stricturing CD, first surgery
at an early age, perioperative complications and
repeated operations for enterocutaneous fistulae,
strictures or enterostomy complications are the
most significant risk factors contributing to intes-
tinal failure (Gearry et al. 2013).
76.2.10 Outcomes of Surgery
for Crohn’s Disease
The rate of postoperative recurrence in children
is very high. In a study including 27 patients
who underwent either resection or stricturo-
plasty with postoperative azathioprine mainte-
nance therapy, the one-year relapse rate was
55% and 18% underwent re-resection (Pacilli
1048
et  al. 2011). Another study with 36 children
undergoing surgery for CD reported 94% rate
of postoperative relapse a median of 1.8 years
after the primary surgical resection and 54%
rate of re-resection in the follow up period of
10 years. The median time from first surgery to
re-resection was 2  years, whereas the median
time without re-­resection was 7 years. Despite a
high recurrence rate, bowel function was
acceptable, and quality of life was comparable
between the patients and their peers (Piekkala
et  al. 2013). Colectomy, repeated abdominal
operations and pelvic infections and fistulae
may have negative effect on sexual functions
(Van Balkom et al. 2012) and cause subfertility
in females.
76.2.11 Conclusion and Future
Directions
CD is an increasing problem in children, espe-
cially in the developed world. In children, CD
is usually more widespread and severe than in
adults. CD commonly presents with signifi-
cant general symptoms and affects growth and
development of the child. The aetiology of CD
is unknown, and there is no permanent cure.
The management of CD is predominantly
medical; surgery is reserved for complications
or recalcitrant and refractory disease. There
are a number of management options in CD,
but it is not known if the modern therapeutic
modalities such as immunomodulatory and
biological drugs have permanently improved
the prognosis of CD or decreased the need of
surgery.
In the future, it is likely that CD will also
increase in developing countries as the standard
of living increases. It is also likely that medical
therapies will develop and treatment strategies
more specific to CD will emerge. This will
require research to reveal the basic aetiological
and pathophysiological mechanisms of
CD. Surgery remains an adjunct to medical treat-
ment modalities, but it is unlikely that the need
for surgery will be completely abolished or even
significantly decreased.
R. J. Rintala and M. P. Pakarinen
References
Adler J, Coran AG, Teitelbaum DH (2012) Ulcerative coli-
tis. In: Coran A, Adzick NS, Krummel TM, Laberge
JM, Shambereger RC, Caldamone AA (eds) Pediatric
surgery, 7th edn. Elsevier, Philadelphia
Alexander F, Sarigol S, Stallion A, Cotman K, Clark H,
Lydzinski B, Fazio V (2003) Fate of the pouch in 151
pediatric patients after ileal pouch anal anastomosis. J
Pediatr Surg 38:78–82
Alves A, Panis Y, Trancart D, Regimbeau JM, Pocard M,
Valleur P (2002) Factors associated with clinically sig-
nificant anastomotic leakage after large bowel resec-
tion: multivariate analysis of 707 patients. World J
Surg 26:499–502
Amil-Dias J, Kolacek S, Turner D, Pärregaard A et  al
(2017) Surgical management of Crohn’s disease in
children: guidelines from the paediatric IBD Porto
group of ESPGHAN.  J Pediatr Gastroenterol Nutr
64:818–835
Auvin S, Molinié F, Gower-Rousseau C, Brazier F et al
(2005) Incidence, clinical presentation and location
at diagnosis of pediatric inflammatory bowel disease:
a prospective population-based study in northern
France (1988–1999). J Pediatr Gastroenterol Nutr
41:49–55
Benchimol EI, Guttmann A, To T, Rabeneck L, Griffiths
AM (2011) Changes to surgical and hospitaliza-
tion rates of pediatric inflammatory bowel disease in
Ontario, Canada (1994–2007). Inflamm Bowel Dis
17:2153–2161
Bergquist A, Said K, Broomé U (2007) Changes over
a 20-year period in the clinical presentation of pri-
mary sclerosing cholangitis in Sweden. Scand J
Gastroenterol 42:88–93
Bernstein CN, Blanchard JF, Houston DS, Wajda A
(2001) The incidence of deep venous thrombosis and
pulmonary embolism among patients with inflamma-
tory bowel disease: a population-based cohort study.
Thromb Haemost 85:430–434
Cameron DJ (1991) Upper and lower gastrointestinal
endoscopy in children and adolescents with Crohn's
disease: a prospective study. J Gastroenterol Hepatol
6:355–358
Castellaneta SP, Afzal NA, Greenberg M, Deere H
et  al (2004) Diagnostic role of upper gastroin-
testinal endoscopy in pediatric inflammatory
bowel disease. J Pediatr Gastroenterol Nutr
39:257–261
Coscia M, Gentilini L, Laureti S, Gionchetti P et  al
(2013) Risk of permanent stoma in extensive Crohn’s
colitis: the impact of biological drugs. Color Dis
15(9):1115–1122
Di Nardo G, Aloi M, Oliva S, Civitelli F, Casciani E,
Cucchiara S (2012) Investigation of small bowel
in pediatric Crohn's disease. Inflamm Bowel Dis
18:1760–1776
Diamond IR, Gerstle JT, Kim PCW, Langer JC (2010)
Outcomes after laparoscopic surgery in children
76  Inflammatory Bowel Disease
with inflammatory bowel disease. Surg Endosc
24(11):2796–2802
Dignass A, Van Assche G, Lindsay JO, Lémann M et al
(2010) European Crohn's and colitis organisation
(ECCO). The second European evidence-based con-
sensus on the diagnosis and management of Crohn's
disease: current management. J Crohns Colitis
4(1):28–62
Dolgin SE (2007) Surgical management of upper gas-
trointestinal tract and small bowel Crohn’s disease.
Semin Pediatr Surg 16(3):172–177
Dotson JL, Hyams JS, Markowitz J, LeLeiko NS et  al
(2010) Extraintestinal manifestations of pediatric
inflammatory bowel disease and their relation to dis-
ease type and severity. J Pediatr Gastroenterol Nutr
51:140–145
Feakins RM (2013 Aug 26) Ulcerative colitis or Crohn’s
disease? Pitfalls and problems. Histopathology. [Epub
ahead of print]
Ford AC, Achkar JP, Khan KJ, Kane SV et  al (2011)
Efficacy of 5-aminosalicylates in ulcerative coli-
tis: systematic review and meta-analysis. Am J
Gastroenterol 106:601–616
Gearry RB, Kamm MA, Hart AL, Bassett P et al (2013)
Predictors for developing intestinal failure in patients
with Crohn’s disease. J Gastroenterol Hepatol
28(5):801–807
Geiger JD, Teitelbaum DH, Hirschl RB, Coran AG (2003)
A new operative technique for restorative procto-
colectomy: the endorectal pull-through combined
with a double-stapled ileo-anal anastomosis. Surgery
134:492–495
Gower-Rousseau C, Dauchet L, Vernier-Massouille G,
Tilloy E et al (2009) The natural history of pediatric
ulcerative colitis: a population-based cohort study. Am
J Gastroenterol 104:2080–2088
Hildebrand H, Finkel Y, Grahnquist L, Lindholm J, Ekbom
A, Askling J (2003) Changing pattern of paediatric
inflammatory bowel disease in northern Stockholm
1990-2001. Gut 52:1432–1434
Hukkinen M, Pakarinen M, Piekkala M, Koivusalo
A, Rintala R, Kolho KL (2014) Treatment of com-
plex perianal fistulas with seton and infliximab in
adolescents with Crohn's disease. J Crohns Colitis
8:756–762
Hyams J, Markowitz J, Lerer T, Griffiths A et al (2006)
Pediatric inflammatory bowel disease collaborative
research group. The natural history of corticoste-
roid therapy for ulcerative colitis in children. Clin
Gastroenterol Hepatol 4:1118–1123
Kleer CG, Appelman HD (2001) Surgical pathology of
Crohn’s disease. Surg Clin North Am 81:13–30
Knights D, Lassen KG, Xavier RJ (2013) Advances in
inflammatory bowel disease pathogenesis: linking
host genetics and the microbiome. Gut 62:1505–1510
Koivusalo A, Pakarinen MP, Natunen J, Ashorn M,
Rintala RJ, Sipponen T, Kolho KL (2009) Sexual
functions in adulthood after restorative proctocolec-
tomy for paediatric onset ulcerative colitis. Pediatr
Surg Int 25:881–884
1049
Kostakis ID, Cholidou KG, Vaiopoulos AG, Vlachos IS,
Perrea D, Vaos G (2013) Fecal calprotectin in pediatric
inflammatory bowel disease: a systematic review. Dig
Dis Sci 58:309–319
Kovacs M, Muller KE, Arato A, Lakatos PL et al (2012)
Hungarian IBD registry group (HUPIR). Diagnostic
yield of upper endoscopy in paediatric patients with
Crohn’s disease and ulcerative colitis. Subanalysis of
the HUPIR registry. J Crohns Colitis 6(1):86–94
Larsen S, Bendtzen K, Nielsen OH (2010) Extraintestinal
manifestations of inflammatory bowel disease: epi-
demiology, diagnosis, and management. Ann Med
42:97–114
Lehtinen P, Ashorn M, Iltanen S, Jauhola R, Jauhonen
P, Kolho KL, Auvinen A (2011) Incidence trends of
pediatric inflammatory bowel disease in Finland,
1987–2003, a nationwide study. Inflamm Bowel Dis
17:1778–1783
Levine A, Griffiths A, Markowitz J, Wilson DC et  al
(2011) Pediatric modification of the Montreal classifi-
cation for inflammatory bowel disease: the Paris clas-
sification. Inflamm Bowel Dis 17:1314–1321
Levine A, Koletzko S, Turner D, Escher JC et al (2014)
The ESPGHAN revised Porto criteria for the diagnosis
of inflammatory bowel disease in children and adoles-
cents. J Pediatr Gastroenterol Nutr 58:795–806
Linden BC, Bairdain S, Zurakowski D, Shamberger RC,
Lillehei CW (2013) Comparison of laparoscopic-­
assisted and open total proctocolectomy and ileal
pouch anal anastomosis in children and adolescents. J
Pediatr Surg 48(7):1546–1550
Loftus EV Jr (2004) Clinical epidemiology of inflamma-
tory bowel disease: incidence, prevalence, and environ-
mental influences. Gastroenterology 126:1504–1517
Maaser C, Sturm A, Vavricka SR, Kucharzik T et al (2019)
ECCO-ESGAR guideline for diagnostic assessment
in IBD part 1: initial diagnosis, monitoring of known
IBD, detection of complications. J Crohns Colitis
13:144–164
Mackner LM, Crandall WV (2005) Long-term psycho-
social outcomes reported by children and adolescents
with inflammatory bowel disease. Am J Gastroenterol
100:1386–1392
Moir CR (2007) Surgical management of Crohn's colitis.
Semin Pediatr Surg 16(3):178–184
Morgan XC, Tickle TL, Sokol H, Gevers D et al (2012)
Dysfunction of the intestinal microbiome in inflam-
matory bowel disease and treatment. Genome Biol
13:R79
Mortellaro VE, Green J, Islam S, Bass JA, Fike FB, St
Peter SD (2011) Occurrence of Crohn's disease in
children after total colectomy for ulcerative colitis. J
Surg Res 170(1):38–40
Nuutinen H, Kolho KL, Salminen P, Rintala R,
Koskenpato J, Koivusalo A, Sipponen T, Färkkilä M
(2011) Capsule endoscopy in pediatric patients: tech-
nique and rsults in our first 100 consecutive children.
Scand J Gastroenterol 46:1138–1143
Nyholm I, Hukkinen M, Koivusalo A, Merras-Salmio
L et  al (2019) Long-term single-Centre outcomes
1050
after Proctocolectomy with Ileoanal anastomosis
for Paediatric ulcerative colitis. J Crohns Colitis
13:302–308
Ono Y, Hirai F, Matsui T, Beppu T et al (2012) Value of
concomitant endoscopic balloon dilation for intes-
tinal stricture during long-term infliximab ther-
apy in patients with Crohn's disease. Dig Endosc
24(6):432–438
Orholm M, Binder V, Sørensen TI, Rasmussen LP, Kyvik
KO (2000) Concordance of inflammatory bowel dis-
ease among Danish twins. Results of a nationwide
study. Scand J Gastroenterol 35:1075–1081
Pacilli M, Eaton S, Fell JM, Rawat D, Clarke S, Haddad
MJ (2011) Surgery in children with Crohn disease
refractory to medical therapy. J Pediatr Gastroenterol
Nutr 52(3):286–290
Pakarinen MP, Natunen J, Ashorn M, Koivusalo A et  al
(2009) Long-term outcomes of restorative procto-
colectomy in children with ulcerative colitis. Pediatrics
123:1377–1382
Pakarinen MP, Koivusalo A, Natunen J, Ashorn M et  al
(2010) Fecal calprotectin mirrors inflammation of
the distal ileum and bowel function after restorative
proctocolectomy for pediatric-onset ulcerative colitis.
Inflamm Bowel Dis 16:482–486
Palder SB, Shandling B, Bilik R, Griffiths AM, Sherman
P (1991) Perianal complications of pediatric Crohn's
disease. J Pediatr Surg 26:513–515
Piekkala M, Kalajoki-Helmiö T, Martelius L, Pakarinen
M, Rintala R, Kolho KL (2012) Magnetic resonance
enterography guiding treatment in children with
Crohn’s jejunoileitis. Acta Paediatr 101:631–636
Piekkala M, Pakarinen M, Ashorn M, Rintala R, Kolho KL
(2013) Long-term outcomes after surgery on pediatric
patients with Crohn disease. J Pediatr Gastroenterol
Nutr 56(3):271–276
Potter DD, Moir CR, Day CN, Harmsen WS, Pemberton
JH (2020) Fertility and sexual function in women
following pediatric Ileal pouch-anal anastomosis. J
Pediatr Surg 55:59–62
Sartor RB (2005) Role of commensal enteric bacteria
in the pathogenesis of immune-mediated intesti-
nal inflammation: lessons from animal models and
implications for translational research. J Pediatr
Gastroenterol Nutr 40(Suppl 1):S30–S31
Sawczenko A, Sandhu BK, Logan RF, Jenkins H, Taylor
CJ, Mian S, Lynn R (2001) Prospective survey of
childhood inflammatory bowel disease in the British
Isles. Lancet 357:1093–1094
Seetharamaiah R, West BT, Ignash SJ, Pakarinen MP et al
(2009) Outcomes in pediatric patients undergoing
staright vs J pouch ileoanal anastomosis: a multicenter
analysis. J Pediatr Surg 44:1410–1417
Shivananda S, Lennard-Jones J, Logan R, Fear N et  al
(1996) Incidence of inflammatory bowel disease
across Europe: is there a difference between north
and south? Results of the European collaborative
R. J. Rintala and M. P. Pakarinen
study on inflammatory bowel disease (EC-IBD). Gut
39:690–697
Tilney HS, Constantinides V, Ioannides A, Tekkis PP,
Darzi AW, Haddad MJ (2006) Pouch-anal anastomosis
vs staright ileoanal anastomosis in pediatric patients: a
meta-analysis. J Pediatr Surg 41:1799–1808
Turner D, Otley AR, Mack D, Hyams J et  al (2007)
Development, validation, and evaluation of a pediatric
ulcerative colitis activity index: a prospective multi-
center study. Gastroenterology 133(2):423–432
Turner D, Levine A, Escher JC, Griffiths AM et al (2012)
Management of pediatric ulcerative colitis: joint
ECCO and ESPGHAN evidence-based consensus
guidelines. J Pediatr Gastroenterol Nutr 55:340–361
Turner D, Ruemmele F, Orlanski-Meyer E, Griffiths AM
et al (2018) Management of paediatric ulcerative coli-
tis, part 1: ambulatory care – an evidence-based guide-
line from European Crohn’s and colitis organization
and European society of paediatric gastroenterology,
hepatology and nutrition. J Pediatr Gastroenterol Nutr
67:257–291
Turunen P, Kolho KL, Auvinen A, Iltanen S, Huhtala H,
Ashorn M (2006) Incidence of inflammatory bowel
disease in Finnish children, 1987-2003. Inflamm
Bowel Dis 12:677–683
Van Assche G, Dignass A, Reinisch W, van der Woude
CJ et al (2010) The second European evidence-based
consensus on the diagnosis and management of
Crohn’s disease: special situations. J Crohn Colitis
4:63–101
Van Balkom KA, Beld MP, Visschers RG, van Gemert W,
Breuknik SO (2012) Long-term results after restor-
ative proctocolectomy with ileal pouch-anal anasto-
mosis at a young age. Dis Colon Rectum 55:939–947
Van Limbergen J, Russell RK, Drummond HE, Aldhous
MC et  al (2008) Definition of phenotypic character-
istics of childhood-onset inflammatory bowel disease.
Gastroenterology 135:1114–1122
van Rheenen PF, Aloi M, Assa A, Bronsky J et al. (2020)
The medical management of paediatric Crohn’s dis-
ease: an ECCO-ESPGHAN guideline update. J Crohn
Colitis. online ahead of print
Vernier-Massouille G, Balde M, Salleron J, Turck D,
Dupas JL, Mouterde O, Merle V, Salomez JL, Branche
J, Marti R, Lerebours E, Cortot A, Gower-Rousseau
C, Colombel JF (2008) Natural history of pediatric
Crohn's disease: a population-based cohort study.
Gastroenterology 135:1106–1113
Ziring DA, Wu SS, Mow WS, Martín MG, Mehra M,
Ament ME (2007) Oral tacrolimus for steroid-­
dependent and steroid-resistant ulcerative colitis in
children. J Pediatr Gastroenterol Nutr 45:306–311
Zmora O, Natanson M, Dotan I, Vinogard I, Nagar H,
Rabau M, Tulchinsky H (2013) Lon-term functional
and quality of life outcomes after IPAA in children.
Dis Colon Rectum 56:198–204
 Paediatric Small Bowel
Transplantation
77
G. L. Gupte, K. Sharif, and A. J. W. Millar

77.1 Introduction but all recipients died of technical complications,

sepsis or rejection. Optimism for success in the
Intestinal transplantation (ITx) is now a well-­ cyclosporine era was short-lived as most grafts
established treatment for infants and children were lost to rejection. A total of 15 cases were
with chronic intestinal failure (IF) who develop reported between 1985 and 1990 with one long-­
life-threatening complications (central venous term survivor. This child received a neonatal
catheter related sepsis, venous thrombosis and donor intestine and has survived 30  years post-­
intestinal failure associated liver disease (IFALD) transplant. The introduction of more effective
also known as PN-associated liver disease) tacrolimus immune suppression by Starzl in 1990
related to long-term parenteral nutrition (PN). allowed for reports of 60% 1-year graft survival
ITx is a challenging solid organ transplant to by 1993. Currently, 90% 1-year survival and 70%
manage as the intestine has been found to be an to 80% 5-year survival is being reported from
organ with special problems of its own immuno- well-­ established intestinal transplant centres
genicity, its interface with the enteric lumen colo- worldwide.
nized with bacteria and transfer of a heavy load The main causes of intestinal failure in the
of immune competent cells of the donor at the paediatric age group are short bowel syndrome,
time of engraftment. In addition, the conse- dysmotility syndromes and congenital diarrhoeal
quences of ischaemic reperfusion injury are more disease (Table  77.1). PN remains the current
severe compared to other solid organ transplants. standard of care for infants and children with
Sporadic cases of ITx were reported in the 1970s intestinal failure.

G. L. Gupte (*)
Birmingham Children’s Hospital, Birmingham, UK
e-mail: girishgupte@nhs.net
K. Sharif
Birmingham Women’s and Children’s Hospital,
Birmingham, UK
A. J. W. Millar
University of Cape Town and Red Cross War
Memorial Children’s Hospital,
Cape Town, South Africa
e-mail: alastair.millar@uct.ac.za

© Springer Nature Switzerland AG 2023 1051

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_77
1052
Table 77.1  Causes of intestinal failure
Short bowel syndrome
(a) Gastroschisis (atresia and dysmotility).
(b) Necrotizing enterocolitis.
(c) Intestinal atresia.
(d) Mid-gut volvulus.
Older children
 Crohn’s disease
 Mesenteric infarction
 Tumours
 Trauma
Dysmotility
(a) Intestinal aganglionosis- Long segment
Hirschsprung‘s.
(b) Megacystis microcolon hypoperistalsis
syndrome.
(c) Neuronal Intestinal dysplasia.
(d) Hollow visceral myopathy.
(e) Idiopathic intestinal pseudo-­
obstructionCongenital diarrhoeal disease.
(a) Microvillous inclusion disease.
(b) Tufting enteropathy.
(c) Autoimmune enteropathy.
(d) Syndromic diarrhea/Tricho-hepatico-enteric
syndrome.
77.2 Indications
Indications for transplant assessment in children
with short gut and established IF are the potential
lethal complications of PN, namely loss of cen-
tral venous access from thrombosis of central
veins, life-threatening recurrent line infections
and PN-induced cholestasis, which can rapidly
progress to establish irreversible intestinal failure
associated liver disease (Kaufman et al. 2020).
There are relatively few contraindications but
complete loss of venous access, presence of other
systemic malignant diseases, progressive neuro-
logical disorders and untreated immune defi-
ciency syndromes because of the risk of
unrestrained graft versus host disease (GVHD)
are considered absolute. Relative contraindica-
tions would be the degree of other systemic
comorbidities. There is no lower age or size limit
and a history of multiple previous laparotomies is
not a contraindication but both may carry a wait-
ing list mortality risk of nearly 50% in the <10 kg
size recipients because of the shortage of size
matched donors.
G. L. Gupte et al.
77.3 Pre-Transplant Assessment
The pre-transplant workup for ITx focuses on
the assessment of central venous access patency
and establishing the diagnosis of irreversible
intestinal failure by assessing the adequacy of
function of the residual bowel and documenta-
tion of urinary tract anatomy in dysmotility
syndromes because of the frequent association
of bladder and ureteric dilatation in these
patients.
It is important to assess the residual bowel for
intestinal adaptation in children with short bowel
syndrome. Patients with short bowel syndrome,
who have the potential to achieve independent
enteral tolerance through intestinal adaptation
and/or concomitant non-transplant surgery (clo-
sure of stomas, intestinal plication and lengthen-
ing procedures), should undergo procedures to
optimise the residual bowel function before the
development of end stage liver disease and portal
hypertension.
Children with short bowel syndrome (SBS)
who develop irreversible IFALD but show the
potential for adaptation to autonomous enteral
feeding should be considered for isolated liver
transplant in the expectation that the adaptation
process would be facilitated by replacement of a
severely diseased liver with relief of portal
hypertension. However, the decision should be
made by a MDT team experienced in managing
intestinal rehabilitation and intestinal
transplantation.
The ideal donor would be that of a stable
cadaver heart-beating donor (preferably CMV
and EBV negative) 20% smaller than the recipi-
ent. However, because of the shortage of size
match donors, techniques have been developed to
accommodate larger sized donors by either
increasing the size of the recipient abdomen (pre-­
transplant insertion of tissue expanders, use of
prosthetic materials to close abdomen at the time
of transplant, and staged abdominal closure) or
reduce the size of the graft by back-table excision
of part of donor liver and/or intestine in the case
of multivisceral transplants.
77  Paediatric Small Bowel Transplantation
77.4 Types of Transplant
Procedures
Types of transplant depends on the severity of
liver disease and portal hypertension. Different
types of transplant are as follows:
1. Isolated intestine for minimal or mild liver
disease and impaired venous access (Fig. 77.1)
(a) Full length small intestine
(b) Living related segmental small intestine
(c) Small bowel and colon
2. Liver and intestine for moderate to severe
liver disease with evidence of significant por-
tal hyper-tension (Fig. 77.2)
(a) ‘Classical’ separate liver and intestine
(b) ‘En bloc’ liver, bile ducts, duodenum,
head of pancreas and intestine in continu-
ity as a composite graft whole or reduced
a b
Fig. 77.1 A reduced size composite liver and small
bowel transplant using segments 2 & 3 of the donor liver,
small bowel from the duodenum with head of pancreas to
the terminal ileum. Note the tissue expanders (white
arrows in a) used to assist in skin closure of a graft size
1053
3. Multi-visceral transplantation, which
includes liver, intestine and pancreas with an
additional organ i.e. stomach and/or kidney
(Fig. 77.2f).
77.4.1 Living Related Intestinal
Transplantation
Living related intestinal transplantation is an
option to address the deaths on the waiting list
and expand the pool of donors. The proponents of
the procedure claim the advantages of reduced
cold ischemia time, the ability to select recipients
in an optimal condition and the opportunity to
plan a tolerance inducing immunosuppression
protocol. However, the graft per se does not offer
any immunological advantage, develops the same
complications related to intestinal transplantation
too large for the recipient abdominal domain. A diagram
representation of the graft is shown in b. The graft may be
further reduced in size by excision of a segment of mid
jejunum
1054
a b
d e
Fig. 77.2  Different types of grafts used for intestinal fail-
ure (a) isolated intestine (b) isolated liver for SBS (c) liver
and intestine for IFALD (d) composite graft of liver and
intestine with duodenum, bile ducts and head of pancreas
and would still need to be maintained on life-long
immunosuppression. Thus, living related ITx has
not been widely adopted in major ITx centres.
77.5 Techniques
of Transplantation
The procurement of organs is a complex proce-
dure requiring a separate team of experienced
transplant surgeons. The donor is given antibiotic
and antifungal agents enterally by nasogastric
tube as selective gut decontamination along with
standard intravenous antibiotic prophylaxis. The
graft must be handled with great care with in-situ
cold flush and storage using the University of
Wisconsin preservation solution. The donor
organs are procured en bloc, i.e. the
hepatoduodenal-­ pancreas organ cluster is kept
intact although the liver may be appropriately
reduced in size as required. It is advisable to keep
the cold ischaemic time to less than 8  h. Both
G. L. Gupte et al.
c
f
(e) reduced size composite graft (f) multi-visceral graft,
which could include the whole pancreas, stomach and
right colon and even spleen
stomach and colon may be included in the graft.
The recipient operation may be challenging as
there is frequently a history of multiple previous
surgeries leading to dense vascularized adhesions
from portal hypertension. Details of the proce-
dures are well described but in principle, the fore-
gut portal circulation is decompressed with a
portocaval shunt. The diseased viscera are
removed and the appropriate graft is then sutured
in situ. It is acceptable for the venous drainage of
the graft to be to the inferior vena cava in isolated
intestinal transplants. In multivisceral trans-
plants, the ‘clean sweep’ operation may be per-
formed whereby all the abdominal contents
anterior to the main vessels are removed from the
gastric body to the distal colon, including the
duodenum, pancreas, spleen and rest of the intes-
tine. We generally tend to preserve the spleen to
reduce the risk of post-operative infections and
graft versus host disease.
A donor arterial conduit from the infrarenal
aorta is anastomosed to an arterial patch includ-
77  Paediatric Small Bowel Transplantation
ing the orifices of both the coeliac axis and supe-
rior mesenteric artery.
The proximal bowel anastomosis may be a
gastro-jejunostomy or jejuno-jejunostomy in
continuity. The distal bowel continuity is restored
by end to end anastomosis with a proximal vent-
ing stoma where distal colon has normal func-
tion. The stoma is used for biopsy access in the
post-transplant period or the distal end may be
brought out as an end stoma. Stomas may be
closed at a variable time interval after the original
engraftment when the graft has ‘settled in’ immu-
nologically speaking. This is usually not consid-
ered until after the first 6  months to a year
post-transplant with a stable graft in the absence
of rejection.
77.5.1 Immunosuppression in Small
Bowel Transplantation
Tacrolimus, a powerful calcineurin 2 inhibitor,
along with steroids have been the mainstay of
immune suppression protocols. Over the past
decade, the uses of different agents like IL2
blockers (anti-CD25 monoclonal antibodies),
Campath 1-H (anti CD52 monoclonal antibodies)
and Thymoglobulin (rabbit anti-thymocyte glob-
ulin) have been used in the induction regime,
which has resulted in improved long-term out-
come following ITx.
77.6 Medical Complications
Intestinal transplant is a complex and challenging
procedure which requires a highly skilled multi-
disciplinary team to manage these patients. There
are several reported complications associated
with the procedure. Some of these are:
77.6.1 Graft Rejection
Rejection usually occurs within the first few
weeks after transplantation. Clinical features of
rejection can be variable and include increased
stomal output, fever, abdominal pain and mal-
1055
aise. Frequently, it is difficult to distinguish rejec-
tion from viral enteritis. There are currently no
reliable non-invasive markers of rejection. Serum
citrulline, a marker of mucosal mass, has shown
some promise, however, the differentiation
between rejection and infection can make the lev-
els difficult to interpret. The diagnosis of rejec-
tion is made on histopathological examination of
graft tissue from stomal/ endoscopic biopsies.
Mild or moderate rejection is treated by optimis-
ing the immunosuppression and with pulse doses
of intravenous methylprednisolone (10–20  mg/
kg to a maximum of 400 mg/dose). Severe rejec-
tion is notoriously difficult to treat and is associ-
ated with a poor prognosis. Treatment strategies
with thymoglobulin/campath have been tried
with good success. Newer agents, such as vedo-
luzimab (monoclonal antibody against α4β7 inte-
grins), have been attempted to treat rejection with
good effect.
77.6.2 Infections
The intestinal transplant registry reports infec-
tions as the cause in 60–70% of the deaths. There
is a high incidence of infection in the post-­
transplant period. The high incidence of bacterial
infections is likely to be due to bacterial translo-
cation from the gut lumen. Broad-spectrum anti-
biotics are routinely administered during the
early post-transplant period, along with selective
decontamination of the gut with oral antibiotics
and antifungal agents. Infectious enteritis is
reported in up to 40% of recipients with viruses
(adenovirus, rotavirus and calci-virus) and proto-
zoa (GiardiaLamblia, Cryptosporidium). As pre-
viously mentioned, prompt recognition of
infectious enteritis can be difficult as the clinico-­
pathologic picture can mimic acute rejection. The
treatments are diametrically opposite; one
requires increased immune suppression (rejec-
tion) and the other requires reduced immune sup-
pression (infection). Rotavirus and adenovirus
enteritis can trigger episodes of acute moderate
to severe rejection and can result in graft loss.
Cytomegalovirus virus (CMV) and Epstein
Barr virus (EBV) infection are seen more
1056
commonly in the paediatric intestinal trans-
plant population because of a primary infec-
tion (seropositive donor to seronegative
recipient). Different tests are used to make the
diagnosis of CMV infection and can be effec-
tively treated with intravenous gancyclovir.
Post-transplant infectious enteritis, CMV and
EBV infections can lead to significant morbid-
ity and mortality.
77.6.3 Post-Transplant
Lymphoproliferative Disorder
(PTLD)
PTLD is believed to be driven by EBV infection.
The incidence of PTLD following intestinal
transplantation, historically, was reported to be as
high as 30–40%, but with judicious EBV, poly-
merase chain reaction (PCR) monitoring has
reduced to around 10–15%. Children presenting
with recurrent infections, hypoalbuminemia,
anaemia, thrombocytopenia and neutropenia
should alert the physician to perform investiga-
tions to exclude PTLD. EBV viral load measured
by PCR is helpful to indicate ongoing EBV virae-
mia, but does not confirm the diagnosis of
PTLD.  Imaging of the abdomen and the chest
with computed tomography (CT) or magnetic
resonance imaging (MRI) for lymphadenopathy
may be necessary to arrive at the diagnosis.
Confirmation of EBV-related PTLD needs histo-
pathological examination of affected tissue and
grading as per the WHO classification.
Endoscopic procedures along with EBER (EBV
in situ hybridisation) staining of the gut biopsies
may be necessary to confirm the diagnosis in
children without lymphadenopathy. Pre-emptive
reduction of immunosuppression (RIS) in the
context of EBV viraemia has been demonstrated
to prevent the progression to PTLD. Polymorphic
PTLD needs reduction of immunosuppression
and rituximab (anti- CD-20) in the majority of
cases. Monomorphic PTLD can be associated
with a bad prognosis and hence needs chemo-
therapy or cytotoxic T cells (depending on local/
national availability) in addition to RIS and
Rituximab (Chiou et al. 2020).
G. L. Gupte et al.
77.6.4 Graft Versus Host Disease
(GVHD)
Surprisingly, despite the high lymphocyte load
within the intestine, the incidence of GVHD has
been reported to be low, 8.5% from a large single
centre experience. Optimisation of tacrolimus
immunosuppression (i.e. reduction of immuno-
suppression) and small duration of high dose ste-
roid administration is the treatment of choice,
usually leading to a favourable outcome. For ste-
roid refractory GVHD, mesenchymal stem cells
and extra-corporeal photopheresis (ECP) treat-
ment at an earlier stage are essential for a suc-
cessful outcome.
77.6.5 Antibody-Mediated Rejection
(ABMR)
ABMR has been recognised as a possible con-
tributor to acute and long-term graft injury. The
mere presence of donor-specific antibodies
(DSA) does not contribute to the diagnosis, but
additional pathological changes (capillaritis) and
C4d deposition (not specific for intestinal trans-
plants as compared to renal transplant) are needed
to establish the diagnosis.
77.7 Surgical Complications
Surgical complications have a significant impact
on the morbidity and mortality following intesti-
nal trans-plantation (ITx). Reported incidence of
surgical complications following ITx is 40%.
77.7.1 Intestinal Perforation
Intestinal perforation (IP) was historically
reported in 30–40% recipients following intesti-
nal transplantation. This usually occurs within
the first two weeks post-transplant and is some-
times difficult to manage. The use of steroids and
viral infections are thought to be contributory
factors in intestinal perforation but there is no
definite evidence. The symptoms and signs of
77  Paediatric Small Bowel Transplantation
acute peritonitis can be masked as the children
are heavily immunosuppressed and on high dose
steroids, thus needing a high index of suspicion.
Abdominal imaging and contrast studies may be
needed to confirm the diagnosis. Early second
look operations to exclude IP in high-risk patients
with minimal signs are frequently required.
Surgical control of the perforation is achieved
either by primary closure or by insertion of a T
tube into the perforation to create a controlled
enteric fistula. With standardisation of the surgi-
cal techniques and reduction in cold ischemic
time, the incidence of intestinal perforation has
significantly reduced.
77.7.2 Abdominal Compartment
Syndrome (ACS)
Abdominal Compartment Syndrome (ACS) is a
relatively less recognised and an under-reported
complication following ITx. There are several
contributing factors including swelling of the
graft in the first few days following ITx and
reduced en-bloc liver small bowel transplant
(LSBTx) in infants with small abdominal cavities
and contracted scarred abdominal walls. These
patients need skilled surgical management.
Innovative surgical techniques have been devel-
oped which include insertion of tissue expanders
pre-operatively; staged closure utilising a silastic
pouch to contain the graft and shoelace sutures to
the skin, which can be tightened daily to stretch
the skin and achieve a degree of ‘skin creep’.
When it is apparent that the skin can be approxi-
mated without tension, the silastic patch is
replaced by absorbable collagen protheses
(Surgisis Gold® in infants & Permacol® in older
children).
77.7.3 Pancreatic Complications
Historically, a shortage of size matched donors
and high incidence of biliary complications fol-
lowing conventional liver and bowel transplants
(LBTx) resulted in development of innovative
1057
techniques for LBTx. This included the use of
reduced en-bloc liver small bowel transplant
(LSBTx). Some of these children developed pan-
creatitis and pancreatic fluid collection presum-
ably from the cut surface of the donor pancreas.
These collections were usually treated conserva-
tively with a percutaneous drainage tube however
sometimes surgical intervention is required.
Currently, the whole pancreas is preferably
included in the composite graft. This, however,
results in a bigger graft, which in children with a
small abdominal cavity increases the donor to
recipient ratio and difficulties in abdominal clo-
sure, but has resulted in a dramatic reduction in
pancreatic complications. It is important to keep
a high index of suspicion of pancreatitis in chil-
dren, especially those with unexplained abdomi-
nal pain.
77.7.4 Vascular Complications
There is a high incidence of vascular complica-
tion in some of the paediatric solid organ trans-
plantation especially liver and kidney transplant.
This complication has been rarely seen following
ITx using current techniques as vascular recon-
structions are with large arteries.
77.7.5 Other Complications
Other complications such as adhesive intestinal
obstruction, development of biliary sludge with
partial obstruction of donor common bile duct,
wound dehiscence and stomal issues, e.g. pro-
lapse are less frequent but can be successfully
managed with appropriate interventions.
77.8 Nutritional Outcome
and Quality of Life
More than 90% of the survivors can be estab-
lished on full enteral nutrition in the early post-­
transplant period. The median time to establish
children on an oral diet varies as it is dependent
1058
on the pre-transplant oral intake. Children with a
diagnosis of pseudo-obstruction, who have not
eaten normally prior to intestinal transplantation,
find it extremely difficult to establish oral intake
and may need psychological and speech therapy
input. Children do need a close dietetic follow-up
and may need extra caloric requirement over and
above the normal requirements to ensure appro-
priate growth.
In the recent decade, the need for ongoing
psychological and psychiatric support is being
increasingly recognised. Psychological issues, in
particular depression, have already been reported
after other solid organ and bone marrow trans-
plantation (Norsa et al. 2018).
77.9 Outcome
Data from the intestinal transplant registry, as
well as from major intestinal transplant centres
report an improving incidence of 1-year survival
of 90%, 5-year survival of 75% and 10-year sur-
vival of 70% in children with short bowel syn-
drome (Yap et al. 2020; Venick et al. 2020). The
survival in children transplanted for motility dis-
orders and mucosal disorders is less as compared
to those transplanted with short bowel syndrome.
Analysis of data submitted to intestinal transplant
registry has identified the following: hospitalisa-
tion of patient at the time of transplant is still
associated with poor outcome, PTLD and techni-
cal complications contributing to mortality have
reduced, whereas rejection remains the largest
contributor to long-term graft loss.
At Birmingham Children’s Hospital, one hun-
dred intestinal transplants have been done since
1993.
The longest survivor at our hospital is 22 years
post-transplant.
In conclusion, the advances in the surgical
techniques and immunosuppressive strategies
made in the last decade are reflected in the cur-
rent era of improved outcomes following intesti-
nal transplantation. ITx in the next 10 years will
continue to evolve and has the potential to be
considered as an alternative strategy to home PN
in selected patients.
G. L. Gupte et al.
77.10 Conclusion and Future
Directions
In conclusion, the advances in the surgical tech-
niques and immunosuppressive strategies made
in the last decade are reflected in the current era
of improved outcomes following intestinal trans-
plantation. In larger intestinal transplant centres,
the learning curve of experienced teams over the
last two decades with the early recognition of
complications in post-transplant period should
result in less damage to the graft in long term
thus improving long term outcomes. Late com-
plications leading to graft loss due to antibody
mediated rejection, post-transplant lymphopro-
liferative disease and chronic rejection are com-
plications that need to be overcome to improve
long term survival thus establishing intestinal
transplantation as an alternative to home
PN. Intestinal transplantation in the next 10 years
will continue to evolve and has the potential to
be considered as an alternative strategy to home
PN in selected patients.
77.11 Key Points
• Early referral to a transplant centre before
onset of IFALD so that children can be consid-
ered for isolated bowel transplantation, which
may result in better utilisation of donor organs.
• Children under 2 years of age needing a liver
and small bowel transplant have a worse out-
come due to scarcity of size matched donor
organs and high waiting list mortality.
• Innovative surgical techniques of graft size
reduction, use of abdominal tissue expanders
and staged abdominal closure have improved
the chances of small children being consid-
ered for transplantation.
• Improved immune suppression strategies
using induction agents like IL2 blockers and
anti-lymphocyte preparations have reduced
the incidence of acute rejection.
• Advances in both surgical strategies and med-
ical management have resulted in a current
80% 3-year survival following intestinal
transplantation.
77  Paediatric Small Bowel Transplantation
Acknowledgments  We thank Professor J. de Ville de
Goyet for giving permission to publish his figures.
References
Kaufman SS, Avitzur Y, Beath SV et  al (2020) New
insights into the indications for intestinal transplan-
tation: consensus in the year 2019. Transplantation
104:937–946
Norsa L, Gupte GL et al (2018, Jun) Life of patients after
10 years of a successful paediatric intestinal transplan-
tation in Europe. Am J Transplant 18(6):1489–1493
1059
Chiou FK, Beath SV, Morland B, Hartley J, Van Mourik
I, Abdel-Hady M et  al (2020) Comparison of clini-
cal features and outcome of pediatric posttransplant
lymphoproliferative disorder in recipients of small
bowel allograft versus isolated liver transplantation.
Transplantation 104(7):1429–1436
Yap JYK, Roberts AJ, Bines JE (2020) Paediatric intesti-
nal failure and transplantation. J Paediatr Child Health
56(11):1747–1753
Venick RS, Duggan E, Whatley J (2020) Current status
of pediatric intestinal transplantation in the United
States. Curr Opin Organ Transplant 25(2):201–207
 Long-Term Outcomes in Pediatric
Surgery
78
Risto J. Rintala and Mikko P. Pakarinen

78.1 Introduction of patients with congenital defects and those with

The rapid development of pediatric and neonatal
surgery after World War II in many Western
countries led to the formation of pediatric surgery
as an independent specialty. The first pediatric
intensive care units and neonatal surgical units
were opened in the 1950s. These factors led to
rapid changes in the mortality of patients with
congenital malformations and severe surgical
conditions. During the last 50 years, the mortality
severe malformation complexes has continued to
decrease (Figs. 78.1 and 78.2). This development
is due to improved neonatology and pediatric
intensive care, which especially improved treat-
ment possibilities of congenital cardiac defects.
The downside of decreased mortality is that a sig-
nificant percentage of survivors today have per-
manent morbidities and long-term handicaps.
Until recently, there was little information
about long-term consequences of repaired con-

Fig. 78.1  Survival of 100

677 esophageal atresia 1947-1956 (n=100)
90
patients undergoing
80 1956-1960 (n=100)
surgery at Children’s
Hospital, University of 70 1960-1965 (n=101)
Helsinki from 1947 to 60 1965-1971 (n=101)
2007 50 1971-1978 (n=100)
40
1978-1985 (n=86)
30
20 1989-2007 (n=89)
10
0

R. J. Rintala (*) M. P. Pakarinen

Department of Pediatric Surgery, Children’s Hospital, Department of Pediatric Surgery, Hospital for
Helsinki University Central Hospital, Children and Adolescents, University of Helsinki,
Helsinki, Finland Helsinki, Finland
e-mail: risto.rintala@saunalahti.fi e-mail: mikko.pakarinen@hus.fi

© Springer Nature Switzerland AG 2023 1061

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_78
1062
Fig. 78.2 Overall
25
mortality of patients
with anorectal
malformations at 20
Children’s Hospital,
University of Helsinki
from 1946 to 2003 15
10
5
genital defects or acquired pediatric surgical
problems. For pediatric surgeons, the end point
and final outcome measure are the functional out-
come in an adult patient. Recent research has
indicated that a significant proportion of patients
suffer from abnormal organ functions during
childhood and many of these abnormalities are
carried on to adulthood. Often, these have a sig-
nificant impact on quality of life. Today, long-­
term functional results and quality of life are as
important outcome measures as early mortality
and morbidity.
78.2 Justification of Long-Term
Follow-Up in Pediatric
Surgery
Long-term outcome of pediatric surgery con-
cerns several interest groups. The patients’ par-
ents need to get a realistic picture of what is going
to happen to their child who has undergone major
surgery. The information has to be as accurate as
possible without giving inappropriately positive
expectations concerning functional outcomes.
The patient himself/herself needs to get reli-
able information, as early as this can be given, of
potential problems during later life. The infor-
mation needs to include clarification of manage-
ment modalities available to treat these problems.
The parents may adapt to the handicaps of the
patient and may not consider these as major
functional problems. On the other hand, the
R. J. Rintala and M. P. Pakarinen
1946-1953
1953-1963
1963-1973
1973-1983
1983-1993
1993-2003
patients may experience these handicaps as
major factors that limit their social activities. For
example, even minor soiling in a teenager with
an anorectal malformation (ARM) may segre-
gate the patient from many social activities, such
as sports and overnight visits to friends, although
the parents may consider that the patient’s conti-
nence has greatly improved since earlier
childhood.
The surgical team caring for children needs
long-term follow-up data to guide clinical prac-
tice. Particularly in neonatal surgical conditions,
the final functional outcome is not evident until
the child has reached adolescence or adulthood.
Longitudinal follow-up studies are required to
determine the natural history of a neonatal surgi-
cal condition. A profound knowledge of possible
complications that may develop during the devel-
opmental period of a surgical patient may guide
primary treatment and definitely modify the fol-
low-­up of the patient.
Health care administration needs to be aware
of the consequences and costs of pediatric sur-
gery. Although the numbers are not high, the
management costs of a newborn surgical patient
are very high. There is a need for highly special-
ized intensive care facilities. In many cases, the
intensive care period may be long, especially in
patients who have major neonatal surgical defects
combined with cardiac abnormalities. On the
other hand, the mortality of practically all neona-
tal surgical conditions has decreased significantly
during the last decades. This implicates that more
78  Long-Term Outcomes in Pediatric Surgery
sick children survive with more handicaps. It is
likely that these surviving children with serious
primary conditions have more needs for medical
care during their growth period than those with a
less stormy start.
78.3 How Do We Study Long-Term
Outcomes?
One of the main problems in assessing long-term
outcomes in pediatric surgical patients is the lack
of standardization of surgical procedures.
Although pediatric surgical conditions are usu-
ally well-defined and classified, there is no stan-
dardization of procedures used to repair
congenital or acquired surgical problems in chil-
dren. It is likely that the type of surgical p­ rocedure
plays a significant role, but this is very difficult to
demonstrate.
The length of the follow-up period is a crucial
factor. The most valid end point is the outcome
beyond childhood, at maturity. There is, however,
paucity of long-term follow-up data in adults
with congenital defects; therefore, short-term
studies, often with variable age ranges of patients,
have been used as measures of long-term out-
come. These kinds of studies often have signifi-
cant methodological problems. Commonly, there
has been a lack of a healthy control population.
Healthy controls with similar age and sex distri-
bution as the patients are crucial for reliable out-
come analysis. Most of the long-term follow-up
studies have been cross-sectional, which does not
give completely valid data on the natural longitu-
dinal development of the outcomes of pediatric
surgical conditions.
Long-term outcome data, in terms of function
and quality of life, is usually based on clinical
history and examination. A major problem is the
validity and agreement of the information
attained from children and parents. Reliable data
can only be collected from older children. The
parents may not be able to provide truly reliable
functional data for several reasons. They may not
want to report unfavorable results to a surgeon
who has been responsible for the treatment of
1063
their child. Therefore, it is important that when
functional data is collected personally or by tele-
phone interviews, the person performing the que-
ries has not been involved in the actual
management of the patient. The parents may also
ignore minor and moderate functional defects in
a child who has had a congenital problem from
birth or, in the case of smaller children, may con-
sider them to be part of normal functional
maturation.
The low incidence of neonatal surgical condi-
tions poses a significant problem for outcome
analysis. Institutional series are commonly too
small to allow reliable comparisons between dif-
ferent management modalities. Pooling data from
multiple centers could also be a powerful tool for
long-term outcome studies, but there is an
increased opportunity for bias. Standardized
management and follow-up protocols are very
difficult to set up, and long-term outcome studies
commonly require many years of follow-up that
is very difficult to organize in a multi-center
setting.
The study design in the evaluation of long-­
term outcomes is inherently problematic.
Randomized controlled trials with enough power
are usually unavailable for long-term outcome
studies; the patient series are too small and there
are too many confounding factors, such as vari-
able surgical skills and techniques, inhomoge-
neous patient material and problems in recruiting
patients for follow-up visit/examinations.
Observational studies are more commonly used
for analysis of long-term outcomes. These can
be performed prospectively, retrospectively or as
a cross-sectional study. Prospective matched,
controlled cohort studies are powerful in provid-
ing information on natural history of a congeni-
tal defect or disease. Case control studies are
useful for cross-sectional studies, e.g., gastroin-
testinal morbidity in adult esophageal atresia
(EA) patients. The critical point in case control
studies is the matching of controls (Killelea et al.
2006).
Recently, measures to assess the functional
status and quality of life have become widely
available also for children. These instruments are
1064 R. J. Rintala and M. P. Pakarinen

designed to measure health-related quality of life 78.4.1 Long-Term Outcome

(HRQoL) and today are validated for many lan- in Children with Congenital
guages and cultural backgrounds. Quality of life Diaphragmatic Hernia
instruments can usually be divided in two catego-
ries: generic instruments that assess overall Including abortion, stillbirths and pre-hospital
HRQoL and disease-specific instruments that are deaths, the total mortality caused by congenital
validated and available for patients with, for diaphragmatic hernia (CDH) is estimated at close
example, renal diseases, gastrointestinal disease, to 50% (Stege et al. 2003). This is mainly owing
etc. A typical generic HRQoL instrument is to the new ventilatory strategies the survival of
SF-36 (Ware and Sherbourne 1992) that has 36 congenital diaphragmatic hernia (CDH) has
items and assesses 7 domains. SF-36 is not vali- improved considerably from 50% in 1990 to over
dated for children under 16 years of age; there- 85% in high-volume centers (Waag et al. 2008).
fore, other instruments need to be used. Of the Consequently, the group of CDH survivors
commonly used pediatric HRQoL generic instru- includes an increasing number of children with
ments, Child Health Questionnaire (CHQ) high-risk diaphragmatic hernia and a history of
(Killelea et  al. 2006) includes 87 items and 11 extensive therapeutic measures including vasoac-
dimensions, and Pediatric Quality of Life tive drugs, advanced ventilatory support and
Inventory (PedsQL) (Varni et al. 2007) that is a extracorporeal membrane oxygenation with
very flexible and easy to use tool has 23 items in exposition to prolonged periods of reduced brain
four domains. Both of these come in several oxygenation. Long-term morbidity includes
forms for different age groups and also for chronic respiratory disease, neurodevelopmental
parents/caregivers. A typical disease-specific
­ problems, gastroesophageal reflux and musculo-
instrument is gastrointestinal quality of life index skeletal disorders. Considering long-term studies
(GIQLI) that has 36 items in five domains in adult CDH survivors, it should be borne in
(Eypasch et al. 1995). A significant problem with mind that the majority of the patients may not
the generic validated instruments is that they are have been high-risk cases by today’s criteria.
relatively crude and do not take into account spe-
cific outcome variables of the commonly rare 78.4.1.1 Chronic Respiratory Tract
congenital surgical conditions. Disease
Insufficient airway branching during fetal devel-
opment of the lungs causes hypoplasia of the
78.4 Long-Term Outcomes ipsilateral and contralateral lungs with reduced
of Specific Pediatric Surgical gas exchange surface, abnormal sacculo-alveolar
Conditions maturation and abnormal, muscularized distal
arterioles (Moya et al. 1995). Lung function can
The long-term outcome of defects that are life-­ be further compromised during the initial treat-
threatening in the newborn period is influenced ment by oxygen toxicity, barotrauma and volume
by the severity of the anatomy, success of the trauma. A large proportion of the survivors have
anatomic reconstruction and occurrence of oper- broncho-pulmonary dysplasia (van den Hout
ative complications. Associated malformations et al. 2010), and some require oxygen treatment,
have a significant impact on the outcome, too. even years after discharge.
There is, however, scarcity of long-term follow- Recent studies indicate that in CDH patients,
­up data of many pediatric surgical conditions. In postnatal lung growth is impaired and does not
the following section, examples of surgical con- fully compensate the initial lung hypoplasia and
ditions with recent consistent long-term follow- after infancy pulmonary ventilation increases in
­up data are considered, illustrating the factors relation to perfusion. When the patients grow into
that are involved in the outcome, especially adulthood, the total lung volume increases, but
beyond childhood. there is a slight deterioration in pulmonary diffu-
78  Long-Term Outcomes in Pediatric Surgery
sion capacity. A similar deterioration of lung
function is also observed in critically ill, non-­
CDH children, indicating that the deterioration is
attributed not only to residual sequelae of CDH
but also to intensive care treatment (van den Hout
et  al. 2010; Hayward et  al. 2007; Spoel et  al.
2013). Lung function tests in adult CDH survi-
vors show restrictive, obstructive or combined
ventilatory impairment in 50% and bronchial
hyperreactivity in 35% of cases. In the ipsilateral
lung, vital capacity, ventilation, perfusion and
volume reach approximately 50% of the expected
value. The degree of ventilatory impairment and
bronchial hyperreactivity correlates with the ini-
tial clinical severity of the affliction. Long-term
respiratory problems seem to be particularly
severe in patients treated with extracorporeal
membrane oxygenators (ECMO) and long peri-
ods of artificial ventilation and with patch closure
of the diaphragm (van den Hout et  al. 2010;
Hayward et  al. 2007; Spoel et  al. 2013).
Approximately 30–40% of the survivors have
chronic lung disease, 12% had asthma, and 7%
reported an increased susceptibility to respiratory
infections. Subjective physical performance is
below average in 10–13% of the patients (Spoel
et al. 2013; Vanamo et al. 1996a; Koivusalo et al.
2005a). In Finnish adult CDH survivors, (median
age 39 years, mostly low-risk patients), scores in
the Respiratory Symptoms–Related Quality of
Life questionnaire were similar to those of
healthy controls, and only 2% had decreased
quality of life because of respiratory symptoms
(Koivusalo et  al. 2005a). Respiratory problems
will likely remain a severe issue when high-risk
CDH patients surviving today reach their adult-
hood. In all CDH patients, a long-term respira-
tory follow-up extending into adulthood by a
pulmonologist is indicated. Education and early
intervention of the harmful effects of smoking on
lung function and encouragement on physical
activity are of utmost importance.
78.4.1.2 Gastroesophageal Reflux
Disease
A significant percentage of CDH patients (20–
81%) suffer from gastroesophageal reflux disease
1065
(GERD) following surgical repair. GERD after
CDH repair can cause significant respiratory and
nutritional problems and be unmanageable with
medical therapy. In 26 Finnish CDH patients
treated between 1990 and 2006, significant early
GERD developed in 27% and 15% required anti-­
reflux surgery within 6  months of the repair.
GERD symptoms may also manifest later in life.
Within 10  years after the repair, 53% of the
patients eventually developed significant GERD
defined by pathologic 24 h esophageal pH moni-
toring, endoscopically diagnosed moderate or
severe esophagitis or the need for anti-reflux sur-
gery (Koivusalo et  al. 2008a). Another Finnish
study assessed 60 adult CDH survivors for GERD
from 1948 to 1980. Early GERD developed in
18% of the patients and two (3%) patients devel-
oped a reflux-associated esophageal stricture and
eventually underwent fundoplication. Thirty
years after the repair, 63% had symptomatic
GERD or endoscopically diagnosed esophagitis,
and Barrett’s esophagus was found in 13% of the
patients (Vanamo et  al. 1996b). Although CDH
survivors were found to have tenfold more GERD
symptoms compared with a control population,
the effect of GERD symptoms on quality of life
was limited (Koivusalo et al. 2005a). Long-term
surveillance for CDH-associated GERD should
extend beyond childhood, and a targeted endo-
scopic surveillance program is recommended.
78.4.1.3 Failure to Thrive and Grow
Failure to thrive and impaired growth is common
among CDH survivors. The etiological factors
behind a failure to thrive include increased respi-
ratory work, gastroesophageal reflux (GER) and
aversion to feeding. To obtain an adequate caloric
intake, a substantial number of patients require
feeding gastrostomy and fundoplication. Despite
aggressive nutritional therapy, weight may be
below the 25th percentile in almost 60% of the
patients at the age of 3  years (Muratore et  al.
2001). Body mass index (BMI) in two-thirds of
60 adult Finnish CDH survivors has been found
to be slightly below to that of general population,
and in 6% of the patients, BMI was less than 18
(Koivusalo et al. 2005a).
1066 R. J. Rintala and M. P. Pakarinen

78.4.1.4 Skeletal Abnormalities tion between low, medium and high educational
The most common structural anomalies are chest levels were similar in CDH patients and control
asymmetry and pectus deformities, followed by subjects, but a significantly lower percentage of
vertebral anomalies, such as kyphosis and scolio- CDH survivors (9%) reached higher academic
sis. These deformities may originate from abnor- levels (vocational college or university degree)
mal fetal development or may be associated with compared with controls (21%) (Koivusalo et al.
surgical management. Among 60 Finnish adult 2005a).
CDH survivors (median age 30 years), 49% had
anterior chest asymmetry, 18% had funnel chest, 78.4.1.6 Health-Related Quality
and 27% had scoliosis. Chest asymmetry and of Life
scoliosis were more common among patients Although the reports on the effect of CDH on
with ventilatory impairment than among those long-term health, growth, neurologic develop-
with normal spirometric findings. Most of the ment and cardiovascular performance seem con-
deformations were mild and rarely required sur- tradictory, it is likely that, especially in the most
gical treatment. The skeletal deformations were severely afflicted patients, CDH has a deleterious
more common in patients who had large defects effect on quality of life. The quality of life of
and a patch repair (Vanamo et al. 1996c). Periodic Finnish adult CDH survivors (median age
and regular follow-up is suggested to detect and 39 years) and healthy controls was assessed with
prevent development of functionally significant a SF-36 questionnaire. There were no differences
deformities. in different SF -36 domains or in overall scores
between CDH patients and controls. However,
78.4.1.5 Neurological Deficits the percentage of CDH patients with low SF -36
Patients with CDH are vulnerable to abnormal overall score indicating low quality of life was
brain development and maturation and conse- 1.5 times higher than expected. Medical condi-
quently to impairment or delay in their neurocog- tions that were related to low quality of life in
nitive development. The most common CDH patients were GERD, recurrent intestinal
neurological deficits include hypotonia, hearing obstruction, recurrent abdominal pain, musculo-
loss and impaired visual-motor integration, oral-­ skeletal disorders and exercise-related shortness
motor programming, behavioral attention and the of breath (Koivusalo et al. 2005a).
lack of concentration. Neuromuscular hypotonic- Although the impact of long-term symptoms
ity is common in CDH survivors. The original on the quality of life of CDH patients who mainly
severity of CDH appears to be predictive of have had a low-risk primary condition seems
adverse neurodevelopmental outcome. A wide- subtle, future studies may disclose still more del-
spread neurodevelopmental delay was found in eterious effects on quality of life in the growing
3% of infants and in 6% of preschool-aged chil- population of survivors with high-risk CDH.
dren (Danzer et al. 2013). However, the neurode-
velopmental functioning of the majority of CDH
children is in the average range of normal chil- 78.4.2 Long-Term Outcome
dren at early preschool and preschool age. in Children with Esophageal
There is not much data regarding neurodevel- Atresia
opmental afflictions in adult CDH survivors. A
study of Finnish adult CDH survivors—most of Since the first successful repair in 1941, the mor-
them had not had a high-risk disease—used a test tality of esophageal atresia (EA) has significantly
of psychosocial functioning comprising of decreased from 60% in the 1960s (Louhimo and
Rosenberg Self-Esteem Scale, Beck’s Depression Lindahl 1983) to 2% during 1991–2012
Index and Cohen Test for Life Management (Koivusalo et al. 2013). Consequently, a signifi-
Ability, and similar results were reported in CDH cant number of EA survivors have reached adult-
survivors and in healthy controls. The distribu- hood. The most important issues in the long-term
78  Long-Term Outcomes in Pediatric Surgery 1067

outcome are associated with esophageal morbid-
ity including gastroesophageal reflux (GER),
dysphagia and pre-neoplastic mucosal changes of
the esophagus and the risk of esophageal cancer.
Other major issues of long-term morbidity are
respiratory symptoms and morbidity caused by
various associated anomalies and syndromes.
Long-term endoscopic follow-up of adult EA
survivors is recommended for patients with
defined risk factors. Our proposed institutional
recommendation is outlined in Table 78.1.
78.4.2.1 Esophageal Morbidity
Gastroesophageal reflux in EA is thought to arise
from a short esophagus and impaired motor func-
tion causing poor luminal acid clearance.
Gastroesophageal reflux may cause significant
problems immediately after EA repair and at a
later age. Negative tests for GER in infancy do
not rule out significant GER later in childhood
(Koivusalo et  al. 2004), and by the age of
10  years, significant GER is detected in up to
50% of the EA patients. Over half of these
patients require anti-reflux surgery (Koivusalo
et al. 2007). In a recent Finnish study of adult EA
survivors (median age 36 years), the prevalences
of GER symptoms and dysphagia were 34% and
85%, respectively. In 80% of the patients, dys-
phagia required an adaptive eating behavior, such
as slow eating, abundant use of drinks and careful
chewing. Endoscopy disclosed macroscopic
Barrett’s esophagus in 11%, hiatal hernia in 28%,
esophagitis in 8% and esophageal stricture in 8%
of the patients. Mucosal histology of endoscopic
biopsies showed esophagitis in 25% of the
patients. Twenty-one percent of the patients had
columnar epithelial metaplasia and 6% intestinal
metaplasia. Esophageal manometry showed non-­
propagating peristalsis and low ineffective distal
wave amplitudes in 80% of the patients. The
occurrence of columnar metaplasia was strongly
related with increased age and ineffective esoph-
ageal peristalsis. Other factors strongly associ-
ated with the occurrence of columnar metaplasia
were long-gap atresia, anastomotic stricture, and
recurrent tracheoesophageal fistula. Columnar
metaplasia and esophagitis were not associated
with the severity of GER symptoms or dysphagia
(Sistonen et al. 2010a).
Barrett’s esophagus with intestinal columnar
metaplasia is a known risk factor for mucosal
dysplasia and adenocarcinoma. Among adult EA
survivors, the estimated occurrence of Barrett’s
esophagus is four times higher than in the normal
population (Sistonen et  al. 2010a). Among EA
survivors, there are a number of published cases
of esophageal carcinoma, but the true risk of
esophageal cancer after EA repair is not known.

Table 78.1  Proposed surveillance protocol in esophageal atresia

Findings at surveillance No Any of following: Barrett Barrett
upper endoscopy at 15 years findings 1. Erosive esophagitis No With dysplasia
of age 2. Gastric (columnar) dysplasia
metaplasia
3. Esophageal stricture
4. Tracheoesophageal
refistula
5. Severe GER
symptoms
6. Continuous GER
medication
Next surveillance upper 30 years Repeat after 5 years Repeat after Confirm dysplasia grade and
endoscopy of age 1 year consider local or operative
40 years ablative treatment
of age
50 years
of age
60 years
of age
1068
A 50–500-fold risk estimates have been proposed
(Sistonen et  al. 2008; Jayasekera et  al. 2012).
Although the effect of endoscopic follow-up on
timely detection and treatment of malignant
lesions is not known, an endoscopic follow-up
program of EA survivors, especially those with
esophagitis or Barrett’s esophagus, does not seem
unreasonable.
A substantial proportion of patients, mainly
those with long-gap EA, undergo esophageal
reconstruction. The techniques most often used
are gastric pull-up, gastric tube, colonic interpo-
sition and jejunal interposition. Long-term results
of the different reconstruction methods are vari-
able. After gastric pull-up and colonic interposi-
tion, good functional results but also significant
long-term morbidity, such as chronic acid reflux,
aspiration pneumonitis, ulceration, deficiency of
iron and vitamin B12 and compression of the
respiratory tract by the dilated intrathoracic
stomach or a redundant colon, have been reported,
whereas long-term data is unavailable of jejunal
interposition (Gallo et al. 2012). Barrett’s esoph-
agus of the remaining native esophagus and
­dysplasia of the interposed colonic segment have
also been described (Lindahl et al. 1990, 1992).
Because of many potential long-term problems,
life-long endoscopic surveillance after esopha-
geal reconstruction is recommended.
78.4.2.2 Respiratory Morbidity
After repair of EA, approximately one-third of
the patients suffer from respiratory symptoms
such as wheezing, asthma, pneumonia and dys-
pnea at adolescence and adulthood (Malmström
et  al. 2008; Sistonen et  al. 2010b). Respiratory
symptoms are thought to be caused by the abnor-
mal structural development of the airways,
GERD and chronic aspiration. A comparison of
respiratory morbidity in 101 Finnish adult EA
survivors (median age 36  years) with healthy
controls showed that EA survivors had lower
respiratory quality of life and more respiratory
symptoms including infections, asthma and aller-
gies. In addition, spirometry, histamine challenge
test and exhaled nitric oxide tests disclosed high
occurrence of ventilatory defects and bronchial
R. J. Rintala and M. P. Pakarinen
hyperresponsiveness among EA survivors
(Sistonen et al. 2010b).
78.4.2.3 Musculoskeletal Morbidity
EA is typically associated with various midline
defects and with anomalies of the vertebrae, the
ribs and the upper limb. Scoliosis may develop
to up to 56% of the patients as a consequence of
vertebral anomalies and thoracotomy-induced
rib fusions. The risk of scoliosis is 13-fold com-
pared with the general population. Scoliosis is,
however, mostly mild and corrective surgery is
seldom needed. Anomalies of the cervical spine
are the most common vertebral anomalies
occurring in 38% of the patients. These anoma-
lies are often overlooked in infancy, and among
adult EA survivors, remarkably few (11%) were
diagnosed during childhood. Symptoms of ver-
tebral anomalies and scoliosis are mostly mild.
Radial ray anomalies are found in 25% of the
patients, commonly thenar aplasias or hypopla-
sias. Again, many of these defects were not
detected during primary treatment period
(Sistonen et al. 2009).
78.4.2.4 Other Issues of Long-Term
Morbidity
Apart from tracheomalacia, upper airway malfor-
mations may occur in 13% of EA patients. Most
of the airway malformations, such as choanal
atresia, tracheoesophageal clefts, cleft palate or
subglottic stenosis, are diagnosed and treated in
early infancy and are likely to cause little long-­
term morbidity. After successful EA repair or
reconstruction, many patients continue to have
insufficient oral intake of food. Patients with
impaired oropharyngeal function, such as those
with CHARGE syndrome and patients with men-
tal retardation, are at risk of permanent depen-
dency of feeding ostomy (Koivusalo et al. 2013).
Approximately 10% of EA patients have associ-
ated malformations of the gastrointestinal tract,
such as duodenal atresia, hypertrophic pyloric
stenosis and anorectal malformations. Although
these conditions are diagnosed and treated in
infancy, anorectal malformations in particular
may have a constant impact on quality of life.
78  Long-Term Outcomes in Pediatric Surgery 1069

78.4.2.5 Quality of Life (Gutierrez et  al. 2011; Goulet and Ruemmele
Despite the high incidence of GER symptoms, 2006). In clinical work, intestinal sufficiency is
dysphagia and respiratory symptoms, the general measured by the proportion of PN required for
quality of life of EA patients is similar to that of adequate growth (D’Antiga and Goulet 2013).
the general population. The educational and By definition, IF results in prolonged depen-
occupational status of EA patients are also simi- dence on PN, regardless of the underlying intes-
lar to the general population. When gastrointesti- tinal disease, which greatly modulates the
nal and respiratory symptoms–related quality of natural history on IF. PN may be partial or total
life is measured, however, EA patients have and temporary or permanent. Most neonates
impaired quality of life when compared with the with IF can be weaned from PN, whereas others
general population (Koivusalo et  al. 2005b; require partial or total PN for several years or
Dellenmark-Blom et al. 2015). even permanently. Some patients develop com-
plications especially during long-­ term PN
including liver failure and catheter-­related sep-
78.4.3 Intestinal Failure sis. These patients may be considered as candi-
dates for intestinal transplantation.
Neonatal intestinal failure (IF) refers to any The most common cause of neonatal IF is
intrinsic intestinal disease leading to an inability short bowel syndrome (SBS), characterized by
to sustain adequate nutrition and normal growth. the reduction of functional gut below the minimal
IF carries a significant morbidity and mortality as amount necessary for the adequate digestive and
well as economic burden. The most important absorptive capacity often associated with accel-
causes of IF include short bowel syndrome erated intestinal transit. SBS may result from
(SBS), intestinal motility disorders and congeni- resection, disease-related intestinal loss or con-
tal mucosal enteropathies. The management of genital short bowel. The primary etiologies of
these complex patients is challenging and neonatal SBS include necrotizing enterocolitis,
requires a multidisciplinary approach for optimal intestinal atresia, gastroschisis and malrotation
outcomes. Recent refinements in surgical and with mid-gut volvulus. Severe intestinal motility
medical treatment have been associated with sig- disorders result in recurrent or chronic intestinal
nificant reductions in morbidity and improve- obstruction in the absence of mechanical occlu-
ments in survival. The great majority of neonates sion due to impaired motor activity and peristal-
with IF can be weaned from parenteral nutrition sis. Of these, total colonic aganglionosis (TCA;
(PN), while intestinal transplantation serves as a Hirschsprung’s disease [HD]) with extended
salvage therapy for those who develop nutritional jejuno-ileal involvement and chronic intestinal
failure due to life-threatening complications, pseudo-obstruction are relatively common causes
such as liver failure or recurrent catheter-­ of IF (Salvia et al. 2008; Kurvinen et al. 2012).
associated bloodstream infections. Chronic intestinal pseudo-obstruction is a very
heterogenous condition including neuropathic
78.4.3.1 Definition and Causes and myopathic forms with or without urinary
of Intestinal Failure tract involvement. Megacystis microcolon intes-
Neonatal IF may be defined as any intestinal tinal hypoperistalsis syndrome represents a rare
disease resulting in the inability to absorb and severe form of dysmotile intestinal obstruc-
enough energy and fluids to sustain normal tion in the newborn, resulting in IF in the vast
growth and development. The definition may majority of cases. Mucosal enteropathies, which
also be based on the absolute or age-adjusted often result in IF, include congenital diseases of
length of the remaining bowel, the extent and enterocyte development, such as microvillus
duration of PN requirement, factual intestinal atrophy and intestinal epithelial dysplasia (Salvia
absorptive capacity or fecal energy loss et al. 2008).
1070
78.4.3.2 Incidence and Mortality
A nationwide Italian study reported a 0.1%
occurrence rate of IF out of 30,353 newborns
(Wales et al. 2004). The incidence of neonatal IF
was 0.05% in a prospective nationwide study
covering all of Finland, with 60,430 newborns
corresponding to 46 per 100,000 live births
(Kurvinen et  al. 2012). IF was defined as the
duration of PN for longer than 4 weeks in both of
these studies (Salvia et al. 2008; Kurvinen et al.
2012). A population-based survey from Canada
estimated the incidence of surgically treated SBS
to be 25 per 100,000 live births (Wales et  al.
2004). The occurrence of SBS increases along
with decreasing birth weight and prematurity.
In recent years, mortality rates have markedly
declined and survival rates of neonatal intestinal
failure currently reach 90% (Salvia et  al. 2008;
Kurvinen et al. 2012). The establishment of spe-
cial multidisciplinary teams for neonates with IF
appears to be the major contributor to the
improved survival rates (Gutierrez et  al. 2011).
During the first 2 years of life, the overall and the
disease-specific mortality of neonatal SBS is
three and five times higher in relation to control
cohort with comparable underlying disease char-
acteristics (Wales et  al. 2004). IF-associated
­mortality has a bimodal distribution (Gutierrez
et al. 2011). Most of the deaths occur during the
first years of life due to an underlying disease or
surgery-­associated complications. Delayed com-
plications of IF, such as intestinal failure–associ-
ated liver disease (IFALD) and sepsis due to
catheter-associated bloodstream infections, con-
tinue to cause some mortality among long-term
survivors requiring PN.  Long-term survival is
markedly higher for IF patients weaned from PN
as opposed to those not weaned.
78.4.3.3 Predictors of Intestinal
Autonomy
The amount, anatomy and functional state of the
remaining intestine and the etiology of IF essen-
tially influence the possibility of achieving intes-
tinal autonomy, which is the most important
predictor of long-term survival. The chances for
weaning off PN are markedly worse in children
with severe intestinal dysmotility or congenital
R. J. Rintala and M. P. Pakarinen
enteropathies compared to those with SBS as an
underlying etiology for IF (Pakarinen et al. 2009).
In SBS patients, the remaining bowel is usually
functionally intact, enabling adaptation, whereas
primarily diseased intestine in motility disorders
and enteropathies rarely improves with time and
is highly resistant to various modes of medical
and surgical treatment. Among SBS patients, the
function of the remaining small intestine can
often be improved by autologous surgical proce-
dures designed to tailor and lengthen dilated
small bowel remnant (Pakarinen et  al. 2013a).
These procedures are much less frequently indi-
cated in intestinal motility disorders, which may
necessitate intestinal transplantation more fre-
quently (Pakarinen et al. 2013b).
Among neonates with SBS, the length of the
remaining small intestine is strongly correlated
with duration of PN, being the most important
single predictor of PN dependence. The presence
of the ileocecal valve (ICV) and residual colon or
ileum is associated with a shorter duration of PN
and improved chances of weaning off PN.  The
vast majority of neonates with more than
35–40  cm of small bowel remaining achieve
intestinal autonomy. In a study of 135 neonates
with SBS, 40% of those with less than 40  cm
residual small bowel and without ICV remained
dependent on PN after 8 years, whereas 80% of
those with 40–80 cm residual small bowel and an
intact ICV were weaned off PN within 1  year
(Goulet et  al. 2004). However, a wide variance
exists and some neonates wean off PN with as
little as 10 cm of small intestine below the liga-
ment of Treitz. Some of this variance may be
explained by the rapid in utero growth of small
intestinal length, doubling during the last trimes-
ter of gestation. Accordingly, instead of the abso-
lute length, the percentage of normal age-adjusted
bowel length may be a more accurate determi-
nant of weaning from PN. Functional destruction
of the remaining intestine may prolong PN,
despite favorable anatomy. For example, motility
of the remaining bowel may be impaired due to
ischemic injury and scarring in necrotizing
enterocolitis.
A retained ICV is beneficial by restraining
colonization of the small intestinal lumen by
78  Long-Term Outcomes in Pediatric Surgery
colonic bacteria and bacterial overgrowth and by
slowing intestinal transit. Because the presence
of ICV is almost invariably associated with partly
retained ileum in the clinical setting, some of the
beneficial functional effects of ICV may be, in
fact, mediated by the residual ileum. Even a short
segment of preserved ileum is advantageous in
several ways. The active absorption of conju-
gated bile acids and vitamin B12 is confined to the
distal ileum. The interruption of the enterohe-
patic circulation of bile acids results in impaired
micelle formation and malabsorption of fat and
fat-soluble vitamins. In children with SBS, sig-
nificant malabsorption of bile acids continues
after weaning off PN, which is markedly more
severe among those without any remaining ileum
(Pakarinen et al. 2010).
78.4.3.4 Growth and Development
Neonatal IF is associated with short stature both
during childhood and in adulthood (Mutanen
et  al. 2013a; Pichler et  al. 2013; Oliemen et  al.
2012), although some catch-up growth may occur
during puberty (Mutanen et al. 2013a). Children
with congenital enteropathy are at the greatest
risk of growth failure compared to those with
SBS or intestinal motility disorders as a cause of
IF (Pichler et  al. 2013). Overall, up to 50% of
children with IF are short with a standard devia-
tion score for height below −2.0 (Mutanen et al.
2013a; Oliemen et  al. 2012). By adulthood, the
majority of them remain below the target height,
while their weight and body composition are less
affected (Oliemen et  al. 2012). Most children
undergo normal pubertal development (Mutanen
et al. 2013a; Oliemen et al. 2012).
The risk of metabolic bone disease is clearly
increased both during and after weaning off PN
(Mutanen et  al. 2013a; Pichler et  al. 2013).
Approximately one-third of children and adoles-
cents show decreased lumbar spine bone mineral
density with standard deviation score below
−2.0, regardless of the underlying etiology of
IF. Bone mineral density seems to fall with time
at least among those who remain dependent on
PN support (Pichler et  al. 2013). Despite
decreased bone mineral density, the incidence of
fractures is not increased, possibly reflecting
1071
decreased physical activity (Mutanen et  al.
2013a; Pichler et al. 2013). A significant propor-
tion of patients have vitamin D deficiency and
associated hyperparathyroidism despite active
vitamin D supplementation (Mutanen et  al.
2013a). The shorter the remaining small bowel,
the lower the serum concentration of vitamin D
(Mutanen et  al. 2013a). These findings under-
score the importance of close ongoing nutritional
monitoring among these children.
78.4.3.5 Intestinal Failure–Associated
Liver Disease (IFALD)
In addition to the short remaining small intestine
and prolonged dependence on PN, cholestasis is
the major predictor of mortality in pediatric
IF. Depending on the definition, IFALD occurs in
up to 60–80% of surgical neonates, who require
prolonged PN (Salvia et al. 2008; Kurvinen et al.
2012). Biochemical signs include conjugated
hyperbilirubinemia associated with increased
gamma-glutamyl transferase and aminotransfer-
ases. IFALD involves hepatic steatosis, periportal
inflammation and cholestasis, which may prog-
ress to bile duct proliferation, fibrosis and biliary
cirrhosis leading to liver failure. Portal hyperten-
sion and associated esophageal varices with sple-
nomegaly and hypersplenism, as well as a
decrease in liver-derived coagulation factors as a
sign of impaired hepatic synthetic capacity, are
usually observed late in the disease course.
Around 10% of children with IF develop choleli-
thiasis at some point (Mutanen et al. 2013b). The
cause of IFALD is multifactorial and still incom-
pletely understood.
Although jaundice is usually the first clinical
sign of IFALD, biochemical cholestasis is not
well correlated with the presence or degree of
histological liver fibrosis, which is a common
finding after long-term PN, despite well-­
preserved biochemical liver function (Kurvinen
et  al. 2011). Moreover, significant liver fibrosis
and steatosis persist in about half of the patients
after weaning off PN (Mutanen et  al. 2013b).
Although laboratory markers of liver function
usually normalize after weaning off PN, liver his-
tology remains abnormal up to 9  years after
weaning off PN in the majority of patients
1072
(Mutanen et  al. 2013b). At present, the clinical
significance of persistent histological changes of
the liver remains unclear but clearly points to an
ongoing need to also follow up these children
after weaning off PN.
78.4.3.6 Quality of Life
Children and adolescents with a history of neona-
tal IF have lower overall health-related quality of
life in relation to a healthy control population
(Olieman et  al. 2012; Sanchez et  al. 2013).
Parent-proxy scores are similarly decreased,
although traditional quality of life tools fail to
assess important impacts of IF on children and
their families (Olieman et al. 2012). On the sub-
scale level, children’s psychosocial health, physi-
cal functioning and school functioning are
significantly impaired (Olieman et  al. 2012).
Patient age and socioeconomic status modify
results so that the worst health-related quality of
life is associated with increasing age and low
socioeconomic status. A decreased health-related
quality of life is not an unexpected finding among
those receiving PN due to significant limitations
to lifestyle and a continuous threat of severe com-
plications. Frequent gastrointestinal symptoms
including diarrhea and loose stools, as well as
different kinds of abdominal complaints, such as
bloating, bowel cramps and flatulence, may also
contribute to a decreased quality of life after
weaning off PN (Oliemen et al. 2012).
78.4.4 Hirschsprung’s Disease
Hirschsprung’s disease (HD) is a rare (1:5000)
congenital disease characterized by an absence of
enteric neurons and ganglion cells of the distal
bowel. Failure of enteric neural crest–derived
cells to colonize the distal bowel results in func-
tional obstruction at the affected region and sec-
ondary bowel dilatation proximal to the
aganglionic segment. Clinical manifestations of
HD include failure or delayed passage of meco-
nium, severe constipation, intestinal obstruction
and failure to thrive and may be even fatal due to
HD-associated enterocolitis. Mutations in more
R. J. Rintala and M. P. Pakarinen
than ten genes have been associated with HD, but
the majority of identified mutations are in RET
explaining around 40% of familial and 20% of
sporadic cases. Between 75% and 80% of all
patients have rectosigmoid HD in which the
aganglionosis is confined to the rectosigmoid
colon. A more extensive colonic or total colonic
aganglionosis (TCA) with a variable ileal
involvement both occur in about 10% of the
patients. Aganglionosis extending to the proxi-
mal small bowel is exceedingly rare. Overall, the
long-term functional outcome expectancy of HD
in terms of fecal and urinary continence is rela-
tively optimistic today. The outcomes have
improved significantly due to an increasing
understanding of the pathological anatomy and
physiology of the defect and modern surgical
techniques. Although some degree of bowel dys-
function and defects in fecal continence remain
permanent in many patients, the great majority of
patients reaching adolescence and adulthood are
able to maintain themselves socially continent
and live normal lives (Jarvi et al. 2010; Heikkinen
et al. 1995). In some, however, this may require
special measures such as bowel management pro-
grams. Psychosocial implications of HD in adult-
hood including quality of life issues have recently
gained significant attention (Hartman et al. 2004).
78.4.4.1 Bowel Function: Classic
Rectosigmoid Aganglionosis
Outcomes of HD regarding bowel function in
childhood and adolescence are well characterized
(Bai et al. 2002; Baillie et al. 1999; Rintala and
Pakarinen 2006; Conway et  al. 2007). During
childhood, different degrees of fecal incontinence
and constipation are major functional derange-
ments, whereas enterocolitis is an exceptional
problem in the long term (Jarvi et  al. 2012).
Controlled follow-up studies that assess long-­
term outcomes of bowel function among HD
patients are still few (Jarvi et al. 2010; Heikkinen
et  al. 1995; Conway et  al. 2007). Comparison
with healthy controls carefully matched for
demographic variables, such as age and gender, is
essential in order to obtain reliable results (Jarvi
et al. 2010, 2012). Very few investigations have
78  Long-Term Outcomes in Pediatric Surgery
solely concentrated on adults with operated HD
(Jarvi et al. 2010; Heikkinen et al. 1995; Conway
et al. 2007; Ieiri et al. 2010). These studies have
mainly included patients who have been operated
using the Duhamel procedure for rectosigmoid
aganglionosis. Consequently, our current knowl-
edge on true long-term outcomes of bowel func-
tion in adulthood relies largely on the results of
the Duhamel procedure among patients with rec-
tosigmoid disease. It is noteworthy that no long-­
term functional outcome studies after endorectal
pull-through, not to mention transanal procedure,
are yet available.
Available controlled follow-up studies extend-
ing to adolescence uniformly demonstrate
impaired bowel function of variable degree,
including an increased incidence of fecal inconti-
nence and constipation in relation to age-matched
controls (Jarvi et al. 2010; Heikkinen et al. 1995;
Diseth et al. 1997; Reding et al. 1997; Bai et al.
2002; Baillie et  al. 1999; Conway et  al. 2007).
Table 78.2 displays published controlled studies
on bowel function in adolescence after operative
treatment of HD.  These investigations suggest
that approximately half of the patients have
impaired bowel function, mainly resulting from
different degrees of fecal incontinence and con-
stipation. The results of functional assessment of
bowel function are very similar among adults
with operated HD, as shown in Table  78.3,
suggesting that no major alterations in bowel
­ et al. 2010). It is evident that bowel function in
function occur after adolescence. In both age
groups, only approximately half of the patients
achieved optimal functional outcome, defined as
bowel function score comparable to that of the
control population, in the absence of fecal incon-
tinence, constipation or stoma.
Long-term bowel function in adulthood was
assessed in a population-based, cross-­sectional,
controlled follow-up study including 92 patients,
who had been operated on for HD in childhood
(Jarvi et al. 2010). The great majority (94%) of
the patients had rectosigmoid aganglionosis and
most (78%) had undergone the Duhamel proce-
dure. The overall bowel function score assessing
ability to hold back defecation, sensation of urge
to defecate, frequency of bowel actions, fecal
1073
soiling, fecal accidents, constipation and social
problems related to bowel function was signifi-
cantly inferior to that of controls (17.1 ± 2.8 vs
19.1 ± 1.2; P < 0.0001). Patients reported signifi-
cantly more often difficulties to hold back defe-
cation, fecal soiling, constipation and social
problems related to bowel function. The percent-
ages of patients and controls who reported any
degree of respective functional impairment were
40% vs 17%, 48% vs 22%, 30% vs 9% and 29%
vs 11%. A total of 14% of the patients had frank
fecal accidents compared to 0% among controls.
The proportion of patients with fecal soiling
more frequently than once a week, fecal acci-
dents more often than once a week or constipa-
tion treated with laxatives or enemas were 13%,
2.2% and 10%, respectively. The frequency of
bowel actions was one to two times per day in
60% of the patients and similar to that of con-
trols. No significant problems regarding sensa-
tion of urge to defecate were reported by the
patients. In a multivariate analysis, only an
increasing age was found to be a significant pre-
dictor of poor functional outcome and age was
inversely associated with bowel function score.
These results were supported by another study
demonstrating bowel function comparable to
healthy controls in 47% of 49 patients after the
Duhamel procedure, with a mean age of 20 years,
using an identical bowel function scoring (Ieiri
adolescence and adulthood is clearly impaired
among adults operated on for HD, at least after
the Duhamel procedure. Derangements of bowel
function occur in approximately half of the
patients. On the other hand, socially disturbing
fecal soiling is relatively uncommon in adults
and had little impact on their psychosocial func-
tioning (Jarvi et al. 2010; Heikkinen et al. 1995;
Diseth et  al. 1997), most likely reflecting more
advanced coping methods and efficient adapta-
tion to the problem in adulthood.
78.4.4.2 Evolution of Bowel Function
with Advancing Age
Overall bowel function of patients with HD
seems to improve with time, at least up to a cer-
 1074

Table 78.2  Overview of controlled bowel function studies among adolescents with operated Hirschsprung’s disease
Patients Follow-up age Main Rectosigmoid Impaired Constipation Bowel function Optimal
Author Year studied Controls (y)a operation aganglionosis (%) continenceb (%) (%) score outcomec (%)
Diseth 1997 22 33 16 (10–20) Duhamel 64 36 16 Decreased 52
Reding 1997 27 39 12 (5–26) Swenson 100 NR NR Decreased 52
Bai 2002 45 44 11 (8–16) Swenson 100 45 7 Decreased 49
Baillie 1999 80 22 8 (5–16) Duhamel 71 NR NR Decreased 42
NR, not reported
a
Median or mean (range)
b
Includes any degree of fecal incontinence such as soiling as described by authors
c
Defined as bowel function score comparable to that of controls in the absence of fecal incontinence, constipation or stoma
R. J. Rintala and M. P. Pakarinen
Table 78.3  Summary of follow-up studies on bowel function in adults operated for Hirschsprung’s disease in childhood
Patients Follow-up age Main Rectosigmoid Impaired Constipation Bowel function Optimal
78  Long-Term Outcomes in Pediatric Surgery

Author Year studied Controls (y)a operation aganglionosis (%) continenceb (%) (%) score outcomec (%)
Jarvi 2010 92 53 43 (18–68) Duhamel 94 48 30 Decreased 46
Ieiri 2010 43 No 33 (19–55) Duhamel 79 36 33 NR 67
Heikkinen 1995 102 81 31 (15–39) Duhamel 95 45 5 Decreased 54
Conway 2007 49 20 20 ± 3.6 Duhamel 81 NR NR Decreased 47
NR, not reported
a
Median or mean (range/standard deviation)
b
Includes any degree of fecal incontinence such as soiling as described by authors
c
Defined as bowel function score comparable to that of controls in the absence of fecal incontinence, constipation or stoma
1075
1076 R. J. Rintala and M. P. Pakarinen

tain age (Diseth et al. 1997; Baillie et al. 1999; 78.4.4.4 Total Colonic
Rintala and Pakarinen 2006; Conway et al. 2007; and Panintestinal
Ieiri et al. 2010). The critical age for this improve- Aganglionosis
ment is puberty (Bai et  al. 2002). It is unclear The long-term outcomes of TCA extending to
what the mechanisms behind this age-related adolescence are clearly inferior in relation to rec-
improvement are and whether bowel function tosigmoid HD. A central, long-term complication
continues to improve after adolescence. Taking associated with TCA is recurrent enterocolitis.
into account very similar overall outcomes of The reported rates of post pull-through enteroco-
bowel function reviewed for adolescents and litis range from 2% to 27% in rectosigmoid HD
adults in Tables 78.2 and 78.3, it seems unlikely and between 20 and 55% in TCA, respectively.
that a significant improvement of overall bowel Soiling and fecal incontinence are common in
function occurs after puberty. Conversely, bowel TCA patients. At least one-third of the patients
function appears to deteriorate along with suffer from frank fecal incontinence in later
advancing adult age, as discussed above (Jarvi childhood or adolescence. This has significant
et al. 2010). psychosocial implications because the inconti-
nence in many of these patients is a continuous
78.4.4.3 Urinary and Sexual Function daily problem. Moreover, a significant percent-
Low pelvic dissection during rectosigmoidec- age of these patients require nighttime diapers
tomy is associated with a risk of neural damage (Wildhaber et  al. 2005). Because of intractable
interfering genitourinary function, with the risk incontinence or recurrent enterocolitis, some
being higher when the dissection is performed patients opt for permanent or long-term bowel
outside the rectal wall. Day- or nighttime enure- diversion. The percentage of TCA patients end-
sis has been reported in addition to urodynamic ing up with a permanent ileostomy ranges
abnormalities in occasional children with oper- between 5% and 18%.
ated HD.  Urinary and sexual dysfunction also In patients with TCA, stooling frequency sig-
occurs in a small minority of adult and adolescent nificantly decreases over time. The numerous
patients with HD (Ieiri et al. 2010). Some series bowel movements during early postoperative
report no adult or adolescent patients with uri- years often cause significant discomfort because
nary incontinence or sexual dysfunction of defective fecal continence and perianal rash.
(Heikkinen et  al. 1995; Diseth et  al. 1997), but Typically, stooling frequency decreases from 5 to
controlled long-term follow-up studies specifi- 6 bowel movements per day at 5 years of age to
cally designed to investigate urinary and sexual around 3–4 per day at adolescence. Growth retar-
function following surgical treatment of HD are dation during childhood is common, extending to
lacking. However, in their detailed analysis of adolescence and affecting mainly the body
178 patients with a mean age of 10 years, Moore weight (Escobar et  al. 2005). In recent reports,
et  al. found that a significant percentage of long-term severe failure to thrive is, however,
patients suffer from micturition disturbances and uncommon.
sexual dysfunction. The respective figures were Near total or panintestinal aganglionosis rep-
9.8% and 11% (Moore et al. 1996). Sexual dys- resents the most extensive and a rare form of HD,
function included dyspareunia, erectile dysfunc- with extension of the aganglionosis to the proxi-
tion and even infertility. The majority of the mal small intestine (Ruttenstock and Puri 2009).
patients with micturition disturbances suffered This condition carries extremely high mortality
from different degrees of urinary incontinence. and uniform permanent dependence on paren-
As expected, urinary dysfunction was more com- teral nutrition without intestinal transplantation.
mon in patients following Swenson or Duhamel Initially, these patients can be salvaged by myec-
operation than in those with endorectal pull-­ tomy and myotomy of the retained proximal jeju-
through ((Moore et al. 1996). num. In addition to providing perquisite for stable
78  Long-Term Outcomes in Pediatric Surgery
parenteral nutrition without massive stomal fluid
losses, the operation may facilitate significant
enteral tolerance preventing development of liver
disease (Pakarinen et al. 2009). The mainstay of
treatment is parenteral nutritional support, and
every effort should be made to avoid life-­
threatening complications of long-term paren-
teral nutrition, such as intestinal failure–associated
liver disease, frequent septicemia and loss of
venous access sites when intestinal transplanta-
tion serves as a salvage therapy.
78.4.4.5 Syndromic Hirschsprung’s
Disease
Down’s syndrome (trisomy 21) is the most com-
mon syndrome associated with HD.  The fre-
quency of Down’s syndrome in patients with HD
ranges from 2% to 15%. Mortality of Down’s
syndrome combined with HD is increased due to
frequently occurring comorbidities. Many of the
fatalities are attributed to cardiac problems such
as atrioventricular septal defects that are common
in Down’s syndrome.
Down’s syndrome patients are more suscepti-
ble to chronic enterocolitis than non-syndromic
HD patients, while chronic enterocolitis may lead
to the development of inflammatory bowel dis-
ease–like colonic inflammation. Functional prog-
nosis of HD patients with Down’s syndrome is
inferior to otherwise healthy patients. They
develop bowel control much slower than HD
patients without chromosomal aberrations. A sig-
nificant proportion of them, and especially those
with chronic enterocolitis, suffer from fecal
incontinence as adolescents or adults.
HD is also associated with other more uncom-
mon syndromes and neurocristopathies. Patients
with Mowat–Wilson syndrome, resulting from
mutations in the SIP 1 gene, have typical dysmor-
phic features, severe intellectual disability,
microcephaly and seizures and commonly
HD. These patients often display frequent bouts
of severe enterocolitis and develop bowel control
very slowly. Cartilage-hair hypoplasia is a
metaphyseal chondrodysplasia associated with
growth failure, impaired immunity and high inci-
dence of HD.  It is caused by a mutation in the
RMRP gene. Cartilage-hair hypoplasia patients
1077
with HD have worse prognosis than other HD
patients (Mäkitie et al. 2002). Due to an immune
defect, they have extremely high incidence of
pre- and postoperative enterocolitis that can even
be fatal.
78.4.4.6 Hirschsprung’s Disease
and Cancer
The phenotype of HD, multiple endocrine neo-
plasia type 2 (MEN2A) and familial medullary
thyroid cancer are all associated with germline
mutations of the RET proto-oncogene.
Approximately 2.5% to 5% of HD patients carry
MEN2A RET mutations with an increased risk of
developing medullary thyroid carcinoma. The
incidence of medullary thyroid carcinoma and
other cancers was assessed in a population-based
cohort of 156 adult patients treated for HD
(Table 78.4) (Pakarinen et al. 2005). Among 156
patients and 4816 person years at risk, two cases
of medullary thyroid cancer were observed giv-
ing the standardized incidence ratio of 550 (95%
CI 67–2000) (Pakarinen et al. 2005). The risk of
cancers other than medullary thyroid carcinoma
was not increased. Of note, only one patient had
been treated for rectal carcinoma.
To further elucidate the risk of thyroid malig-
nancies in adults with HD, a total of 91 of the 156
patients agreed to participate in the follow-up
part of the study; they underwent ultrasound
examination of the neck with a thyroid needle
biopsy when indicated, determination of serum
calcitonin and extraction of DNA from whole
blood samples. Exons 10, 11, 13 and 16 of RET
were sequenced in all, and all exons of RET in 43
individuals, including those with thyroid cancer,
Table 78.4 Cancer cases among 156 adults with
Hirschsprung’s disease
Malignancy N
Medullary thyroid cancer 2
Papillary thyroid cancer 1
Adenocarcinoma of the rectum 1
Adenocarcinoma of the prostate 1
Non-Hodgkin’s lymphoma 1
Acute lymphocytic leukemia 1
Glioma of the brain 1
N, number of cases
1078
RET mutations, suspicious clinical findings and
familial and long-segment disease. These investi-
gations revealed one additional patient with
increased serum calcitonin as a sign of medullary
thyroid cancer and one papillary carcinoma of the
thyroid. Medullary thyroid cancer cases had
mutations in susceptibility codons (C611R and
C620R), while no clinical evidence of thyroid
cancer was observed in four patients with suscep-
tibility mutations in codons 609 and 791. These
findings suggested that medullary thyroid can-
cer–associated RET mutations were restricted to
exons 10 and 13, affecting approximately 5% of
unselected adults with HD.
78.4.4.7 Quality of Life
Some earlier reports describing outcome in
adult or adolescent patients showed very little
limitations with regard to occupation, social
contacts or physical activities (Heikkinen et al.
1995; Diseth et al. 1997). More recent data sug-
gest that children and adolescents with HD
report comparable overall quality of life in rela-
tion to healthy reference subjects, although
there are individuals with very low level of qual-
ity of life (Jarvi et al. 2010). Low level of self-
competence seemed to be associated with
female gender, advancing age and a severe form
of the disease (Hartman et al. 2008). Adults with
HD show decreased overall physical quality of
life when compared to the general population
(Hartman et  al. 2004). However, in the same
study, all the specific domains of quality of life
among adults with HD were comparable to
those of the general population. Interestingly,
psychosocial functioning appears to have a
stronger effect on quality of life than disease-
specific factors, such as constipation and fecal
incontinence. Adults with operated HD demon-
strate only marginally lower gastrointestinal
quality of life in relation to randomly chosen
matched controls mainly due to lower scores in
questions assessing functional bowel symptoms
(Jarvi et al. 2010). Furthermore, social problems
associated with impaired bowel function are
more prevalent among adult patients than con-
trols. A worrying finding is that the degree of
impairment of bowel function increases with
R. J. Rintala and M. P. Pakarinen
increasing adult age, which may have a signifi-
cant negative impact on quality of life of older
patients with HD.
78.4.5 Anorectal Malformations
The spectrum of anorectal malformations ranges
from minor aberrations in the position of the anal
opening with very little long-term functional
consequences to very severe malformation com-
plexes, such as congenital cloacal malformations,
which have a serious impact on the patient’s life-
time bowel and urogenital function as well as
quality of life.
In the literature, there is a great variation in the
functional results after repair of anorectal malfor-
mations. There is generally no agreed method to
assess the bowel function of patients with anorec-
tal malformations, and the main problem in com-
paring different series is the highly variable
criteria used in the evaluation of fecal continence.
Therefore, the outcomes can be expressed only in
very general terms. The final outcome may not be
assessed until the patients have reached adult-
hood and, as independent individuals, can evalu-
ate the social consequences of possibly defective
bowel control.
Scoring methods based on subjective parame-
ters with or without clinical examination have
been designed to get quantitative information
about the bowel function. The most commonly
used classifications are the Kelly-score (Stephens
and Smith 1971), the Templeton score (Templeton
and Ditesheim 1985), the Holschneider score
(Holschneider 1983; Rintala et al. 1992) and the
Rintala score (Rintala and Lindahl 1995). All
these classifications ultimately categorize the
outcomes as good, fair or poor. Descriptive, non-­
scoring methods to assess and describe functional
outcome were advocated by the Wingspread
group of pediatric surgeons (Stephens and Smith
1986). This scoring method and the descriptive,
non-scoring outcome classification of Alberto
Pena, which is based on his extensive series
(Pena 1995), are the basis of the new Krickenbeck
outcome classification method (Holschneider
et al. 2005).
78  Long-Term Outcomes in Pediatric Surgery 1079

There are still very few reports on outcomes of However, larger series of Ong et al. (Ong and
ARM in adults. Most adult series include patients Beasley 1990) and Rintala et  al. (1997) have
that have been operated on by traditional meth- demonstrated that a significant percentage of
ods, such as abdominoperineal or sacro-­ patients have abnormal anorectal function at the
abdominoperineal and sacroperineal adult age. Ong et al. reported 35 patients with a
pull-throughs. There are few reports on adult out- follow-up of more than 15 years. Although the
comes following posterior sagittal anorectoplasty majority were considered to have good conti-
(PSARP), the gold standard of ARM repair today. nence according to commonly used clinical
Laparoscopy-assisted repairs of anorectal mal- scoring methods, only 13 (37%) of the patients
formations have gained significant popularity in were clean at all times. Seventeen patients
recent years. The main indication for laparo- (49%) had fecal smearing and 5 (14%) had
scopic repairs has been recto-bladder neck and intermittent soiling (Ong and Beasley 1990).
high prostatic fistulas, but some surgeons have Rintala et al. used a quantitative scoring method
used laparoscopic techniques for milder anoma- to compare the bowel function of 83 patients
lies, such as bulbar or vestibular fistulas (Rentea with that of healthy individuals with similar age
et al. 2020). and sex distribution. All controls had good fecal
continence, 76% with completely normal bowel
78.4.5.1 Mild Anomalies function. In contrast, 60% of the patients with a
with Perineal Fistula: Long- low ARM had good continence, and completely
Term Bowel Function normal bowel function was observed only in
Traditionally, the long-term results in mild mal- 15%. Social problems related to deficient fecal
formations are considered to be good in the great control were reported by 39% of the patients.
majority of patients (Kiesewetter and Chang Other health problems were reported by 52% of
1977; Trusler and Wilkinson 1962; Nixon and the patients but only by 6% of the controls. In a
Puri 1977). More critical reviews on the long-­ more recent report by Kyrklund et al. males and
term outcome have clearly demonstrated a sig- females with mild anomalies, treated using
nificant number of children with functional comprehensive methods, had a bowel function
defects, the most common of which is chronic that was comparable with that of healthy con-
constipation followed by soiling (Yeung and trols after a mean follow-up of 12  years
Kiely 1991; Ong and Beasley 1990). These series (Kyrklund et al. 2017).
have demonstrated that approximately half of the It is evident that at adult age, a significant pro-
patients have significant aberrations in the ano- portion of patients with mild ARM still have
rectal function. problematic bowel function, such as constipation
There are only a few functional outcome stud- and fecal smearing, both of which may have sig-
ies of low ARM with a follow-up extending to nificant psychosocial consequences.
adulthood. Most earlier series report favorable
functional outcome in most patients with low 78.4.5.2 High Anomalies: Long-Term
anomalies. Karkowski (Karkowski et  al. 1973) Bowel Function
reported good continence in 12 (80%) of his 15 Reports concerning long-term results for high
patients with low malformations. Nixon et  al. anomalies are highly variable. Most series grade
(Nixon and Puri 1977) found entirely normal the results as good, fair or poor. It has to be
bowel control in 23 (74%) of his 31 adult or ado- remembered that in most reported series, a good
lescent patients. The remaining patients had outcome does not mean that the patient has a nor-
occasional or frequent soiling. Recently, Iwai mal bowel function. The patients with a good
et  al. (Hashish et  al. 2010) found good bowel result have usually been considered socially con-
function with no soiling in all seven patients tinent, which implies that the defects in bowel
(mean age 26  years) with follow-up to function do not cause significant social disability.
adulthood. In the era before the posterior sagittal anorecto-
1080
plasty, the reported percentages of patients, eval-
uated with clinical criteria, with a “good” result
varied between 6% and 56%. The percentage of
poor results, which means more or less total
incontinence, varied between 10% and 70% of
the affected patients (Stephens and Smith 1971;
Nixon and Puri 1977; Taylor et  al. 1973). The
plausible explanation for the variation is differ-
ences in the strictness of assessment criteria. The
two large series of Templeton (Templeton and
Ditesheim 1985) and Rintala (Rintala et al. 1991)
both used a quantitative multifactorial evaluation
for continence. Both of these series identified a
lower percentage of poor results than the other
series using mainly qualitative criteria. These
multifactorial quantitative assessments seemed to
grade continence higher than a qualitative clini-
cal assessment, probably because they are more
sensitive to patient’s social adaptation to abnor-
mal anorectal function (Rintala et al. 1991). Only
Taylor et  al. (1973) and Rintala (Rintala et  al.
1991) have reported the incidence of patients
with a completely normal bowel function without
soiling or staining in any circumstances; and in
both series, only 7.5% of the patients had full
bowel control compared to that of healthy chil-
dren. It is probable that this grim figure reflects
the true incidence of unequivocally good long-­
term outcome in the patients treated with abdom-
inoperineal and sacral approaches.
There are still only a few reports concerning
long-term functional outcome following poste-
rior sagittal anorectoplasty, and the results have
been contradictory. Some surgeons report a dis-
mal outcome with most patients requiring adjunc-
tive measures to maintain social continence
(Langemeijer and Molenaar 1991). On the other
hand, in Pena’s series (Pena 1995), approxi-
mately one-third of the patients with high or
intermediate anomalies could be considered
totally continent. In the series of Rintala et  al.
(Rintala and Lindahl 1995), in which the bowel
function of the patients were compared to that of
healthy children with similar age and sex distri-
bution, 35% of the patients had an age-­appropriate
normal bowel function. A fair outcome with
intermittent soiling requiring frequent change of
underwear or protective aids or poor outcome
R. J. Rintala and M. P. Pakarinen
with intractable constipation or total inconti-
nence was found in 30% of the patients. Similar
results were found in a more recent report by
Kyrklund et al. (2017).
Many authors report an improvement in fecal
continence with increasing age in patients oper-
ated by the abdominoperineal procedure with or
without a sacral approach (Templeton and
Ditesheim 1985; Kiesewetter and Chang 1977;
Rintala et al. 1991). In Rintala’s series, the inci-
dence of good outcome increased from 35% in
the age group between 5 and 10 years to 58% in
patients between 11 and 15 years. The improve-
ment in fecal continence is more clearly shown in
the series including adolescent or adult patients
(Kiesewetter and Chang 1977; Nixon and Puri
1977). It is not clear whether this improvement is
a true improvement of sphincter function or just
adaptation to their handicap. On the other hand,
reports by Pena and Rintala (Rintala and Lindahl
1995; Pena 1995) on patients operated on by pos-
terior sagittal anorectoplasty have shown that in
favorable cases, the patients gain normal or near
normal bowel function as early as 3 years of age,
provided that the inherent functional complica-
tions related to the procedure, especially consti-
pation that often causes overflow soiling, are
treated vigorously. On the other hand, perceived
worsening stooling patterns with age have
recently been reported in patients with high
anomalies treated by PSARP (Hashish et  al.
2010). These controversial results reflect the dif-
ficulties in assessing functional outcomes in
patients with anorectal malformations.
As with low anomalies, pediatric surgical lit-
erature has only a few reports concerning func-
tional outcome in high anomalies at adult age
(Nixon and Puri 1977; Rintala et  al. 1994;
Hassink et al. 1993). Again, there is a wide vari-
ation in the outcomes; this most likely reflects
the diversity of assessment criteria. Nixon et al.
(Nixon and Puri 1977) found normal bowel con-
trol in 7 (15%) of 47 adolescent and young adult
patients. Twenty-nine patients (62%) had occa-
sional soiling and 11 (23%) had frequent soiling
or a colostomy. Rintala et al. (1994) studied 33
adult patients with a mean age of 35 years using
a questionnaire-based scoring system. Healthy
78  Long-Term Outcomes in Pediatric Surgery
adults with an age and sex distribution similar to
the patients were used as controls. None of the
patients reached a score indicating normal
bowel function and only 6 (18%) had a good
continence score. All controls had good scores
and 80% had a score indicating normal bowel
function. Thirty-­one (94%) of the 33 patients
reported some degree of fecal soiling. Hassink
et al. (1993) evaluated 58 patients with a median
age of 26 years by using similar scoring meth-
ods as Rintala et al. Additionally, in this study,
none of their patients met the criteria for normal
bowel habits. About 80% of the patients had
soiling. In these series, most patients had an
abdominoperineal repair as a primary recon-
struction. In Nixon’s (Nixon and Puri 1977)
series, 68% of the patients had major secondary
surgery to improve continence. In Rintala’s
series (Rintala et al. 1994), 30% of the patients
and in Hassink’s series (Hassink et  al. 1993)
half of the patients had secondary sphincter
repairs. Although many patients report being
satisfied with their current level of fecal conti-
nence (Rintala et al. 1994), objective evaluation
of the data concerning bowel function provides
a different picture. It is likely that the adult
patients have accepted their handicap. The
patients have developed measures to cope with
unsatisfactory bowel control, such as by staying
near toilets, by wearing liners or diapers, by
having regular enemas or by having dietary
restrictions (Rintala et  al. 1994; Hassink et  al.
1993; Hassink et al. 1994).
Patients with defective continence benefit
from bowel management programs (Pena and
Hong 2000). Retrograde bowel enemas are effec-
tive but not as well tolerated as antegrade wash-
outs. Malone antegrade colonic enema (MACE)
has proven to be a powerful modality to treat
fecal incontinence of various etiologies (Malone
et  al. 1990). A great majority of patients with
ARM have gained continence following this pro-
cedure (Koivusalo et  al. 2008b; Hassett et  al.
2009; Rangel et al. 2011).
Today, the gold standard of anorectal repair in
ARM is posterior sagittal anorectoplasty by
deVries and Pena (1982). There are, however, no
literature reports concerning bowel function of
1081
exclusively adult patients who have undergone
primary repair of ARM by posterior sagittal ano-
rectoplasty. Rintala et  al. (Rintala and Lindahl
1995) found normal or good continence (no
social restrictions) in 64% of adolescents who
had undergone PSARP during early childhood.
The same group compared adolescent and young
adult patients who had undergone either posterior
sagittal anorectoplasty or sacro- or sacro-­
abdominoperineal pull-through. The continence
outcome was significantly better in those who
had undergone PSARP (Rintala and Lindahl
1999). Similar findings were recently reported by
Danielson et al. (2015).
Rintala et al. reviewed the functional outcome
in 69 patients who were older than 15 years at the
time of assessment. Normal continence was
found in 43% of the patients. Thirty-six percent
had minor problems, such as constipation or
occasional staining, causing little social restric-
tions. Fourteen patients (25%) had significant
continence problems causing frequent soiling
and used protective pads or changed underwear
or had undergone MACE procedure (Rintala and
Pakarinen 2010). More recently, Kyrklund et al.
reported that 76% of males with rectourethral fis-
tulas and 85% of females with vestibular fistulas
developed social continence after a follow-up
period of 12 years (Kyrklund et al. 2017). Males
with bulbar fistulas had significantly better out-
come than those with prostatic and bladder neck
fistulas.
The future role of minimally invasive laparo-
scopic repair of high malformations in terms of
fecal continence remains to be seen. At the
moment, there is no evidence that minimally
invasive techniques would produce better (or
worse) functional outcomes than posterior sagit-
tal anorectoplasty; however, there may be a
higher incidence of complications related to
retained distal rectourinary communication, such
as urethral diverticula or residual rectal stump.
78.4.5.3 Urinary Tract Problems
Urinary tract anomalies occur in more than 40%
of all patients with ARM; however, the actual
incidence of long-term urological morbidity,
especially in adults, is difficult to assess because
1082
only a few long-term follow-up reports have spe-
cifically addressed urological problems.
Rintala et  al. found some degree of urinary
incontinence in 11% of 83 adult patients with
mild anomalies. In high anomalies, urinary
incontinence was found in 33% of 33 patients
(Rintala et  al. 1994). More recently, Kyrklund
et  al. found no difference in lower urinary tract
function between patients with mild anorectal
malformation and healthy controls (Kyrklund
et  al. 2017). Hassink et  al. (1993) reported uri-
nary incontinence in 22% of their 58 adult
patients with high anomalies.
78.4.5.4 Genital Anomalies, Fertility
and Sexual Problems
Genital anomalies are commonly found in male
patients with ARM, especially in patients with
high lesions. An incidence of 26% has been
reported by Metts et al. (1997). Hypospadias and
undescended testes are the most common defects.
Ejaculatory duct and seminal vesicle anomalies
occur mainly in males with high anomalies.
Weak or missing erections or retrograde ejac-
ulations occur (Rintala et  al. 1994) in a signifi-
cant proportion of patients. Normal erection
angle has been reported in only 53% of adult
patients with high or intermediate anomalies
(Konuma et al. 2006). The same group also found
abnormal or retrograde ejaculation in 41% of
their patients. Lower but significant incidences of
sexual dysfunction have been reported by
Kyrklund et  al. (2017), Rintala et  al. (1994),
Hassink et al. (1993, 1994).
The most common genital anomalies in
females with ARM are vaginal and uterine sep-
tation anomalies and vaginal agenesis. Vaginal
anomalies have been found in 30% and uterine
anomalies in 35% of female patients (Hall et al.
1985). The repair of female anorectal anomalies
may cause scarring of the genital tract, and dis-
tal vagina and vaginal outlet are most commonly
affected. Genital tract and sexual function have
been reported to be impaired in almost half of
the patients because of vaginal scarring (Hall
et al. 1985; Matley et al. 1990). Vaginal scarring
may cause dyspareunia and may interfere with
deliveries (Warne et  al. 2003). On the other
R. J. Rintala and M. P. Pakarinen
hand, a recent report found similar outcomes of
sexual function in patients with vestibular fis-
tula and healthy matched controls (Kyrklund et
al. 2017).
Late gynecological problems are especially
common in patients with cloaca. Obstruction of
the Müllerian structures with subsequent cystic
menstrual blood collections has been reported to
develop in a high percentage of postpubertal girls
with cloaca (Levitt et al. 1998; Warne et al. 2003).
However, in a series of 21 adult cloaca patients
(Levitt et  al. 1998), patent and adequate vagina
were found in 86% of the patients. Of these 21
adult patients, 57% were sexually active.
There is little information about fertility in
patients with ARM, although successful pregnan-
cies have been reported (Konuma et al. 2006). In
Rintala’s series of 83 adult patients (Rintala et al.
1992) with low malformations, 47 (57%) had off-
spring of their own. In the same study, 54% of the
healthy controls with similar age and sex distri-
bution had children of their own. On the other
hand, in another study from the same institution
(Kyrklund et al. 2017) concerning high malfor-
mations, only 39% of the patients had children,
which was significantly less than healthy con-
trols, 60% of whom had offspring. Obviously, the
low frequency of offspring in patients with high
anomalies reflects true infertility in a significant
percentage of patients. On the other hand, some
patients may avoid sexual contacts because of
defective fecal continence. Rintala (Rintala et al.
1992; Rintala et  al. 1994) reported that 20% of
the patients with high anomalies and 13% of the
patients with low anomalies avoided sexual inter-
course because of poor bowel control.
78.4.5.5 Vertebral Anomalies
and Myelodysplasias
In the literature, there are essentially no reports
concerning late problems related to vertebral
anomalies in patients with ARM. In a report from
the authors’ institution, 16% of adult patients had
spine-related symptoms, mainly chronic back
pain (Rintala et al. 1991).
Recently, much attention has been placed on
the occurrence of myelodysplasias in patients
with anorectal anomalies (Rivosecchi et al. 1995;
78  Long-Term Outcomes in Pediatric Surgery
Sato et al. 1993; Levitt et al. 1997). The impact of
spinal abnormalities (intraspinal lipomas, fatty
filum, tethered cord) on long-term bladder and
bowel function or neurological symptoms in
lower extremities is unclear. Some investigators
suggest that worsening of the function due to spi-
nal anomalies is possible (Sato et  al. 1993).
However, there appears to be no evidence to sup-
port prophylactic detethering of patients who do
not have specific symptoms related to tethering
(Levitt et al. 1997; Tuuha et al. 2004). There is
also no evidence to support the concept that
asymptomatic tethered cord impairs fecal or uri-
nary continence in patients with ARM (Levitt
et  al. 1997; Tuuha et  al. 2004; Kyrklund et  al.
2016; Taskinen et al. 2002). On the other hand,
screening for spinal abnormalities is indicated in
all patients with ARM and also in patients with
low anomalies (Rivosecchi et  al. 1995; Beek
et al. 1995). Normal vertebral anatomy on spinal
X-rays does not preclude the presence of spinal
cord abnormalities (Levitt et  al. 1997). In the
newborn, screening can be performed by ultra-
sound or by MRI at any age.
The impact of spinal and sacral defects on
fecal continence in adults has not been reported
in the literature. There is obviously an associa-
tion as patients with sacral and spinal problems
are most common in patients with high malfor-
mations who also have worse bowel function
than those with milder defects. There are no
studies that have specifically addressed the asso-
ciation of sacral and spinal status with bowel or
urinary function in patients with anorectal
malformations.
78.4.5.6 Quality of Life in Adult
Patients with Anorectal
Malformations
Despite improvements in understanding etiology
and pathology and developments in surgical tech-
niques, there is significant morbidity during
childhood and adolescence that is associated with
anorectal malformations. This is reflected by psy-
chological and social problems of both the
patients and their parents (Grano et  al. 2012).
Other factors that may affect the psychosocial
well-being of patients are multiple and often pro-
1083
tracted hospitalizations and repeated surgical
procedures, including long-term anal dilatations
(Diseth and Emblem 1996).
At adult age, defective fecal continence has
significant social consequences. Fecal conti-
nence problems are a strong predictor of quality
of life in the areas of social functioning, emo-
tional functioning and body image. Urinary
incontinence predicts poor sexual functioning
(Grano et  al. 2011). The continence-related
social problems are more common in patients
with high lesions (Rintala and Lindahl 1995;
Rintala et al. 1994). The main problem is fecal
soiling that restricts social activities. In Rintala’s
series (Rintala et  al. 1994), 85% of the adult
patients with high ARM reported social disabil-
ity related to soiling. Other problems especially
disturbing occupational life were inability to
hold back flatus and fecal urgency. Hassink
et  al. (Hassink et  al. 1994) reported that adult
patients had significantly lower educational
level than expected.
The literature reports on quality of life of adult
patients with ARM show significant incidence of
emotional problems (Rintala and Lindahl 1995;
Kyrklund et al. 2017; Rintala et al. 1994; Hassink
et al. 1994; Diseth and Emblem 1996; Hartman
et al. 2004). Up to 58% of psychiatric diagnoses
has been reported by Diseth et  al. (Diseth and
Emblem 1996). Psychosocial functioning also
appears to be more affected in patients with
severe anomalies and worse fecal continence
(Grano et al. 2011).
78.5 Conclusion and Future
Directions
The development of neonatal surgery and
improving early outcomes have brought into
light a growing group of individuals that in the
past would have died or suffered from serious
functional defects. Although early outcomes in
these patient groups are well recognized, amaz-
ingly few reports on long-term outcomes beyond
childhood are available. The medical teams tak-
ing care of neonatal surgical patients should
have a responsibility to follow-up their patients
1084
throughout the whole growth period to reveal the
true, long-term outcomes. The caregivers of the
patients have the right to know the effect of the
neonatally repaired condition on growth and
development and also the cognitive and psycho-
social consequences and the effects on fertility
and sexuality. The follow-up should also con-
tinue during adult age as aging may impact func-
tional outcome. The potential risks of malignancy
and future inheritance also need to be addressed.
The research on long-term outcomes is facili-
tated by the development of patient registries
and tracking systems. Reliable, long-term out-
come studies also require validated and stan-
dardized research tools and healthy control
material.
References
Bai Y, Chen H, Hao J et  al (2002) Long-term outcome
and quality of life after the Swenson procedure for
Hirschsprung’s disease. J Pediatr Surg 37:639–642
Baillie CT, Kenny SE, Rintala RJ et  al (1999) Long-­
term outcome and colonic motility after the Duhamel
procedure for Hirschsprung’s disease. J Pediatr Surg
34:325–329
Beek FJA, Boemers TML, Witkamp TD et al (1995) Spine
evaluation in children with ARM.  Pediatr Radiol
25:S28–S32
Conway SJ, Craigie RJ, Cooper LH et  al (2007) Early
adult outcome of the Duhamel procedure for left-sided
Hirschsprung disease  – a prospective serial assess- Hartman EE, Oort FJ, Aronson DC et al (2004) Critical
ment study. J Pediatr Surg 42:1429–1432
D’Antiga L, Goulet O (2013) Intestinal failure in children:
the European view. J Pediatr Gastr Nutr 56:118–126
Danielson J, Karlbom U, Graf W, Olsen L, Wester T Hartman EE, Oort FJ, Sprangers MA et al (2008) Factors
(2015) Posterior sagittal anorectoplasty results in bet-
ter bowel function and quality of life that pull-through
procedures. J Pediatr Surg 50:1556–1559
Danzer E, Gerdes M, D'Agostino JA, Partridge EA, Hashish MS, Dawoud HH, Hirschl RB et  al (2010)
Hoffman-Craven CH, Bernbaum J, Natalie E, Rintoul
NE, Flake AW, Adzick NS, Hedrick HL (2013)
Preschool neurological assessment in congenital dia-
phragmatic hernia survivors: outcome and perinatal Hassett S, Snell S, Hughes-Thomas A, Holmes K (2009)
factors associated with neurodevelopmental impair-
ment. Early Hum Dev 89:393–400
de Vries PA, Peña A (1982) Posterior sagittal anorecto-
plasty. J Pediatr Surg 17:638–643
Dellenmark-Blom M, Chaplin JE, Gatzinsky V, Jönsson Hassink EA, Rieu PN, Severijnen RS et  al (1993) Are
L, Abrahamson K (2015) Health-related quality of
life among children, young people and adults with
esophageal atresia: a review of the literature and
recommendations for future research. Qual Life Res Hassink EA, Rieu PN, Brugman AT et  al (1994)
24:2433–2445
R. J. Rintala and M. P. Pakarinen
Diseth TH, Emblem R (1996) Somatic function, mental
health and psychosocial adjustment of adolescents
with anorectal anomalies. J Pediatr Surg 31:638–643
Diseth TH, Bjornland K, Novik TS et  al (1997) Bowel
function, mental health, and psychosocial function
in adolescents with Hirschsprung’s disease. Arch Dis
Child 76:100–106
Escobar MA, Grosfeld JL, West KW et al (2005) Long-­
term outcomes in total colonic aganglionosis: a
32-year experience. J Pediatr Surg 40:955–961
Eypasch E, Williams JI, Wood-Dauphinee S et al (1995)
Gastrointestinal quality of life index: development,
validation, and application of a new instrument. Br J
Surg 82:216–222
Gallo G, Zwaveling S, Groen H, Van der Zee D, Hulscher
J (2012) Long-gap esophageal atresia: a meta-analysis
of jejunal interposition, colon interposition, and gas-
tric pull-up. Eur J Pediatr Surg 22:420–425
Goulet O, Ruemmele F (2006) Causes and management
of intestinal failure in children. Gastroenterology
130:S16–S28
Goulet O, Ruemmele F, Lacaille F, Colomb V (2004)
Irreversible intestinal failure. J Pediatr Gastroenterol
Nutr 38:250–269
Grano C, Aminoff D, Lucidi F, Violani C (2011) Long-­
term disease-specific quality of life in adult anorectal
malformation patients. J Pediatr Surg 46:691–698
Grano C, Aminoff D, Lucidi F, Violani C (2012) Long-­
term disease-specific quality of life in children
and adolescent patients with ARM.  J Pediatr Surg
47:1317–1322
Gutierrez IM, Kang KH, Jaksic T (2011) Neonatal
short bowel syndrome. Semin Fetal Neonat Med
16:157–163
Hall R, Fleming S, Gysler M et al (1985) The genital tract
in female children with imperforate anus. Am J Obstet
Gynecol 151:169–171
factors affecting quality of life of adult patients with
ARM of Hirschsprung’s disease. Am J Gastroenterol
99:907–913
affecting quality of life of children and adolescents
with anorectal malformations of Hirschsprung’s dis-
ease. JPGN 47:463–467
Long-term functional outcome and quality of life in
patients with high imperforate anus. J Pediatr Surg
45(1):224–230
10-year outcome of children born with anorectal mal-
formation, treated by posterior sagittal anorectoplasty,
assessed according to the Krickenbeck classification. J
Pediatr Surg 44:399–403
adults content or continent after repair for high anal
atresia? A long-term follow-up study in patients 18
years of age and older. Ann Surg 218:196–200
Quality of life after operatively corrected high ano-
78  Long-Term Outcomes in Pediatric Surgery
rectal malformation. A long-term follow-up study in
patients 18 years of age and older. J Pediatr Surg
29:773–776
Hayward MJ, Kharasch V, Sheils C, Friedman S, Dunleavy
M-J, Utter S, Zurakowski D, Jennings R, Wilson JM
(2007) Predicting inadequate long-term lung develop-
ment in children with congenital diaphragmatic her-
nia: an analysis of longitudinal changes in ventilation
and perfusion. J Pediatric Surg 42:112–116
Heikkinen M, Rintala RJ, Louhimo I (1995) Bowel
function and quality of life in adult patients with
operated Hirschsprung’s disease. Pediatr Surg Int
10:342–344
Holschneider AM (1983) Elektromanometrie des
Enddarms. Urban & Schwarzenberg, Munich-Wien-­
Baltimore, pp 213–218
Holschneider A, Hutson J, Pena A et al (2005) Preliminary
report on the international conference for the develop-
ment of standards for the treatment of anorectal mal-
formations. J Pediatr Surg 40:1521–1526
Ieiri S, Nakatsuji T, Akiyoshi J et  al (2010) Long-term
outcomes and quality of life of Hirschsprung disease
in adolescents who have reached 18 years or older –
a 47-year single institute experience. J Pediatr Surg
45:2398–2402
Jarvi K, Laitakari E, Koivusalo A, Rintala RJ, Pakarinen
MP (2010) Bowel function and gastrointestinal quality
of life among adults operated for Hirschsprung disease
during childhood. A population-based study. Ann Surg
252:977–981
Jarvi K, Koivusalo A, Rintala RJ, Pakarinen MP (2012)
Evaluation of bowel function and fecal continence
in 594 Finnish individuals aged 4-26 years. Dis Col
Rectum 55(6):671–676
Jayasekera CS, Desmond PV, Holmes JA, Kitson M,
Taylor AC (2012) Cluster of 4 cases of esophageal
squamous cell cancer developing in adults with surgi-
cally corrected esophageal atresia—time for screening
to start. J Pediatr Surg 47:646–651
Karkowski J, Pollock WF, Landon CW (1973) Imperforate
anus. Eighteen to thirty year follow-up study. Am J
Surg 126:141–147
Kiesewetter WB, Chang JHT (1977) Imperforate anus: a
five to thirty year follow-up perspective. Prog Pediatr
Surg 10:110–120
Killelea BK, Lazar EL, Vitale MG (2006) Principles of
outcome analysis. In: Stringer MD, Oldham KT,
Mouriquand PDE (eds) Pediatric surgery and urol-
ogy, Chapter 2, 2nd edn. Cambridge University Press,
Cambridge, pp 17–28
Koivusalo A, Pakarinen M, Rintala RJ, Lindahl H (2004)
Does postoperative pH monitoring predict compli-
cated gastroesophageal reflux in patients with esopha-
geal atresia? Pediatr Surg Int 20:670–674
Koivusalo A, Pakarinen M, Vanamo K, Lindahl H, Rintala
RJ (2005a) Health-related quality of life in adults after
repair of congenital diaphragmatic defects: a question-
naire study. J Pediatr Surg 40:1376–1381
Koivusalo A, Pakarinen MP, Turunen P, Saarikoski H,
Lindahl H, Rintala RJ (2005b) Health-related quality
1085
of life in adult patients with esophageal atresia: a ques-
tionnaire study. J Pediatr Surg 40:307–312
Koivusalo A, Pakarinen MP, Rintala RJ (2007) The
cumulative incidence of significant gastrooesopha-
geal reflux in patients with oesophageal atresia
with a distal fistula: a systematic clinical, pH-met-
ric, and endoscopic follow-up study. J Pediatr Surg
42:370–374
Koivusalo AI, Mikko P, Pakarinen MP, Lindahl HG,
Rintala RJ (2008a) The cumulative incidence of sig-
nificant gastroesophageal reflux in patients with con-
genital diaphragmatic hernia—a systematic clinical,
pH-metric, and endoscopic follow-up study. J Pediatr
Surg 43:279–282
Koivusalo AI, Pakarinen MP, Pauniaho SL, Rintala RJ
(2008b) Antegrade continence enema in the treatment
of congenital fecal incontinence beyond childhood.
Dis Colon Rectum 51:1605–1610
Koivusalo AI, Pakarinen MP, Rintala RJ (2013) Modern
outcomes of oesophageal atresia: single centre expe-
rience over the last twenty years. J Pediatr Surg
48:297–303
Konuma K, Ikawa H, Kohno M et al (2006) Sexual prob-
lems in male patients older than 20 years with anorec-
tal malformations. J Pediatr Surg 41:306–309
Kurvinen A, Nissinen MJ, Gylling H, Miettinen TA,
Lampela H, Koivusalo AI, Rintala RJ, Pakarinen MP
(2011) Effects of long-term parenteral nutrition on
serum lipids, plant sterols, cholesterol metabolism and
liver histology in pediatric intestinal failure. J Pediatr
Gastr Nutr 53:440–446
Kurvinen A, Nissinen MJ, Andersson S, Korhonen P,
Ruuska T, Taimisto T, Kalliomäki M, Lehtonen L,
Sankilampi U, Arikoski P, Saarela T, Miettinen TA,
Gylling H, Pakarinen MP (2012) Parenteral plant ste-
rols and intestinal failure associated liver disease in
neonates: a prospective nationwide study. J Pediatr
Gastr Nutr 54:803–811
Kyrklund K, Pakarinen MP, Taskinen S et al (2016) Spinal
cord anomalies in patients with anorectal malforma-
tions without severe sacral abnormalities or menin-
gomyelocele: outcomes after expectant, conservative
management. J Neurosurg Spine 25:782–789
Kyrklund K, Pakarinen MP, Rintala RJ (2017) Long-term
bowel function, quality of life and sexual function in
patients with anorectal malformations treated during
the PSARP era. Sem Pediatr Surg 26:336–342
Langemeijer RATM, Molenaar JC (1991) Continence
after posterior sagittal anorectoplasty. J Pediatr Surg
26:587–590
Levitt MA, Patel M, Rodriguez G et al (1997) The teth-
ered spinal cord in patients with anorectal malforma-
tions. J Pediatr Surg 32:462–468
Levitt MA, Stein DM, Pena A (1998) Gynecologic con-
cerns in the treatment of teenagers with cloaca. J
Pediatr Surg 33:188–193
Lindahl H, Rintala R, Sariola H, Louhimo I (1990)
Cervical Barrett's esophagus: a common complica-
tion of gastric tube reconstruction. J Pediatr Surg
25:446–448
1086
Lindahl H, Rintala R, Sariola H, Louhimo I (1992) Long-­
term endoscopic and flow cytometric follow-up of
colon interposition. J Pediatr Surg 27:859–861
Louhimo I, Lindahl H (1983) Esophageal atresia: primary
results of 500 consecutively treated patients. J Pediatr
Surg 18:217–229
Mäkitie O, Heikkinen M, Kaitila I, Rintala R (2002)
Hirschsprung’s disease in cartilage-hair hypoplasia
has poor prognosis. J Pediatr Surg 37:1585–1588
Malmström K, Lohi J, Lindahl H et al (2008) Longitudinal
follow-up of bronchial inflammation, respiratory
symptoms, and pulmonary function in adolescents
after repair of esophageal atresia with tracheoesopha-
geal fistula. J Pediatr 153:396–401
Malone PS, Ransley PG, Kiely EM (1990) Preliminary
report: the antegrade continence enema. Lancet
336:1217–1218
Matley PJ, Cywes S, Berg A et al (1990) A 20-year fol-
low-­up of children born with vestibular anus. Pediatr
Surg Int 5:37–40
Metts JC III, Kotkin L, Kasper S et  al (1997) Genital
malformations and coexistent urinary tract and spinal
anomalies in patients with imperforate anus. J Urol
158:1298–1300
Moore SW, Albertyn R, Cywes S (1996) Clinical outcome
and long-term quality of life after surgical correction of
Hirschsprung’s disease. J Pediatr Surg 31:1496–1502
Moya FR, Thomas VL, Romaguera J, Mysore MR,
Maberry M, Bernard A, Freund M (1995) Fetal lung
maturation in congenital diaphragmatic hernia. Am J
Obstet Gynecol 173:1401–1405
Muratore CS, Utter S, Jaksic T, Lund DP, Wilson JM
(2001) Nutritional morbidity in survivors of congeni-
tal diaphragmatic hernia. J Pediatr Surg 36:1171–1176
Mutanen A, Mäkitie O, Pakarinen MP (2013a) Risk of
metabolic bone disease is increased both during and
after weaning off parenteral nutrition in pediatric
intestinal failure. Horm Res Paediatr 25:227–235
Mutanen A, Lohi J, Heikkilä P, Koivusalo A, Rintala R,
Pakarinen MP (2013b Mar 4) Persistent abnormal
liver fibrosis after weaning off parenteral nutrition in
pediatric intestinal failure. Hepatology. https://doi.
org/10.1002/hep.26360
Nixon HH, Puri P (1977) The results of treatment of ano-
rectal anomalies: a thirteen to twenty year follow-up. J
Pediatr Surg 12:27–37
Olieman JF, Penning C, Poley MJ, Utens E, Hop W,
Tibboel D (2012) Impact of infantile short bowel syn-
drome on long-term health-related quality of life: a
cross sectional study. J Pediatr Surg 47:1309–1316
Oliemen JF, Penning C, Spoel M, Ijsseltijn H, van den
Hoonaard TL, Excher JC, Bax NM, Tibboel D (2012)
Long-term impact of infantile short bowel syn-
drome on nutritional status and growth. Br J Nutr
107:1489–1497
Ong NT, Beasley SW (1990) Long-term functional results
after perineal surgery for low anorectal anomalies.
Pediatr Surg Int 5:238–240
Pakarinen MP, Rintala RJ, Koivusalo A, Heikkinen
M, Lindahl H, Pukkala E (2005) Increased inci-
R. J. Rintala and M. P. Pakarinen
dence of medullary thyroid carcinoma in patients
treated for Hirschsprung’s disease. J Pediatr Surg
40:1532–1534
Pakarinen MP, Koivusalo AI, Rintala RJ (2009) Outcomes
of intestinal failure – a comparison between children
with short bowel and dysmotile intestine. J Pediatr
Surg 44:2139–2144
Pakarinen MP, Kurvinen A, Gylling H, Miettinen TA,
Pesonen M, Kallio M, Koivusalo AI, Nissinen MJ
(2010) Cholesterol metabolism in pediatric short
bowel syndrome after weaning off parenteral nutri-
tion. Dig Liver Dis 42:554–559
Pakarinen MP, Kurvinen A, Koivusalo AI, Iber T, Rintala
RJ (2013a) Long-term controlled outcomes after
autologous intestinal reconstruction surgery in treat-
ment of severe short bowel syndrome. J Pediatr Surg
48:339–344
Pakarinen MP, Kurvinen A, Koivusalo AI, Makisalo H,
Jalanko H, Rintala RJ (2013b) Surgical treatment and
outcomes of severe pediatric intestinal motility dis-
orders requiring parenteral nutrition. J Pediatr Surg
48:333–338
Pena A (1995) Anorectal malformations. Semin Pediatr
Surg 4:35–47
Pena A, Hong A (2000) Advances in the management of
anorectal malformations. Am J Surg 180:370–376
Pichler J, Chomtho S, Fewtrell M, Macdonald S, Hill SM
(2013) Growth and bone health in pediatric intestinal
failure patients receiving long-term parenteral nutri-
tion. Am J Clin Nutr 97(6):1260–1269
Rangel SJ, Lawal TA, Bischoff A et al (2011) The appen-
dix as a conduit for antegrade continence enemas in
patients with anorectal malformations: lessons learned
from 163 cases treated over 18 years. J Pediatr Surg
46:1236–1242
Reding R, de Ville de Goyet J, Gosseye S et  al (1997)
Hirschsprung’s disease: a 20-year experience. J
Pediatr Surg 32:1221–1225
Rentea RM, Halleran DR, Wood RJ, Levitt MA (2020)
The role of laparoscopy in anorectal malformations.
Eur J Pediatr Surg 30:156–163
Rintala R, Lindahl H (1995) Is normal bowel function
possible after repair of intermediate and high anorectal
malformations. J Pediatr Surg 30:491–494
Rintala RJ, Lindahl H (1999) PSARP is superior to sacro-
perineal/sacroabdominoperineal pull-through - a long
term follow-up study in males with high anorectal
anomalies. J Pediatr Surg 34:334–337
Rintala RJ, Pakarinen M (2006) Hirschsprung’s disease.
In: Stringer M, Oldham KT, Mouriquand PD (eds)
Pediatric surgery and urology. Long-term outcomes.
Cambridge University Press, Cambridge, pp 385–400
Rintala RJ, Pakarinen MP (2010) Outcome of anorectal
malformations and Hirschsprung’s disease beyond
childhood. Semin Pediatr Surg 19:160–167
Rintala R, Lindahl H, Louhimo I (1991) Anorectal mal-
formations - results of treatment and long term follow-
­up of 208 patients. Pediatr Surg Int 6:36–41
Rintala R, Mildh L, Lindahl H (1992) Fecal continence
and quality of life in adult patients with an oper-
78  Long-Term Outcomes in Pediatric Surgery
ated low anorectal malformation. J Pediatr Surg
27:902–905
Rintala R, Mildh L, Lindahl H (1994) Fecal continence
and quality of life in adult patients with an operated
high or intermediate anorectal malformation. J Pediatr
Surg 29:777–780
Rintala RJ, Lindahl HG, Rasanen M (1997) Do children
with repaired low anorectal malformations have nor-
mal bowel function? J Pediatr Surg 32:823–826
Rivosecchi M, Lucchetti MC, De Gennaro M et  al
(1995) Spinal dysraphism detected by magnetic reso-
nance imaging in patients with anorectal anomalies:
incidence and clinical significance. J Pediatr Surg
30:488–490
Ruttenstock E, Puri P (2009) A meta-analysis of clinical
outcome in patients with total intestinal aganglionosis.
Pediatr Surg Int 25:833–839
Salvia G, Guarino A, Terrin G, Cascioli C, Plaudetto R,
Indrio F, Lega L, Stronati M, Corvaglia L, Tagliabue P,
De Curtis M (2008) Neonatal onset intestinal failure:
an Italian multicenter study. J Pediatr 153:674–676
Sanchez SE, McAteer JP, Goldin AB, Horslen S, Huebner
CE, Javid JP (2013 May 5) Health-related quality of
life in children with intestinal failure. J Pediatr Gastr
Nutr 57(3):330–334
Sato S, Shirane R, Yoshimoto T (1993) Evaluation of teth-
ered cord syndrome associated with anorectal malfor-
mations. Neurosurgery 32:125–127
Sistonen SJ, Koivusalo A, Lindahl H, Pukkala E, Rintala
RJ, Pakarinen MP (2008) Cancer after repair of
­esophageal atresia: population-based long-term fol-
low-up. J Pediatr Surg 43:602–605
Sistonen SJ, Helenius I, Peltonen J, Sarna S, Rintala RJ,
Pakarinen MP (2009) Natural history of spinal anoma-
lies and scoliosis associated with esophageal atresia.
Pediatrics 124:e1198–e1204
Sistonen SJ, Koivusalo A, Nieminen U, Lindahl H, Lohi
J, Kero M, Kärkkäinen PA, Färkkilä MA, Sarna
S, Rintala RJ, Pakarinen MP (2010a) Esophageal
morbidity and function in adults with repaired
esophageal atresia with tracheoesophageal fistula:
a population-­based long-term follow-up. Ann Surg
251:1167–1173
Sistonen S, Malmberg P, Malmström K, Haahtela T,
Sarna S, Rintala RJ, Pakarinen MP (2010b) Repaired
oesophageal atresia: respiratory morbidity and pulmo-
nary function in adults. Eur Respir J 36:1106–1112
Spoel M, van der Cammen-van Zijp M, Hop W, Tibboel
D, de Jongste JC, IJsselstijn H (2013) Lung function in
young adults with congenital diaphragmatic hernia, a
longitudinal evaluation. Pediatr Pulmonol 48:130–137
Stege G, Fenton A, Jaffray B (2003) Nihilism in the
1990s: the true mortality of congenital diaphragmatic
hernia. Pediatrics 112:532–535
Stephens FD, Smith ED (1971) Ano-rectal malformations
in children. Year Book Medical Publishers, Chicago
Stephens FD, Smith ED (1986) Classification, identifica-
tion and assessment of surgical treatment of anorectal
anomalies. Pediatr Surg Int 1:200–205
1087
Taskinen S, Valanne L, Rintala R (2002) Effect of spinal
cord abnormalities on the function of the lower urinary
tract in patients with anorectal abnormalities. J Urol
168:1147–1149
Taylor I, Duthie HL, Zachary RB (1973) Anal continence
following surgery for imperforate anus. J Pediatr Surg
8:497–503
Templeton JM, Ditesheim JA (1985) High imperforate
anus  - quantitative result of long-term fecal conti-
nence. J Pediatr Surg 20:645–652
Trusler GA, Wilkinson RH (1962) Imperforate anus: a
review of 147 cases. Can J Surg 5:269–277
Tuuha SE, Aziz D, Drake J et al (2004) Is surgery neces-
sary for asymptomatic tethered cord in anorectal mal-
formation patients? J Pediatr Surg 39:773–777
van den Hout L, Reiss I, Felix JF, Hop WC, Lally PA,
Lally KP, Tibboel D (2010) Risk factors for chronic
lung disease and mortality in newborns with congeni-
tal diaphragmatic hernia. Neonatology 98:370–380
Vanamo K, Rintala RJ, Sovijärvi A, Jääskeläinen J,
Lindahl H, Louhimo I (1996a) Long-term pulmonary
sequelae in survivors of congenital diaphragmatic
defects. J Pediatr Surg 131:1096–1100
Vanamo K, Rintala RJ, Lindahl H, Louhimo I (1996b)
Long-term gastrointestinal morbidity in patients with
congenital diaphragmatic defects. J Pediatr Surg
31:551–554
Vanamo K, Peltonen J, Rintala RH, Lindahl H,
Jääskeläinen J, Louhimo I (1996c) Chest wall and
spinal deformities in adults with congenital diaphrag-
matic defects. J Ped Surg 31:851–854
Varni JM, Limbers CA, Burwinkle TM (2007) Impaired
health-related quality of life in children and adoles-
cents with chronic conditions: a comparative analysis
of 10 disease clusters and 33 disease categories/severi-
ties utilizing the PedsQL 4.0 Generic Core Scales.
Health Qual Life Outcomes 5:43–58
Waag KL, Loff S, Zahn K, Ali M, Hien S, Kratz M, Neff
W, Schaffelder R, Schaible T (2008) Congenital dia-
phragmatic hernia: a modern day approach. Semin
Pediatr Surg 17:244–254
Wales PW, de Silva N, Kim J, Lecce L, To T, Moore A
(2004) Neonatal short bowel syndrome: population-­
based estimates of incidence and mortality rates. J
Pediatr Surg 39:690–695
Ware JE Jr, Sherbourne CD (1992) The MOS 36-item
short-form health survey (SF-36). I.  Conceptual
framework and item selection. Med Care 30:473–483
Warne SA, Wilcox DT, Creichton S et  al (2003) Long
term gynecological outcome of patients with persis-
tent cloaca. J Urol 170:1493–1496
Wildhaber BE, Teitelbaum DH, Coran AG (2005) Total
colonic Hirschsprung’s disease: a 28-year experience.
J Pediatr Surg 40:203–207
Yeung CK, Kiely EM (1991) Low anorectal anomalies: a
critical appraisal. Pediatr Surg Int 6:333–335
 Part IX
Hepatobiliary
 Biliary Atresia
Hannah Thompson and Mark Davenport
79.1 Introduction
Biliary atresia (BA) can be a devastating disease
of infancy, invariably leading, if untreated, to cir-
rhosis, liver failure and death. It is the common-
est indication for paediatric liver transplantation
throughout the developed world.
In common with North America and Western
Europe, the incidence in the United Kingdom is
about 1 in 17,000 live births. It is clearly higher
in countries such as Japan, and probably China,
at about 1 in 8000. All large series have a slight
female preponderance, though this is most obvi-
ous in the syndromic forms. These are a group of
infants who are distinguished by the presence of
other abnormalities and maybe a poorer progno-
sis. About 10% of infants in European and North
American series (but only <2% in Chinese and
Japanese series) will have a specific constellation
of anomalies which we have termed the Biliary
Atresia Splenic Malformation (BASM) syn-
drome. Such anomalies include: polysplenia or
asplenia; situs inversus; preduodenal portal vein;
absence of the inferior vena cava; malrotation
and congenital heart abnormalities.
H. Thompson · M. Davenport (*)
Department of Paediatric Surgery, Kings College
Hospital, London, UK
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_79
79
79.2 Etiology
The cause of BA is unknown and it is most likely
that a number of factors may induce or cause the
final common pathology of biliary inflammation,
luminal obliteration and fibrosis (etiological
heterogeneity).
A key study from Houston, TX was published
in 2011 (Harpavat et  al.). This retrospectively
identified levels of conjugated and total bilirubin
in the first 48  hours of life and found that the
elevated conjugated (not total) levels could be
distinguished in those infants that had later pre-
sented as BA to their hospital. This strongly sug-
gested an obstructed biliary tree at the time of
birth.
79.2.1 Congenital Embryopathy
Infants with BASM have a parallel range of
anomalies, which could only have arisen at key
points in organ development within the embry-
onic phase of development (up to the eighth week
of gestation). A genetic aetiology in these infants,
although perhaps likely, has not been convinc-
ingly shown in the majority. Thus, there are cer-
tainly genes where there is obvious overlap
between biliary and visceral (including acquisi-
tion of normal right-left determination) develop-
ment (e.g. CFC-1, INV), but proof of linkage is
lacking in clinical practice. Recently, an American
1091
1092
study identified mutations in the PKD1L1 gene,
possibly linked to cilial abnormalities, in about
10% of their cohort. There is also some evidence
that such infants have been exposed to an abnor-
mal intrauterine environment during the first tri-
mester (e.g. maternal diabetes).
There is a remarkable similarity between the
normal biliary appearance of the fetal porta hepa-
tis at ∼12  weeks’ gestation and what is seen
pathologically in biliary atresia. There is a sieve-­
like appearance with multiple, microscopic bili-
ary channels, some of which will be selected and
conglomerate into the two macroscopic hepatic
ducts with the remainder being deleted. One
hypothesis suggests that BA may result if there is
developmental arrest and failure of maturation at
this stage.
79.2.2 Viral Exposure
There are a range of hepatotropic cholangio-
pathic viruses (e.g. Reovirus type 2, cytomegalo-
virus, rotavirus) which can possibly cause biliary
damage during the perinatal period. Such viruses
are ubiquitous and are a common source of gas-
trointestinal symptoms during infancy, and yet
the vast majority clearly do not cause BA. Mouse
models are available where perinatal viral expo-
sure leads to biliary pathology reminiscent of
BA, with a similar host immunopathological
response. Serological evidence in humans is con-
tradictory and although some studies using
RT-PCR and PCR techniques have shown isola-
tion of both viral RNA and DNA, others have not.
We reviewed our own series for those with IgM
antibodies to CMV, finding them in about 10% of
cases. In comparison to CMV IgM-ve infants,
they were older at presentation; had a greater
degree of fibrosis and mononuclear infiltrate in
their liver biopsies, and had a poorer response to
surgery. Their Guthrie dried blood spot was also
positive for CMV DNA where we could find it.
This represents the first real clinical validation,
though of course not proof, for a viral-mediated
BA sub-group. CMV is much more prevalent in
Asia and South Africa and is possible reason for
the increased overall prevalence in those regions.
H. Thompson and M. Davenport
It may not be direct viral damage that causes the
problem more an immune-mediated injury in a
normally developed bile duct. However, given
the prevalence of such viruses in the community
it seems odd that BA remains such a rare disease,
and it may be that there is some (possibly) genetic
predisposition to develop biliary destruction with
the virus acting as an initiating immune trigger.
79.3 Pathology
Both intra- and extra-hepatic parts of the biliary
tree are affected and is characterised by cholan-
giolar plugging and proliferation, moderate giant
cell transformation, variable degrees of fibrosis,
an inflammatory infiltrate in the liver and fibroin-
flammatory luminal obliteration within the extra-
hepatic bile ducts. In about 5% of cases, there is
cyst formation within the biliary tree which may
or may not contain bile (Figs. 79.1a and b).
The level at which the lumen is obliterated is
the basis for the commonest classification in clin-
ical use (Table 79.1). Type 3 is the usual manifes-
tation of this with a level of obstruction within
the porta hepatis combined with atrophic non-­
bile containing gallbladder and obliterated extra-
hepatic biliary tree (Fig. 79.2).
Inflammation appears to play an important
pathogenic role in biliary atresia. Cell adhesion
molecules are proteins involved in cell-cell bind-
ing and the initiation and maintenance of the
inflammatory response. Intercellular adhesion
molecule (ICAM-1) and vascular cell adhesion
molecule (VCAM-1) are expressed on both hepa-
tocytes and sinusoidal cell membranes and act to
bind circulating leukocytes by interaction with
cell-surface integrins leading to transendothelial
migration. There is abnormal expression of the
adhesion molecules, ICAM-1 and VCAM-1, but
not E-selectin, on the sinusoidal endothelium of
virtually all infants with BA, with abnormal
expression of MHC class 2 antigens and upregu-
lation of pro-inflammatory cellular adhesion
molecules ICAM and VCAM in up to 40% of
cases of BA. In our original study, the small cell
infiltrate was largely composed of CD4+ T lym-
phocytes and CD56+ NK cells and more recent
79  Biliary Atresia 1093

a b

Fig. 79.1  Cystic biliary atresia, detected antenatally at typically abnormal intrahepatic biliary ductule “cloud”
22  weeks gestation. Translucent obliterated distal com- appearance. Cut surface of resected specimen appeared
mon bile duct (yellow sling) and thick-walled mucus-­ solid (i.e. Type 3)
containing cyst (a) The cholangiogram (b) showed

Table 79.1 Classification (Japanese Association of

Paediatric Surgeons) of Biliary Atresia
Type Incidence Description
1 5% Level of obstruction within
common bile duct.
The gallbladder therefore contains
bile. Typically these are cystic. The
more proximal intrahepatic ducts are
abnormal (scanty, irregular and
etiolated), sometimes with a hazy,
cloud-like appearance rather than as
actual ducts
2 3% Level of obstruction within the
common hepatic duct Fig. 79.2  Extra-abdominal mobilised liver with atrophic
The gallbladder will not contain bile solid gallbladder just visible within its fossa
but a transection of the proximal
remnant should show two distinct
bile-containing lumens. ary atresia, not seen in familial cholestatic syn-
3 >90% Obstruction is within the porta dromes. These may be resident as Kupffer cells
hepatis with no visible bile-­ or systemic and recruited and, when activated
containing proximal lumen.
The gallbladder may look normal but and proliferating, appear to be a key component
if so will only contain clear mucus. A in the prolongation of the inflammatory response.
communication with the duodenum Certainly, there is some evidence that the degree
may be shown on cholangiography of macrophage infiltration and activation may
have prognostic importance and worsen
results indicate that the T cell infiltrate is pre- outcome.
dominantly Th1 and Th17 cells with the former An abnormal parallel expression of soluble
having a particular association in those with cellular adhesion molecules and a range of solu-
CMV IgM antibodies. ble cytokines can be found in the circulation of
Macrophages/monocytes also seem to play an infants with BA. Thus, there are elevated levels
important role in the presentation of antigenic of soluble ICAM-1 and soluble VCAM-1 both at
material (possibly derived from viruses) and in the time of surgery and for up to 6 months after.
the initiation of later fibrosis, a hallmark of bili- Similarly, there appears to be a non-polarized
1094
soluble cytokine response characterized by often
gross elevations of TNF-a, interferon-γ, IL-2,
IL-4, IL-18, etc.
It has been known since the 1960s that there is
an animal form of BA which occurs in sheep.
This arose from observations in Australian sheep
farms where it occurred following specific cli-
matic conditions and where the pregnant ewe
grazed in periods of drought on a particular
weed—the Red Crumbweed (dysphania spp.).
Later, a specific toxin was identified from plant
samples, now known as bilatresone that caused
specific injury to the fetal biliary tract. While it
seems that exposure to this toxin is unlikely to
explain human BA, Red Crumbweed not being a
usual part of the diet, it has been used to elucidate
various cellular pathways in the laboratory.
79.4 Clinical Features
The key clinical triad observable in virtually all
infants with biliary atresia is conjugated jaun-
dice, pale, acholic stools and dark urine
(Fig. 79.3). There is an inability to excrete conju-
gated (i.e. water soluble) bilirubin into the gastro-
intestinal tract, which is then excreted into the
urine thereby darkening its colour. Some infants
will have had an abnormal antenatal ultrasound
Fig. 79.3  Triad of
clinical features
H. Thompson and M. Davenport
scan (∼5%) because of the cystic change in the
biliary tree, and some infants present because of
the other abnormalities associated with BASM
(e.g. cardiac anomalies or malrotation). A small
proportion of infants will present with features of
vitamin K-dependent coagulopathy and bleeding,
particularly in those communities where routine
neonatal administration of parenteral vitamin K
is not usual practice. Clinical features of cirrhosis
such as ascites, splenomegaly or a hard liver are
unusual in infants presenting within 80  days of
age.
79.4.1 Diagnosis
It is possible to make the correct diagnosis in
>80% of cases before a laparotomy. The key
diagnostic investigations in our centre involve a
detailed ultrasound of the liver and biliary tree
followed by a percutaneous liver biopsy. Other
centres have omitted the liver biopsy but rely
more on the surgeon at the time of laparotomy to
make the diagnosis.
Other techniques, such as ERCP (endoscopic
retrograde cholangiopancreatography), can be
used although are practically confined to the larg-
est of specialist centres. Currently, MRCP (mag-
netic retrograde cholangiopancreaticography)
79  Biliary Atresia
does not appear to offer much advantage over
standard ultrasonography simply because of the
lack of detail required for a positive diagnosis.
Similarly, radioisotope hepatobiliary imaging
lacks sufficient discrimination in borderline cases
to be really useful.
79.4.2 Differential Diagnosis
This is long and needs perseverance to work out.
Surgical causes other than BA are uncommon but
may include obstructed choledochal malforma-
tions, spontaneous perforation of the bile duct and
the inspissated bile syndrome. Ultrasonography
and MRCP will show a dilated common bile duct
(usually >5 mm) or obvious cystic anomaly with
a significant choledochal malformation. There
should be some free peritoneal fluid and perhaps
an echogenic mass at the porta hepatis in sponta-
neous biliary perforation with bile-staining of the
skin in most. Infants with inspissated bile syn-
drome are usually premature and the ultrasound
shows a dilated biliary tree with inspissated bile
present. These infants usually require percutane-
ous cholangiography and occasionally surgery to
clear the obstructed bile ducts.
Common medical causes include neonatal
hepatitis, a1-antitrypsin deficiency, giant cell
hepatitis, CMV hepatitis, Progressive Familial
Intrahepatic Cholestasis (PFIC) and cystic fibro-
sis. Appropriate tests of exclusion include liver
biopsy, viral serology, genetic studies, and a
sweat test. Biliary hypoplasia may be a feature of
Alagille’s syndrome (abnormal “elfin” facies,
butterfly vertebrae, pulmonary stenosis) and can
certainly cause diagnostic confusion, even at lap-
arotomy and cholangiography.
79.5 Management
There are two possible strategies for infants with
BA.  Most infants are suitable for an attempt at
restoration of bile flow and preservation of their
native liver using a Kasai-type portoenterostomy
(KPE). If this fails, or the child develops signifi-
cant complications of chronic liver disease, then
1095
liver transplantation will be required if it is avail-
able. The alternative is simply to list all infants
with BA for liver transplantation as a primary
procedure. In most institutions, this strategy is
reserved only for infants where an attempt at
KPE would be regarded as futile and these tend
to be late-presenters (>100  days) who have
already obvious features of cirrhosis such as
ascites, etc. In the national England and Wales
series, this was used in 4% of all infants present-
ing with BA.
79.5.1 Kasai Portoenterostomy
The aim of surgery is to excise all extrahepatic
biliary remnants allowing a wide portoenteros-
tomy onto a portal plate, denuded of all tissue
(Fig.  79.4). In most, this will expose sufficient
transected microscopic bile ductules which retain
connections with the primitive intrahepatic bile
ductule system to allow restoration of at least a
degree of biliary drainage (Fig. 79.5). This should
be the object not only in Type 3 BA, but also in
those with Types 1 and 2 BA.
Surgery consists of a liver mobilisation and
eversion onto the abdominal wall (Fig.  79.1) to
allow full exposure of the porta hepatis and a
detailed meticulous dissection. The dissection
itself must be wide and expose the origin of the
umbilical vein from the left portal vein in the fossa
of Rex, the extrahepatic bifurcation of the right
Fig. 79.4  Close-up of porta hepatis showing appropriate
level of transection
1096
Fig. 79.5  H&E photomicrograph of transected proximal
biliary remnants from Type 3 biliary atresia showing a
predominantly fibro-inflammatory stroma within which
are multiple biliary ductules (10–100 μm in diameter)
posterior portal pedicle (Rouviere’s fossa) and the
caudate lobe behind the portal vein posteriorly. A
portoenterostomy using a retrocolic Roux loop
(∼40 cm) completes the reconstruction.
Laparoscopic-assisted portoenterostomy has
been described and is in use in a few centres in
China and Japan. Realistically, it doesn’t offer
any advantages in terms of shorter hospital stay,
quicker time to full feeds, etc. and is a signifi-
cantly longer procedure to perform. Furthermore,
a different less radical dissection has to be per-
formed, and the jejuno-jejunostomy is done by
extracting the intestine outside by widening the
umbilical wound. It is not performed in any cen-
tre in Europe or North America.
79.5.2 Adjuvant Therapy
There is currently no consensus on a standard
adjuvant drug regimen but there is some evidence
of benefit in some agents used.
The pathogenic effect of inflammation in bili-
ary atresia can be mediated with adjuvant corti-
costeroids. While there certainly has been
disagreement surrounding their use, a high-dose
regime of steroids has been shown to improve
clearance of jaundice and perhaps increase native
liver survival in single-centre single-surgeon pro-
spective series. This could be set against the pub-
lished results of the START randomised,
placebo-controlled trial in North America.
H. Thompson and M. Davenport
However, this trial was powered to detect a 25%
improvement which was somewhat optimistic.
There was a difference of about 15% in the ste-
roid arm in those comping to KPE at <70 days,
interestingly the same difference in the afore-
mentioned single centre, single surgeon series. A
short (4–6 weeks) course of high-dose steroids is
used in most centres in Europe and Asia. In the
latter this is often repeated if there is recurrence
of jaundice post-clearance.
Post-operative cholangitis is associated with
poor outcomes and antibiotics may reduce the
risk of post-operative/ recurrent cholangitis.
Their use, dose, duration and type are highly
variable however. Most commonly in Europe, a
combination of piperacillin-tazobactam and gen-
tamicin is thought to provide appropriate cover.
Ursodeoxycholic acid is used to treat cho-
lestasis in the adult population and is very com-
monly used as adjuvant therapy in BA due to its
upregulation of bile-acid transporters and there-
fore choleretic effect. No large randomised con-
trolled trials have been conducted, but there is
some evidence in a crossover trial that does indi-
cate a biochemical benefit.
We have reported our experience with anti-­
viral therapy as an adjuvant in infants with CMV
associated, BA but placebo-controlled trials have
not been conducted. We used a variety of regi-
mens, initially intravenous ganciclovir but now
oral valganciclovir alone and this resulted in
improvement in jaundice clearance rate (21% to
75% clearance) and a reduction of the need to
transplant in this BA subset.
Vitamin supplementation is an important con-
sideration for biliary atresia patients as they suf-
fer nutritional deficiencies due to reduced enteric
bile acids facilitating fat absorption. Invariably
they are vitamin D deficient at presentation,
regardless of the subtype and this is more pro-
nounced in non-Caucasian infants. This defi-
ciency often persists after KPE despite
replacement and particularly in icteric infants, so
higher-dose supplementation and regular routine
measurement are essential. Appropriate dietetic
advice on calorie and protein supplementation
should also be sought, particularly if growth fail-
ure is shown.
 79  Biliary Atresia
79.5.3 Prognostic Factors
Biliary atresia remains a rare disease and one of
the key prognostic factors is related to the experi-
ence of the surgeon and the centre where it is
treated. The other main feature influencing out-
come is the age at which the KPE is carried out.
Biliary atresia is a progressive disease which can
only be halted by restoration of effective biliary
drainage. This effect is actually difficult to quan-
tify and almost certainly is not a straightforward
linear relationship (Fig.  79.6). However, proba-
bly there is a finite time point beyond which the
liver is irretrievable, so the worse outcomes are
seen in the late Kasai’s (arbitrarily >100  days)
with established cirrhosis. Early KPE (<45 days)
is associated with better native liver survival and
clearance of jaundice particularly in patients
receiving adjuvant steroid therapy. It may be that
the effects of steroids are limited or inhibited by
an increasing degree of fibrosis and onset of
cirrhosis.
Infants with BASM have poorer prognosis but
this seems to be due to the higher numbers with
clinically significant cardiac anomalies (see
later). The degree of liver fibrosis at the time of
KPE seems comparable to age-matched non-­
syndromic infants. Premature infants with BA
also have similar outcomes to term infants,
despite typically a delayed diagnosis. The com-
Isolated BA
100
clearance jaundice (%)
80 P = 0.005
60
40
20
0
0 25 50 75
Age at KPE
Fig. 79.6 Effect of Age on Clearance of Jaundice:
England and Wales data (1999–2018). Age-cohort analy-
sis—Each black circle is the probability of an age-defined
cohort (e.g. <30 days, 31–40 days, etc. at KPE). The sta-
tistic is a Chi2 for trend = 8.2; P = 0.005
1097
monest reason for their prematurity is that they
are one of twins.
CMV IgM + ve has already been highlighted
as a negative prognostic factor.
Cardiac anomalies are a major cause of mor-
tality in BA and we have recently used the term
Cardiac-Associated Biliary Atresia (CABA) to
emphasise this. Concurrent cardiac disease has
been identified in up to 10% of infants with BA—
and in some this is untreatable. We also recom-
mend that the cardiac abnormality is treated first
to allow normalization of cardiac physiology,
before the KPE.
79.6 Complications
The main complication is simply that the proce-
dure doesn’t work. About 10–20% of all Kasais
will have no effect on the course of the liver dis-
ease with persisting acholic stools and increasing
jaundice. This is probably a function of the lack of
exposed biliary ductules at the porta hepatis rather
than the degree of established liver fibrosis and
can only be treated by transplantation.
Alternatively, restored bile flow in about 50% will
be sufficient to allow their bilirubin levels to fall
to normal with an excellent long-term prognosis.
Nonetheless, it is impossible to say that, even in
these children, they are “cured” as, if biopsied, the
liver will show cirrhosis in over half. The remain-
der will have a degree of restoration of bile flow as
evidenced by a fall in bilirubin, and although
some maintain reasonable biochemical liver func-
tion and absence of complications for longer peri-
ods, most tend to deteriorate over time.
79.7 Cholangitis
Re-establishment of bile drainage exposes the
child to the risk of ascending cholangitis. This is
100
seen in up to 40% of most series and is particu-
larly prevalent in the first 2  years. The usual
organisms are enteric in origin (e.g. Escherichia
coli, Pseudomonas, Klebsiella spp.) and present
with pyrexia and a worsening of jaundice and
biochemical liver function.
1098
The diagnosis can be difficult to establish
beyond reasonable doubt but presumptive treat-
ment should not wait for positive cultures (blood
or liver) and should consist of intravenous broad-­
spectrum antibiotics. Most respond within 24  h
and liver function is usually restored fairly
quickly.
Recalcitrant cholangitis can be a problem in
some children and should be initially treated by
inserting a Hickman line and long-term adminis-
tration of intravenous antibiotics, perhaps with
cyclical oral antibiotics. Sometimes, cholangitis
can be associated with parenchymal cyst or
dilated duct formation and some studies have
advocated percutaneous aspiration or internal
drainage to try and improve bile drainage. A brief
course of high-dose steroids has also been advo-
cated by some.
Occasionally, cholangitis occurs as a late
event in otherwise normal children or adoles-
cents. A possible cause in these is a partial
obstruction of their Roux loops, typically where
it goes through the mesocolon, leading to bile
stasis. A combination of radio-isotope scans and
percutaneous cholangiography or even Roux
loop enteroscopy may aid the diagnostic process
and subsequent Roux loop revision is usually
straightforward and successful.
79.8 Portal Hypertension
Portal hypertension has been shown in virtually
all infants at the time of the Kasai operation.
However, subsequent significant portal hyperten-
sion depends both on the degree of established
fibrosis and most importantly the response to sur-
gery. There is a relationship with biochemical
liver function and variceal development. Thus, in
those who fail and need early transplantation,
about 30–40% will have had a significant variceal
bleed. In those who respond well to initial Kasai
but who perhaps have already established fibro-
sis, then variceal development can be delayed. In
these, bleeding may be deferred until 2–4 years
of age.
Infants and children with bleeding oesopha-
geal varices need rapid access to a high-quality
H. Thompson and M. Davenport
paediatric facility with the resources and experi-
ence to manage them. Injection/banding is not a
technique for the occasional endoscopist.
Restoration of circulating blood volume and
pharmacotherapy (e.g. octreotide) should pre-
cede endoscopy and achieve a measure of stabili-
sation. Sometimes, a modified Sengstaken tube
needs to be placed to achieve haemostasis.
Invariably in children, this can only be done
under general anaesthesia but can be lifesaving.
Older children are suited to endoscopic variceal
banding, although injection sclerotherapy retains
a role in treating varices in infants.
Liver transplantation needs to be actively con-
sidered as definitive treatment for portal hyper-
tension where liver function is poor and the child
is already significantly jaundiced. A few centres
will consider portosystemic shunts for those
where the varices become difficult to manage
endoscopically but who have good liver function,
although this is certainly not common.
Splenomegaly is invariably found with portal
hypertension (at least in those with spleens!).
This may cause hypersplenism (thrombocytope-
nia, anaemia and leukopenia) but of itself does
not usually require specific intervention. Some
centres in Japan advocate partial splenic emboli-
zation but the evidence for real benefit is sparse.
79.8.1 Miscellaneous
Less common complications include the devel-
opment of the hepatopulmonary syndrome where
there is chronic hypoxia secondary to the pres-
ence of abnormal pulmonary arteriovenous com-
munications. This is related to established liver
disease and is particularly prevalent in those with
BASM.  It can also be a cause of sudden death,
and if diagnosed, active consideration for liver
transplantation should be given.
79.9 Outcome
The outcome of BA varies significantly across
the world. The best results, in terms of clearance
of jaundice and overall native liver survival, are
79  Biliary Atresia
100
Post KPE
(1999-2020)
Native Liver Survival (%)
80
60
40
20
0
0 5 10
Age (years)
Fig. 79.7  Native Liver Survival in England and Wales
(1999–2020) (n = 831). The 5- and 10-year estimates are
51.3% (95% CI 54.8–47.8) and 46.5% (95% CI
50.1–42.9)
undoubtedly from Japan, but our own results
from the UK and those of Scandinavia come a
close second. Currently, our national results from
England and Wales show a clearance of jaundice
of 65% in those coming to KPE at a median age
of ~50 days. Realistically, centres should at least
be able to achieve a clearance rate of >50%. The
overall native liver survival is more variable
reflecting the attitudes to and availability of trans-
plantation. Our current national estimate is a 5-
and 10- year native liver survival of 45–50%
(Fig. 79.7). Those who do reach adolescence are
still not cured and in truth this is probably not
possible in the vast majority. This cohort of long-­
term survivors still carries significant morbidity
from underlying liver injury and its sequela
including portal hypertension and recurrent chol-
angitis. The potential need for transplantation
should prompt life-long follow-up and certainly
requires a considered and thorough transition to
adult services to further improve their outcomes.
79.10 Conclusion
Biliary atresia remains a rare condition but one
which carries significant morbidity and poten-
tially mortality—even the 5-year death rate
approaches 10% in the best centres. Recent
1099
advances in our understanding of aetiology with
the realisation that this is fundamentally a disease
present at the time of birth and therefore congeni-
tal may help improve attitudes to screening and
time to KPE and may augment more effective
adjuvant therapies. It remains essential that all
those involved in the care of new-born infants are
aware of biliary atresia despite its rarity as miss-
ing these infants and late referrals are the sources
15 20
of much morbidity and potential mortality. Some
countries, particularly Taiwan and Japan, advo-
cate screening programmes based on stool colour
charts. Others, particularly some states in the
USA, are working towards a more blood-based
biochemical discrimination programme. Actual
evidence of benefit and real improvements in
time-to-KPE has yet to be shown though.
Further Reading
Aldeiri B, Giamouris V, Pushparajah K, Miller O, Baker
A, Davenport M (2020, July 20) Cardiac-associated
biliary atresia (CABA): a prognostic subgroup. Arch
dis child. 106(1):68–72. https://doi.org/10.1136/
archdischild-­2020-­319122. Online ahead of print
Burns J, Davenport M (2020) Adjuvant treatments for
biliary atresia. Transl Pediatr 9:253–265. https://doi.
org/10.21037/tp.2016.10.08
Davenport M, Tizzard S, Underhill J, Mieli-Vergani G,
Portmann B, Hadzic N (2006) The biliary atresia
splenic malformation syndrome: a 28-year single-­
center retrospective study. J Pediatr 149:393–400
Harpavat S, Finegold MJ, Karpen SJ (2011) Patients with
biliary atresia have elevated direct/conjugated biliru-
bin levels shortly after birth. Pediatrics 128(6):e1428–
e1433. https://doi.org/10.1542/peds.2011-­1869
Harpavat S, Garcia-Prats JA, Anaya C et  al (2020)
Diagnostic yield of newborn screening for biliary atre-
sia using direct or conjugated bilirubin measurements.
JAMA 323(12):1141–1150. https://doi.org/10.1001/
jama.2020.0837
Parolini F, Hadzic N, Davenport M (2019) Adjuvant
therapy of cytomegalovirus IgM + ve associated bili-
ary atresia: prima facie evidence of effect. J Pediatr
Surg 54:1941–1945. https://doi.org/10.1016/j.
jpedsurg.2018.12.014
Scottoni F, Davenport M (2020) Biliary atresia: poten-
tial for a new decade. Semin Pediatr Surg 29:150940.
https://doi.org/10.1016/j.sempedsurg.2020.150940
Epub 2020 Jul 24
 Choledochal Cyst
Hiroyuki Koga and Atsuyuki Yamataka
80.1 Introduction
Choledochal cyst, or common bile duct dilata-
tion, was first reported by Douglas in 1852. The
condition is relatively rare in Western popula-
tions with an estimated incidence of 1 in 130,000–
150,000 (Lu 1996; Iwai 2018). However, it is
much more common in the Far East, with rates as
high as 1 per 10,000 described in Japan (Sato
et al. 2001).
The pathologic features of this congenital
anomaly usually include an anomalous junction
of the pancreatic and common bile ducts (pancre-
aticobiliary malunion (PBMU)) that creates a
long common channel and an association with
H. Koga · A. Yamataka (*)
Department of Pediatric and Urogenital Surgery,
Juntendo University School of Medicine,
Bunkyo-ku Tokyo, Japan
e-mail: h-koga@juntendo.ac.jp;
yama@juntendo.ac.jp
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_80
80
intrahepatic bile duct (IHBD) dilatation with or
without downstream stenosis; hepatic fibrosis is
not frequent.
Choledochal cysts are usually classified into
three groups, based on configuration, cystic-
type, fusiform-type, and forme fruste-type.
However, other forms and subgroups have been
described based on the cholangiographic appear-
ance of IHBD or PBMU (Fig. 80.1). The major-
ity of choledochal cysts are associated with
PBMU and only occur extremely rarely without
PBMU.
This chapter will focus primarily on cystic-­
type, fusiform-type, and forme fruste-type chole-
dochal cysts associated with PBMU.
1101
1102
a b
d e
Fig. 80.1  Classification of choledochal cysts with
PBMU. (a) Cystic dilatation of the extrahepatic bile duct.
(b) Fusiform dilatation of the extrahepatic bile duct. (c)
Forme fruste-type choledochal cyst. Without PBMU. (d)
Cystic diverticulum of the common bile duct. (e)
80.2 Etiopathogenesis
The etiology remains unknown. Two theories
have been proposed for the etiology of chole-
dochal cyst; congenital weakness of the wall of
the common bile duct due to a primary abnormal-
ity of proliferation during embryologic ductal
development and obstruction distal to the area of
weakness (Jona et al. 1979; Miyano et al. 1980;
Miyano et  al. 1979; Miyano et  al. 1981; Spitz
1977; Suda et  al. 1983, 1991; Wong and Lister
H. Koga and A. Yamataka
c
f
Choledochocele (diverticulum of the distal common bile
duct) (f) Intrahepatic bile duct dilatation alone (Caroli’s
disease). (From Yamataka A, Kato Y, in Puri P., Hollwarth
M. (eds) Pediatric Surgery, Diagnosis and Management,
Springer, p546; Fig. 72.1, 2009 Same figure)
1981). Unequal proliferation of epithelial cells of
primitive bile ducts when they are still solid has
also been postulated. If the cellular proliferation
of the distal portion of the duct is less active,
canalization will result in an abnormally dilated
proximal end. Evidence for an obstructive factor
being involved is based on an experimental study
in which cystic dilatation of the common bile
duct was induced by ligation of the common bile
duct in neonatal lambs (Spitz 1977), but not at
later stages of development.
80  Choledochal Cyst
In 1969, the “long common channel” theory
was proposed (Babbitt 1969) as a new hypothesis
for the etiology of choledochal cyst based on the
reflux of activated pancreatic enzymes into the
common bile duct because of PBMU causing dis-
solution and weakening of the ductal wall. High
levels of amylase in fluid aspirated from chole-
dochal cysts have been used as evidence to sup-
port this theory, and obstruction of the common
bile duct at the distal end of a cyst has been
explained as being a consequence of edema or
fibrosis caused by refluxed pancreatic fluid.
While the “long common channel” theory
appears to explain the etiology of choledochal
cyst well because almost all choledochal cyst
patients have PBMU, we doubt that weakness of
the choledochal wall due to reflux of pancreatic
fluid is the most basic causative factor of cystic
dilatation. We conducted research in puppies
(Miyano et al. 1981) that showed that the chemi-
cal reactivity of refluxed pancreatic fluid in the
bile duct was less caustic than assumed from
active pancreatic enzyme levels, and the general
observation that there are PBMU patients with
elevated gallbladder amylase levels with no signs
of dilatation of the common bile duct (Tanaka
et al. 1993). Incidentally, choledochal cyst can be
diagnosed prenatally as early as the fifth month
of gestation (Bancroft et al. 1994; Gallivan et al.
1996; Schroeder et al. 1989), at which time, the
fetal pancreas has not developed enough to pro-
duce actively functional enzymes, so the exact
role of pancreatic fluid reflux is essentially
unclear. Research on human embryology has also
demonstrated that the pancreaticobiliary ductal
junction actually lies outside the duodenal wall
until the eighth week of gestation, when it
migrates inward toward the duodenal lumen;
thus, PBMU may in fact be related to some arrest
of this migration.
Based on available resources, we feel that an
anomalous pancreaticobiliary ductal junction
combined with stenosis are the most likely caus-
ative factors of choledochal cyst rather than
weakness caused by reflux of pancreatic fluid
with activated pancreatic enzymes, at least in
perinatal and young infants because of its con-
genital nature. Both PBMU and stenosis are also
1103
associated with abnormal development of the
ventral pancreatic duct and biliary duct system.
80.3 Pathophysiology
80.3.1 Cystic/Fusiform-Type
Choledochal Cysts
Erosion, epithelial desquamation, and dysplasia
in the bile duct mucosa without carcinoma are
frequently found with metaplastic changes such
as mucous cells, goblet cells, and Paneth’s cells.
Such hyperplastic and metaplastic epithelia have
been observed to increase with age, especially in
patients older than 14 years of age, and progress
to dysplastic epithelia in adults that can undergo
malignant transformation, giving rise to bile duct
carcinoma. Gallbladder mucosa shows cholecys-
titis, cholesterolosis, adenomyosis or adenomyo-
matosis, polyp and adenoma formation, and
epithelial hyperplasia, which are particularly
characteristic of PBMU.
80.3.2 Forme Fruste-Type
Choledochal Cysts
Bile duct mucosa shows nonspecific changes
such as mucosal ulceration/sloughing, fibrosis,
and inflammatory cell infiltration, indicating that
children with forme fruste-type may be at a high
risk for carcinogenesis of the extrahepatic bile
duct. The process by which epithelial changes
and malignant transformation may occur is the
same as for cystic/fusiform-type choledochal
cysts. Gallbladder mucosa shows diffuse epithe-
lial hyperplasia characterized with or without
metaplasia of pyloric glands, goblet cells, and
Paneth’s cells.
80.4 Diagnosis
80.4.1 Prenatal Diagnosis
Abdominal ultrasonography is probably the best
screening investigation for choledochal cyst. A
1104
typical finding on high-resolution ultrasonogra-
phy is that of a cyst at the porta hepatitis. The
sensitivity and specificity of prenatal ultrasonog-
raphy for diagnosing choledochal cyst is greatly
increased when performed by a skilled and expe-
rienced sonographer. As a result, in recent years,
the number of patients being diagnosed prena-
tally is increasing.
80.4.2 Clinical Features
Choledochal cyst can present at any age, but
more than half of patients are diagnosed within
the first decade of life. Clinical manifestations of
choledochal cysts differ according to the age of
onset (Tanaka et al. 1993; Davenport et al. 1995).
In young infants up to 3 months old, obstruc-
tive jaundice, acholic stools, and hepatomegaly
are characteristic, depending on the degree of
obstruction. Differentiation from cystic type bili-
ary atresia is crucial and is possible by confirm-
ing the presence of an entire biliary tree with
well-developed IHBD. Another presentation is a
large upper abdominal mass without jaundice.
These patients may rarely also have advanced
liver fibrosis.
In younger children after infancy, there are
two general patterns; jaundice secondary to bili-
ary obstruction in the distal cyst in cystic-type
choledochal cyst and distress from abdominal
pain associated with pancreatitis secondary to
PBMU in fusiform-type or forme fruste-type
choledochal cysts.
In adolescence and adulthood, a choledochal
cyst is often associated with cholelithiasis and/or
biliary carcinoma; there may be liver cirrhosis
with or without portal hypertension at the time of
diagnosis irrespective of whether the patient is
symptomatic.
80.4.3 Imaging Studies
For thorough diagnosis of choledochal cyst, the
presence or absence of dilatation of the extrahe-
patic bile duct, PBMU, and a long common
channel must be identified. Magnetic resonance
H. Koga and A. Yamataka
cholangiopancreatography (MRCP) can provide
excellent visualization of the pancreaticobiliary
ducts in patients with choledochal cyst and
allow detection of narrowing, dilatation, and
filling defects with medium to high degrees of
accuracy (Fig.  80.2). Endoscopic retrograde
cholangiopancreatography (ERCP) accurately
visualizes the configuration of the pancreatico-
biliary ductal system in detail and is valuable
especially in cases where fine detail is required
for preoperative planning. However, it is inva-
sive and may induce post-procedural pancreati-
tis, so it is rarely performed nowadays in
children. In fact, MRCP has replaced ERCP at
most centers, even for adult patients. MRCP is
noninvasive and is an alternative to ERCP for
the evaluation of anatomic anomalies of the
pancreaticobiliary tract with the additional
advantage that the pancreatic duct upstream to
an obstruction or an area of stenosis can also be
visualized.
If preoperative imaging can allow clear visual-
ization of the entire biliopancreatic ductal sys-
tem, including the intra- and extrahepatic bile
ducts, and pancreatic duct in detail, intraopera-
tive cholangiography is unnecessary; however, if
information about the pancreaticobiliary tree is
Fig. 80.2  Magnetic resonance cholangiopancreatogra-
phy (MRCP) in a patient with congenital biliary dilata-
tion. Fusiform dilatation of the extrahepatic bile duct,
long common channel (between arrows), protein plugs
(arrowheads), and pancreatic duct can be identified
80  Choledochal Cyst 1105

still considered lacking, for example, with respect

to the presence or absence of PBMU and debris/
protein plugs/stones, it must be performed. If a
cyst is too large, intraoperative cholangiography
via the gallbladder or directly via the common
bile duct is useless because contrast material in
the cyst will cover the proximal pancreatic duct,
preventing PBMU and the common channel from
being visualized. In such cases, intraoperative
cholangiography should be performed separately
for the IHBD and distal common bile duct by a
selective technique after separating the cyst into
proximal and distal segments.

80.5 Differential Diagnosis

In the prenatal period, an intra-abdominal cystic

lesion on high-resolution ultrasonography may
be confused with duodenal atresia, cystic type
biliary atresia, ovarian cyst, duplication cyst, or
mesenteric cyst. The differential diagnosis after
birth includes congenital hepatic fibrosis, Caroli’s
disease with hepatic fibrosis, primary sclerosing
cholangitis, cholangitis due to choledocholithia-
sis, congenital cirrhosis, and congenital stricture
of the common hepatic or common bile duct.
However, a combination of standard imaging
studies can distinguish between these conditions
readily.
80.6 Open Surgical Management
80.6.1 Cyst Excision
Cyst excision is the definitive treatment of choice
for choledochal cyst because of the high
­morbidity and high risk of carcinoma after inter-
nal drainage such as cystoduodenostomy, a com-
monly used procedure in the past. Recently, more
attention has been paid to treatment of intrahe-
patic and intrapancreatic ductal disease such as
IHBD dilatation with downstream stenosis,
debris in the IHBD, and protein plugs or stones in
the common channel. The level of excision of the
common hepatic duct within the pancreas is also
important.
Fig. 80.3  Diagram of the transection of a choledochal
cyst. (From Yamataka A., Kato Y., in Puri P., Hollwarth M.
(eds) Pediatric Surgery, Diagnosis and Management,
Springer, p549; Fig. 72.3, 2009 Same figure)
Prior to commencing specific dissection of a
choledochal cyst, we prefer to incise the anterior
wall of the cyst transversely to open it and allow
the posterior wall of the cyst to be visible directly
from the inside (Fig. 80.3). The choledochal cyst
can be freed from surrounding tissues including
the portal vein and hepatic artery more easily and
safely than by dissecting the cyst free without
incising the anterior wall.
There are usually more adhesions between a
cystic-type choledochal cyst and surrounding
vital structures, such as the portal vein and
hepatic artery in older children, compared with a
fusiform-type choledochal cyst in younger chil-
dren, or a cystic-type choledochal cyst in infants.
If a cyst is extremely inflamed and adhesions are
very dense, mucosectomy of the cyst should be
performed rather than full-thickness dissection to
minimize injury to surrounding structures, such
as the portal vein and hepatic artery (Fig. 80.4).
80.6.2 Intraoperative Endoscopy
Inspection of the biliary tree and pancreatic duct
is valuable for detecting debris that could give
1106 H. Koga and A. Yamataka

Fig. 80.5  Massive debris in the intrahepatic bile duct

observed during intraoperative endoscopy using a pediat-
ric cystoscope

intraoperative endoscopy of the common channel

and IHBD because the side channels of flexible
endoscopes are essentially designed only for flush-
Fig. 80.4  Excision of the distal portion of a choledochal ing and are inadequate for inspection and irriga-
cyst. Arrows indicate the plane for mucosectomy or full-­ tion and laparoscopes do not have side channels.
thickness. (From Yamataka A., Kato Y., in Puri P.,
Hollwarth M. (eds) Pediatric Surgery, Diagnosis and The tip of the scope is inserted into the com-
Management, Springer, p550; Fig. 72.5, 2009 Same mon channel through the distal cyst and into the
figure) IHBD through the common hepatic duct. From
experience, we have found debris in the IHBD or
rise to stones in the future, and irrigation to the common channel, even when no debris was
remove debris is simple and effective (Fig. 80.5). detected on preoperative imaging studies. Debris
We commenced performing intraoperative endos- is clearly more common when IHBD are dilated.
copy of the common channel and IHBD during Intraoperative endoscopy is particularly valuable
cyst excision routinely in 1986 with extremely during the excision of fusiform-type choledochal
effective reduction in postoperative complica- cyst to ensure that any wide intrapancreatic cho-
tions documented on long- and ultra-long-term ledochal cyst is excised adequately, as any rem-
follow-up (Miyano et  al. 1995; Takahashi et  al. nant may contribute to stone formation that may
2010). cause postoperative pancreatitis in the long term.
For intraoperative endoscopy, we use a fine It is less important in cystic-type choledochal
pediatric cystoscope or ureteroscope because they cyst, since the intrapancreatic choledochus is
have a side channel that allows normal saline to short, very narrow, and patients do not usually
flow continuously to dilate ducts, allowing safe present with pancreatitis, because there is usually
visualization/inspection and irrigation for clearing no debris in the common channel.
debris and protein plugs. Without a constant flow Intraoperative endoscopy should be performed
of saline, duct lumens will collapse, greatly com- routinely on all choledochal cyst patients, unless
promising both examination and clearing. Thus, the scope cannot be inserted into the common
despite being used by some surgeons, flexible channel from the distal part of the cyst because it
endoscopes and laparoscopes are not effective for is too narrow.
80  Choledochal Cyst
80.6.3 Excision of the Distal Common
Bile Duct
The distal common bile duct should be resected
as close as possible to the PBMU to prevent
postoperative pancreatitis and/or stone forma-
tion due to residual intrapancreatic choledochal
cyst. Before the introduction of intraoperative
endoscopy, complete and safe excision of the
distal common bile duct within the pancreas
proximal to the PBMU was more difficult in
fusiform-type choledochal cyst patients because
of risk for injury to the pancreatic duct associ-
ated with the distal common bile duct still being
wide at the PBMU and high likelihood that the
residual intrapancreatic portion of the distal
common bile duct would be excessive, if not
excised properly (Fig.  80.6). Thus, if the distal
common bile duct is resected along the red line,
stones may form over time in the residual intra-
pancreatic duct leading to recurrent pancreatitis
or malignant changes, and if resected along the
blue line, that is, just above the PBMU, postop-
erative pancreatitis and stone formation can be
prevented.
80.6.4 Excision of the Common
Hepatic Duct
The common hepatic duct is usually transected at
the level of distinct caliber change. Because
residual choledochal cyst mucosa has been impli-
cated in malignant changes, care must be taken to
excise it completely, especially in older children.
However, if the lumen of the common hepatic
duct at the hepaticojejunostomy anastomosis is
4 mm or less in younger patients with cystic-type
choledochal cyst, we create a Carrel patch with
2–3  mm of choledochal cyst to increase the
lumen to 6–7 mm as an option because a larger
lumen is easier to anastomose technically, and a
study we performed on choledochal cyst histopa-
thology in younger patients followed-up for some
20  years found no evidence of premalignant
transformation in excised specimens and no
malignancies.
1107
80.6.5 IHBD Dilatation
Dilatation of the peripheral portion of the IHBD
in patients with choledochal cyst may be associ-
ated with late complications such as recurrent
cholangitis, stone formation, and anastomosis
stricture. The management of IHBD dilatation is
difficult and can be managed by segmentectomy
of the liver, intrahepatic cystoenterostomy, or
balloon dilatation of the stenotic lesion causing
IHBD dilatation at the time of cyst excision.
However, the actual incidence of the late compli-
cations mentioned appears to be low, especially
in younger children, so specific surgical interven-
tion for IHBD dilatation should be reserved only
for cases with massive dilatation of peripheral
IHBD with severe downstream stenosis. Careful
follow-up of all cases with IHBD dilatation is
mandatory.
80.6.6 Roux-En-Y
Hepaticojejunostomy
End-to-end anastomosis during Roux-en-Y
hepaticojejunostomy is recommended to prevent
elongation of the blind pouch if the ratio between
the diameters of the common hepatic duct and the
proximal Roux-en-Y jejunum at the proposed
site of anastomosis is less than or equal to 1
(common hepatic duct): 2.5 (jejunum) (Fig. 80.7).
If end-to-side anastomosis is unavoidable, the
common hepatic duct should be anastomosed as
close as possible to the closed end of the blind
pouch; if an end-to-side anastomosis is per-
formed far from the closed end of the blind
pouch, elongation of the blind pouch will occur
later in life as the child grows causing bile stasis
in the blind pouch and IHBD (especially if they
are dilated) leading to stone formation (Fig. 80.8).
Some surgeons predetermine the length of the
Roux-en-Y jejunal limb (e.g., 30, 40, 50, or
60 cm) in an attempt to take the size of a child
into consideration, but this can result in the Roux-­
en-­Y jejunal limb being unnecessarily long in
infants and younger children; redundancy and
tortuosity of the Roux-en-Y jejunal limb are
1108
Fig. 80.6  Diagram of intraoperative endoscopy of the
bile duct distal to a cyst with debris and protein plugs. If
the distal common bile duct is resected along the red line,
stones may form over time in the residual intrapancreatic
duct leading to recurrent pancreatitis or malignant
changes, and if resected along the blue line, that is, just
H. Koga and A. Yamataka
Wrong
recommended
above the PBMU, postoperative pancreatitis and stone
formation can be prevented. (From Yamataka A., Kato Y.,
in Puri P., Hollwarth M. (eds) Pediatric Surgery,
Diagnosis and Management, Springer, p551; Fig. 72.6–
72.9, 2009 Same figure)
80  Choledochal Cyst
Fig. 80.7  Adequate Roux-en-Y (RY) hepaticojejunos-
tomy at the time of cyst excision Arrowheads indicate
approximation of the native jejunum and distal RY limb.
Arrows indicate the smooth flow of small bowel contents
without reflux. (From Yamataka A., Kato Y., in Puri P.,
Hollwarth M. (eds) Pediatric Surgery, Diagnosis and
Management, Springer, p553; Fig. 72.12, 2009 Same
figure)
likely to occur later in life as the patient grows,
contributing to cholestasis. Briefly, our technique
for customizing the length of the Roux-en-Y
limb involves identifying the ligament of Treitz
and dividing the jejunum 15 cm distal to the liga-
ment to create a Roux-en-Y jejunal loop. The
Roux limb is brought up to some 3 cm above the
xiphoid process on the anterior abdominal wall
so the Roux-en-Y jejunojejunostomy fits natu-
rally into the splenic flexure after the jejunojeju-
nostomy anastomosis is completed. By doing
this, there will be no redundancy of the Roux-­
en-­Y limb.
When a jejunojejunostomy and Roux-en-Y
limb are used, we recommend that the native
jejunum and the Roux-en-Y jejunal limb be
approximated for up to 5–8 cm cranially from the
jejunojejunostomy to ensure bile in the Roux-­
en-­Y limb and contents of the native jejunum
1109
Fig. 80.8  Inadequate Roux-en-Y (RY) hepaticojejunos-
tomy (HJ) at the time of cyst excision. Note: HJ far from
the closed end of the blind pouch (arrowhead). Double
arrows in the inset indicate elongation of the blind pouch.
The arrow with an asterisk indicates reflux of jejunal con-
tents into the RY limb through a T-shaped RY jejunojeju-
nostomy. (From Yamataka A., Kato Y., in Puri P., Hollwarth
M. (eds) Pediatric Surgery, Diagnosis and Management,
Springer, p554; Fig. 72.13, 2009 Same figure)
flow smoothly into the jejunum distal to the jeju-
nojejunostomy. Without performing this, the
jejunojejunostomy tends to be T-shaped, and
there may be reflux of jejunal contents into the
Roux-en-Y limb leading to dilatation of the jeju-
nal limb and stasis in the Roux-en-Y limb, a situ-
ation we encountered in a patient who had surgery
elsewhere.
80.6.7 Hepaticojejunostomy Versus
Hepaticoduodenostomy
Although hepaticoduodenostomy is an easier,
quicker procedure that allows bile to enter the
duodenum directly, which is more physiological,
postoperative cholangitis and bile gastritis are
1110
known complications with risk for mucosal dam-
age and possible malignant change. Todani et al.
(2002) reported a patient who underwent cyst
excision and hepaticoduodenostomy at 13 months
old and developed hilar bile duct carcinoma
18  years later. Inflammation of the bile duct
mucosa was thought to be related to the reflux of
duodenal contents (including activated pancre-
atic enzymes) into IHBD through the anastomo-
sis which prompted them to abandon
hepaticoduodenostomy in favor of hepaticojeju-
nostomy. We also reported that hepaticoduode-
nostomy caused bilious gastritis due to marked
duodenogastric bile reflux on upper gastrointesti-
nal endoscopy, and histology of biopsied gastric
mucosa showed gastritis (Shimotakahara et  al.
2005).
Overall, hepaticojejunostomy is recom-
mended for biliary reconstruction in children
requiring choledochal cyst excision, because
hepaticoduodenostomy is associated with some
degree of reflux of duodenal contents into the
biliary tree, especially when IHBD dilatation is
present. A long-term prospective randomized
controlled study is warranted to compare the out-
comes of laparoscopic hepaticoduodenostomy
and hepaticojejunostomy.
80.7 Minimally Invasive Surgical
Management
80.7.1 Laparoscopic Surgery
Minimally invasive surgery is indicated for cho-
ledochal cyst and is a valid option for treatment
where available. Although technically more
­challenging than conventional open surgery, the
general concepts are the same. The patient is
positioned at the foot of the operating table and
the table is tilted 15° or to 30° if necessary, to be
in a reverse Trendelenburg position. The surgical
team is initially positioned with the surgeon at
the patient’s feet, an assistant with a scope on the
surgeon’s left, and another assistant on the sur-
geon’s right, with the monitor placed towards the
patient’s head. We use conventionally placed tro-
cars (right upper quadrant, left paraumbilical, left
H. Koga and A. Yamataka
Fig. 80.9  Trocar positions for laparoscopic cyst excision.
A 30° 5 mm or 10 mm laparoscope has been introduced
through a GelPOINT® mini access platform (Applied
Medical, Rancho Santa Margarita, USA) inserted in a
2  cm umbilical incision. Additional 5  mm trocars have
been inserted in the right upper quadrant and left upper
quadrants, respectively as working trocars. The left upper
quadrant trocar (left subcostal trocar in the anterior axil-
lary line) is for exposing the porta hepatis. An additional
3.9  mm trocar has been placed in the epigastrium for
intraoperative endoscopy. Another additional 3.9 mm tro-
car (asterisk) may be placed in the lateral right subcostal
area for an assistant to grasp the cyst to facilitate safe dis-
section of the cyst by the operating surgeon
upper quadrant; scope in the umbilicus) to free
the cyst and transect it at mid-level (Fig. 80.9).
Intraoperative endoscopy is performed
through an additional 3.9  mm trocar in the epi-
gastrium for inserting a fine ureteroscope in the
same way as described earlier in intraoperative
endoscopy during open cyst excision except that
the tip of the uteroscope is inserted into the com-
mon channel through the distal cyst under laparo-
scope guidance to remove debris. The exact level
of transection of the distal common bile duct is
also determined using intraoperative endoscopy,
if the orifice of the pancreatic duct in the com-
mon channel can be identified. After the cyst is
freed, the distal part is divided as close as possi-
ble to the pancreaticobiliary junction and the
stump is ligated with an Endoloop (Ethicon,
Somerville, New Jersey, USA). If intraoperative
endoscopy cannot be performed because the
opening of the intrapancreatic choledochus and
common channel are too small, commonly
80  Choledochal Cyst 1111

observed in cystic-type choledochal cyst cases, instrument grip dexterity, tremor reduction,
intraoperative cholangiography may be per- motion scaling, and degrees of freedom are
formed by placing an endoscopic metal clip at the improved compared with standard laparoscopic
distal end of the dissected cyst to confirm the instruments. These advantages improve dissect-
required extent of further dissection distally since ing, suturing, and knot tying; thus, the hepatico-
the clip and the confluence between the common jejunostomy anastomosis is far easier than with
channel, intrapancreatic choledochus, and pan- conventional laparoscopic surgery. However, at
creatic duct can be visualized. If dissection is present, it is not widely accepted for use in pedi-
inadequate, the cyst can be further dissected dis- atric patients because currently available robotic
tally and intraoperative cholangiography repeated hardware lacks the diversity that is required to
as above, until cyst dissection is adequate. treat children of various sizes. From experience,
The proximal cyst is excised leaving 5 mm of robotic surgery is less suitable for dissecting tis-
the common hepatic bile duct. Another two tro- sue in children than conventional laparoscopy
cars are added for the hepaticojejunostomy; lat- because of space constraints.
eral right subcostal and between the lateral right With ongoing technical improvements, robotic
subcostal and right upper quadrant trocars. To surgery will no doubt be the key to achieving the
customize the length of the Roux loop during ultimate goals of minimally invasive surgery, i.e.,
laparoscopic repair, pneumoperitoneum is inter- painless, scarless treatment with negligible
rupted, the jejunum 15 cm distal to the ligament downtime, and be particularly advantageous for
is exteriorized through the umbilical port site, small diameter anastomoses.
and the Roux-en-Y jejunal loop is created extra-
corporeally according to the same technique
mentioned earlier. The hepaticojejunostomy 80.7.3 Hybrid Surgery
anastomosis is performed using interrupted 5/0
or 6/0 absorbable sutures with the right upper While laparoscopy provides magnified images,
quadrant port as a needle holder in the right allowing dissection and surgical maneuvers to be
hand, the 5 mm port for the scope, and the 3 mm performed precisely with reduced tissue injury,
subcostal port as a needle receiver in the left and better cosmesis, rigid instrumentation with
hand. Both the right and left edge sutures are limited degrees of freedom, fulcrum effects, and
exteriorized and used as traction sutures during two-dimensional vision lead to poor ergonomics
anastomosis of the anterior wall to facilitate for surgeons, thus hindering the accurate and safe
accuracy. From experience, if a hepaticojejunos- placement of sutures, especially during anasto-
tomy is performed without extra trocars, the moses. Robotic surgery improves fine maneuvers
quality of the anastomosis deteriorates, espe- and combined, a hybrid of robotic and laparo-
cially when the lumen diameters are 9  mm or scopic approaches could allow surgery for chole-
less. dochal cyst to be performed optimally using the
best approach for each maneuver; i.e., cyst exci-
sion using laparoscopy and the hepaticojejunos-
80.7.2 Robotic Surgery tomy anastomosis using robotic surgery.
Additional benefits of a hybrid approach would
Robotic surgery was first used for treating chole- be to take advantage of the diversity provided by
dochal cyst in 2006 (Woo et  al. 2006). Robotic laparoscopy because of a larger range and variety
surgery overcomes limitations encountered dur- of instruments that contribute to smoother prepa-
ing laparoscopic surgery by enhancing visualiza- ration of bowel for anastomosis, and the relative
tion and instrument control. The inbuilt camera freedom created by switching to optimize maneu-
system provides high magnification with three-­ verability and visibility, both of which are not
dimensional stereo visualization that the operat- possible with robotic surgery once docking has
ing surgeon can control directly as required, and been completed.
1112
a b
Fig. 80.10 Robotic hepaticojejunostomy. (a) Trocar
positions for conventional laparoscopy. (b) Changing
from laparoscopic to da Vinci trocars. The left trocar has
As an example of the potential of hybrid sur-
gery, we recently treated a child with PBMU with-
out dilatation of the common hepatic duct
(Fig. 80.10) using robotic surgery for the hepatico-
jejunostomy anastomosis and laparoscopy for the
rest of the procedure. The diameter of the common
hepatic duct was 4 mm, a technically challenging
size even in the most experienced hands, and
robotic hepaticojejunostomy anastomosis was
easier than expected and saved struggling to per-
form the same anastomosis laparoscopically.
Robotic surgery would seem to be well suited
for performing tedious procedures such as small
diameter anastomoses successfully with less
stress on the operating surgeon and probably con-
sistently superior results compared with laparos-
copy or open surgery. Hybrid surgery will
ultimately be limited only by the skill and cre-
ativity of operating surgeons.
H. Koga and A. Yamataka
c
been replaced and the right trocar is being replaced. (c)
Docking of the da Vinci cart to the trocars. (d) Robotic
hepaticojejunostomy
80.8 Postoperative Outcome
and Complications
A satisfactory surgical outcome with low mor-
bidity is expected after open or minimally inva-
sive cyst excision in the short- to mid-term.
However, with longer follow-up, reports of com-
plications increase and careful extended follow-
up should be considered mandatory for all
­
postoperative choledochal cyst patients until a
complication finally develops.
From an American survey of 198 choledochal
cyst patients in 1981 (Kim 1981), common late
complications were cholangitis, obstructive jaun-
dice, pancreatitis, stone formation, and portal
hypertension; 36 patients were lost to follow-up;
115 patients were alive without liver disease;
causes of death in 14 were: biliary atresia, chol-
angitis with sepsis, hepatic failure, and carci-
80  Choledochal Cyst
noma. Malignancy can arise from the
intrapancreatic terminal choledochus, the hepati-
cojejunostomy anastomosis site, and the
IHBD.  In a series of 40 adult patients who had
cyst excision at the age of 16 years or more, two
died from cholangiocarcinoma; one had a pri-
mary cyst excision at the age of 25 years.
We reviewed 286 children who had open cyst
excision for choledochal cyst performed at the
age of 15 years or less between 1964 and 2019.
Mean age at initial onset of symptoms including
abdominal pain and jaundice was 5 years or less
(n = 231) and between 6 and 15 years (n = 55);
overall mean age at onset was 3.2  years. Mean
age at cyst excision was 4.3 years. Mean follow-
­up was 28.4  years. Treatment was primary cyst
excision (n = 252), cyst excision converted from
intestinal drainage (n = 5), and cyst excision after
other interventions such as percutaneous transhe-
patic cholangiodrainage, T-tube drainage, and
cholecystectomy (n  =  29). Surgical procedures
performed were Roux-en-Y hepaticojejunostomy
(n  =  266), standard hepaticoduodenostomy
(n  =  19), and jejunal interposition hepaticojeju-
nostomy (n = 1). 156 children had intraoperative
endoscopy performed. There was no operative
mortality. There were 25 complications such as
ascending cholangitis, intrapancreatic terminal
choledochus calculi, pancreatitis, and bowel
obstruction that arose in 20 children (6.9%) and
15 required surgical intervention (revision of
hepatic enterostomy, percutaneous transhepatic
cholangioscopic lithotomy, excision of residual
intrapancreatic terminal choledochus, endo-
scopic sphincterotomy, pancreaticojejunostomy,
or laparotomy for bowel obstruction). In patients
who underwent cyst excision at the age of 5 years
or less, there were no major complications such
as intrahepatic stone formation, intrapancreatic
terminal choledochus calculi, and anastomotic
stricture of the hepaticoenterostomy. None of the
156 children who had intraoperative endoscopy
developed stones, anastomotic strictures, or chol-
angitis. No malignancy was identified.
Results of mid- to long-term follow-up of lap-
aroscopic cyst excisions published recently (Qiao
et  al. 2015; Shen et  al. 2015; Zhen et  al. 2015)
indicate that experienced laparoscopic surgeons
1113
can achieve results as good as those for open sur-
gery with no significant differences in bile leak-
age or wound infections. However, while
laparoscopic cyst excision and Roux-en-Y recon-
struction may be safe and effective in the hands
of skilled laparoscopic surgeons, the hepaticoje-
junostomy anastomosis remains extremely chal-
lenging, technically. For robotic and hybrid
surgery, total anastomotic time was shorter than
for a comparable laparoscopic hepaticojejunos-
tomy although time per suture during the anasto-
mosis itself was similar. Interestingly, the
coefficient of variation for the time to taken place
one suture during a robotic hepaticojejunostomy
was shorter than for laparoscopic hepaticojeju-
nostomy (Koga et al. 2019).
We present our experience of 63 laparoscopic
choledochal cyst excisions between 2009 and
2020. Cases requiring conversion to open lapa-
rotomy (n = 2) and mini-laparotomy (n = 2) were
excluded, leaving 59 cases; 47 females and 12
males. Choledochal cysts were fusiform-type in
31 cases and cystic-type in 28 cases. Mean age
(range) at surgery was 4.8 (0.3–14.1) years, and
mean weight at surgery was 16.9 (5.5–47.0) kg.
Ten patients had IHBD dilatation. There were no
intraoperative complications. Estimated mean
blood loss was minimal at
15  mL.  Hepaticojejunostomy diameters were
6–9 mm in 23/31 fusiform cases and 19/28 cystic
cases; more than 10  mm in 8/31 fusiform-type
cases and 9/28 cystic-type cases.
Hepaticojejunostomy was performed laparoscop-
ically in 45 cases and robotically in 14 cases.
Intraoperative endoscopy of both the common
channel and IHBD was performed in 37 cases (28
fusiform-type; 9 cystic-type); the remaining 22
had intraoperative endoscopy of IHBD alone
because the ureteroscope could not be inserted
into the intrapancreatic choledochus and com-
mon channel. Protein plugs were present in the
common channel in all 28 fusiform-type cases
(massive in 7, moderate in 18, minimal in 3); suc-
cessfully cleared by irrigation with normal saline
from the side channel of the ureteroscope. Debris
were present in all 22 cases that had intraopera-
tive endoscopy of the IHBD alone (moderate in
9, minimal in 13). There were no debris in the
1114
IHBD of the 25 who had intraoperative endos-
copy of both the common channel and intrahe-
patic bile ducts. Although all patients are well
after a mean follow-up of 6.6  years (range:
3 months to 11.6 years) with cosmetically esthetic
wounds, there were three postoperative compli-
cations. The first was pancreatitis that developed
8  months postoperatively from newly formed
3x3mm debris because all massive protein plugs
seen on intraoperative endoscopy were cleared
thoroughly by irrigation. Pancreatitis was treated
by conservative medical management and there
have been no relapses. The second was duodenal
obstruction in a cystic-type case. At exploratory
laparoscopy, the third part of the duodenum was
found to be compressed by the Roux-en-Y limb
that had been fixed inadequately to the colonic
mesentery. Sutures between the Roux-en-Y limb
and the colonic mesentery were released laparo-
scopically. Postoperative recovery was unevent-
ful. The third was anastomotic leak treated by
minilaparotomy.
80.9 Conclusion
Cyst excision and Roux-en-Y hepaticojejunos-
tomy are the treatment of choice in both children
and adults with choledochal cyst. Roux-en-Y
hepaticojejunostomy in children is different
from that in adults, since the Roux-en-Y limb or
blind pouch can grow and elongate as the child
grows. Intraoperative endoscopy at the time of
cyst excision is useful to prevent postoperative
complications, especially those that develop
over time.
References
Babbitt DP (1969) Congenital choledochal cysts: new eti-
ological concept based on anomalous relationships of
the common bile duct and pancreatic bulb. Ann Radiol
(Paris) 12:231–240
Bancroft JD, Bucuvalas JC, Ryckman FC et  al (1994)
Antenatal diagnosis of choledochal cyst. J Pediatr
Gastroenterol Nutr 18:142–145
Davenport M, Stringer MD, Howard ER (1995) Biliary
amylase and congenital choledochal dilatation. J
Pediatr Surg 30:474–477
H. Koga and A. Yamataka
Gallivan EK, Crombleholme TM, D'Alton ME (1996)
Early prenatal diagnosis of choledochal cyst. Prenat
Diagn 16:934–937
Iwai N (2018) Choledochal cyst. In: Loaty PD, Flake AW,
Rintaka RJ, Hitson JM (eds) Rickham’s neonatal sur-
gery. Springer-Verlag, London, pp 855–865
Jona JZ, Babbitt DP, Starshak RJ et al (1979) Anatomic
observations and etiologic and surgical con-
siderations in choledochal cyst. J Pediatr Surg
14:315–320
Kim SH (1981) Choledochal cyst: survey by the surgi-
cal section of the American Academy of Pediatrics. J
Pediatr Surg 16(3):402–407
Koga H, Murakami H, Ochi T et  al (2019) Comparison
of robotic versus laparoscopic hepaticojejunostomy
for choledochal cyst in children: a first report. Pediatr
Surg Int 35(12):1421–1425
Lu S (1996) Biliary cysts and strictures. In: Kaplowitz N
(ed) Liver and biliary disease. Williams & Wilkins,
Baltimore, pp 739–753
Miyano T, Suruga K, Suda K (1979) Abnormal
choledocho-­pancreatico ductal junction related to the
etiology of infantile obstructive jaundice diseases. J
Pediatr Surg 14:16–26
Miyano T, Suruga K, Chen SC (1980) A clinicopathologic
study of choledochal cyst. World J Surg 4:231–238
Miyano T, Suruga K, Suda K (1981) “The choledocho-­
pancreatic long common channel disorders” in rela-
tion to the etiology of congenital biliary dilatation
and other biliary tract disease. Ann Acad Med Singap
10:419–426
Miyano T, Yamataka A, Kato Y et al (1995) Choledochal
cysts: special emphasis on the usefulness of intraop-
erative endoscopy. J Pediatr Surg 30:482–484
Qiao G, Li L, Li S et al (2015) Laparoscopic cyst excision
and roux-Y hepaticojejunostomy for children with
choledochal cysts in China: a multicenter study. Surg
Endosc 29:140–144
Sato M, Ishida H, Konno K et  al (2001) Choledochal
cyst due to anomalous pancreatobiliary junction in
the adult: sonographic findings. Abdom Imaging
26:395–400
Schroeder D, Smith L, Prain HC (1989) Antenatal diag-
nosis of choledochal cyst at 15 weeks’ gestation: etio-
logic implications and management. J Pediatr Surg
24:936–938
Shen HJ, Xu M, Zhu HY et al (2015) Laparoscopic ver-
sus open surgery in children with choledochal cysts: a
meta-analysis. Pediatr Surg Int 31:529–534
Shimotakahara A, Yamataka A, Yanai T et al (2005) Roux-­
en-­Y hepaticojejunostomy or hepaticoduodenostomy
for biliary reconstruction during the surgical treatment
of choledochal cyst: which is better? Pediatr Surg Int
21:5–7
Spitz L (1977) Experimental production of cystic dila-
tation of the common bile duct in neonatal lambs. J
Pediatr Surg 12:39–42
Suda K, Miyano T, Konuma I et al (1983) An abnormal
pancreatico-choledocho-ductal junction in cases of
biliary tract carcinoma. Cancer 52:2086–2088
80  Choledochal Cyst
Suda K, Matsumoto Y, Miyano T (1991) Narrow duct seg-
ment distal to choledochal cyst. Am J Gastroenterol
86:1259–1263
Takahashi T, Shimotakahara A, Okazaki T et  al (2010)
Intraoperative endoscopy during choledochal cyst
excision: extended long-term follow-up compared
with recent cases. J Pediatr Surg 45:379–382
Tanaka K, Nishimura A, Yamada K et  al (1993) Cancer
of the gallbladder associated with anomalous junction
of the pancreatobiliary duct system without bile duct
dilatation. Br J Surg 80:622–624
Todani T, Watanabe A, Toki A et al (eds) (2002) Hilar duct
carcinoma developed aſter cyst excision followed by
1115
hepaticoduodenostomy. Pancreaticobiliary maljunc-
tion, Igaku Tosho, Tokyo, pp 17–21
Wong KC, Lister J (1981) Human fetal development of
the hepato-pancreatic duct junction--a possible expla-
nation of congenital dilatation of the biliary tract. J
Pediatr Surg 16:139–145
Woo R, Le D, Albanese CT, Kim SS (2006) Robot-assisted
laparoscopic resection of a type I choledochal cyst in
a child. J Laparoendosc Adv Surg Tech A 16:179–183
Zhen C, Xia Z, Long L et al (2015) Laparoscopic excision
versus open excision for the treatment of choledochal
cysts: a systematic review and meta-analysis. Int Surg
100:115–122
 Hepatic Cysts and Abscesses
Priya Ramachandran
81.1 Hepatic Cysts
81.1.1 Simple Non-Neoplastic Cyst
These cysts are also known as congenital hepatic
cysts. They are uncommon in children and are
often identified as incidental findings during
imaging procedures. Cysts are unilocular with
fine septations and contain serous fluid. Hepatic
cysts do not communicate with bile ducts. Large
cysts produce atrophy of the adjacent hepatic tis-
sue but the vessels and ducts are spared.
Protruding vessels give a convoluted appearance
to the cyst lining. Small cysts do not compress
adjacent hepatic tissue. Intracystic bleeding can
change the morphological appearance of the
cysts. The lining of the cyst wall is composed of
an inner layer of cuboidal or columnar epithe-
lium and an outer wall, which contains a thin
layer of connective tissue and compressed liver
tissue. Most congenital hepatic cysts are asymp-
tomatic. However, large cysts have been reported
to cause respiratory distress and hydroureterone-
phrosis. Haemorrhage into the cyst or rupture
into the peritoneal cavity causes abdominal pain.
These cysts may be detected on routine prenatal
screening. Hepatobiliary iminodiacetic acid
(HIDA) scans and magnetic resonance cholangi-
P. Ramachandran (*)
Department of Pediatric Surgery, Kanchi Kamakoti
CHILDS Trust Hospital, Chennai, Tamil Nadu, India
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_81
81
opancreatography (MRCP) can be performed to
distinguish them from choledochal cysts and cys-
tic biliary atresia. Asymptomatic cysts do not
require treatment. In the case of large cysts, sur-
gical excision is warranted. Surgical techniques
used in adults, such as cyst deroofing and packing
of the cyst cavity with omentum, have also been
performed in children. At laparotomy, the cyst
fluid may be examined for biliary communica-
tion. Laparoscopic deroofing of the cyst is also a
safe and effective method of treatment. Cyst res-
olution has been reported following percutaneous
injection of ethanol into the cyst.
81.1.2 Fibrocystic Disease
of the Liver
These include a group of congenital disorders
such as polycycstic liver disease, congenital
hepatic fibrosis, choledochal cysts, biliary ham-
artomas and Caroli’s disease. Abnormal develop-
ment of the ductal plate leads to duct dilation and
cyst formation.
81.1.2.1 Polycystic Disease
Liver cysts can occur in patients with autosomal
dominant polycystic kidney disease. In children,
the kidney is more severely affected than the liver
and although liver cysts have been reported, com-
plications are rare. Liver involvement is charac-
terized by numerous cysts spread through the
1117
1118 P. Ramachandran

liver parenchyma. They rarely present before otics for cholangitis. Surgical management to
puberty and treatment is reserved for symptom- relieve obstruction involves removal of stones and
atic patients and is generally supportive of liver hepatico jejunostomy or external drainage.
and kidney function. Ultrasonography shows Localized disease can be managed by lobectomy,
numerous thin walled hypoechoic cysts. Liver but liver transplantation is warranted in severe
transplantation is advised in advanced disease. cases. Caroli’s disease is also included as a part of
Fifty percent of children with autosomal the classification of choledochal cysts, which
recessive polycystic disease of the kidney will have been dealt with elsewhere.
have liver involvement in the form of hepatic
fibrosis. However, large cysts in the liver are rare.
Most children have symptoms of portal hyperten- 81.1.3 Cystic Neoplasms of the Liver
sion because of hepatic fibrosis.
Intra hepatic hemangioma, previously referred to
81.1.2.2 Congenital Hepatic Fibrosis as hemangioendothelioma, present as hypoechoic
This condition occurs in various syndromes and lesions on ultrasound or CT scans and may be
presents with the clinical features of portal hyper- mistaken for abscesses. They are usually self-­
tension. It is caused by the maldevelopment of limiting and present before 6 months of age.
small size interlobular bile ducts. The intrahe- Focal, diffuse and multifocal lesions may be
patic bile ducts are dilated and tortuous and por- seen. Multiple lesions can present with cardiac
tal fibrosis is present. Dilated bile ducts mimic failure and hypothyroidism. Most children are
liver cysts. The outcome is good if the portal successfully treated with low doses of proprano-
hypertension is well controlled. lol. Large lesions may require selective embolisa-
tion or even liver transplantation. There may be
81.1.2.3 Biliary Hamartoma small multifocal hypoechoic lesions scattered
These are characterized by ductal plate malfor- throughout the liver, which are diagnostics of
mations of the small intralobular bile ducts. They multifocal hemangioendotheliomas.
are rare and patients have multiple focal similar Mesenchymal hamartomas are the second
size lesions spread throughout the liver. CT scan most common tumors of the liver in children.
shows hypoattenuated lesions, which are non-­ They present as a benign large multicystic liver
contrast enhancing. They can coexist with simple mass in children younger than 3 years of age.
hepatic cysts and polycystic liver and kidney They arise from mesenchymal cell rests that
disease. become isolated from the normal portal triad and
differentiate independently. Alpha fetoprotein
81.1.2.4 Caroli’s Disease levels are normal. CT scans show a fluid-filled
Caroli’s disease is very rare and presents in infants mass with internal septations and no calcifica-
and young adults. This condition is characterized tion. Most tumors are amenable to surgical exci-
by multiple cystic dilations of the intrahepatic bile sion. Some tumors undergo spontaneous
ducts. It is caused by ductal plate malformations regression. Malignant transformation to embryo-
of the large intrahepatic bile ducts. Segmental bile nal sarcoma has also been reported.
duct dilation is a variant of this disease. It affects Cystic teratomas are rare and present with cal-
only one part of the liver and is rare in children. cification in the margins of the cyst. Serum alpha
Children present with recurrent episodes of chol- fetoprotein is mildly elevated in some cases.
angitis and cholelithiasis. Ultrasonography dem- Resection is the treatment of choice because of
onstrates multiple anechoic intrahepatic bile the potential for malignancy exists especially if
lakes. Treatment includes broad spectrum antibi- there are immature elements.
81  Hepatic Cysts and Abscesses
81.1.4 Parasitic Cysts
Hydatid disease of the liver is caused by two spe-
cies, namely Echinococcus granulosus and
Echinococcus multilocularis. This infection is
endemic in the Middle East, Australia and South
Africa. It commonly occurs between 5 and
15 years of age and the liver is involved in 60% of
children. The hydatid cysts in the liver are slow
growing, although in children, they grow rapidly
and may appear by 3 years of age.
The parasite has a life cycle linked to the inter-
mediate host and a definitive host. The worm
lives in the intestine of animals like the dog
(definitive hosts), and the ova are consumed by
humans and animals like sheep and cattle who
are the intermediary hosts. Humans are affected
by contact with contaminated environments and
infected animals.
In the intermediate host, the ova that are not
destroyed by the gastric juice reach the duode-
num, where the alkaline intestinal contents dis-
solve the membrane of the ovum thereby
releasing the embryos. The embryos penetrate
the bowel wall and reach the liver through the
blood stream where they are filtered by the small
capillaries of the liver. The embryos develop into
cysts in the liver. These cysts have a pericyst
layer composed of connective tissue, epithelial
cells, giant cells and eosinophils. This layer is
densely adherent to the liver tissue. Within the
peri-cyst layer is the endocyst layer composed of
an outer thick layer, which allows permeation of
nutrition into the cyst and an inner germinal
layer, which secretes hydatid fluid and produces
scolices and daughter cysts. The hydatid fluid is
very antigenic and spillage of this fluid can result
in development of cysts in the peritoneal cavity.
Small cysts remain uncomplicated, while
large cysts cause pressure on the adjacent liver
tissue and this leads to right upper quadrant pain
and mass, obstructive jaundice and cholangitis.
Large cysts also cause pain and may rupture into
the peritoneal cavity or into the pleural space.
Sometimes there can be secondary bacterial
infection in the cyst.
Asymptomatic cysts may be incidental find-
ings during ultrasound examinations. Especially
1119
in children from endemic areas, hydatid cysts
must be considered in the differential diagnosis
of hypoechoic lesions of the liver.
Ultrasound scans are accurate in the diagnosis
of these cysts. CT scans show calcification of the
cyst wall and the presence of daughter cysts. The
reliability of the Casoni’s skin test is less than the
indirect hemagglutination tests and the enzyme
linked immunosorbent assays (ELISA) using
urine or saliva.
Mebendazole and Albendazole are the drugs
of choice in the treatment of hydatid disease and
must be continued for 1–2 months to prevent
recurrence. Surgery is restricted to children with
large cysts (<5  cm in diameter) or complicated
cysts that have ruptured. During laparotomy, the
large cysts are aspirated and carefully opened to
avoid spillage. The endocyst wall is carefully
removed by blunt dissection and the residual cav-
ity is closed with mass sutures. Cyst perforation
must be treated urgently and high dose steroids in
the postoperative period reduce the effects of
anaphylaxis. Recently, sonography-guided per-
cutaneous aspiration of the cyst with injection of
hypertonic saline or absolute alcohol has been
reported to be successful.
81.2 Hepatic Abscesses
Hepatic abscesses were initially reported in chil-
dren with perforated appendicitis, but the inci-
dence has decreased with the advent of antibiotics.
However, intraabdominal infections in immuno-
compromised children still continue to be the
source of hepatic abcesses. Bacterial infections are
transmitted to the liver via the hepatic artery, bile
ducts or portal veins. Most pyogenic abscesses in
the western world are seen in children with immu-
nodeficiencies like chronic granulomatous disease
and those who are immunosupressed from chemo-
therapy and transplants. However, in Asia and
Africa, pyogenic abscesses are still seen in neo-
nates with omphalitis. In about a fifth of the chil-
dren, a source for the hepatic abcess cannot be
identified. Almost half of all liver abscesses occur
in children younger than 5 years of age and a quar-
ter occur during the first year of life.
1120
Pyogenic abscesses may arise in children who
have recurrent cholangitis secondary to choleli-
thiasis and choledochal cysts, which causes
inflammation and obstruction of the bile ducts.
Following a portoenterostomy for biliary atresia,
intrahepatic biliomas may develop into pyogenic
abscesses in the presence of secondary infection.
Hematomas of the liver sustained from blunt
abdominal trauma may develop into pyogenic
abscesses. In generalized sepsis, bacterial seed-
ing of the liver leads to pyogenic abscesses. Such
abscesses have been reported in patients with
pneumonia, osteomyelitis and endocarditis.
Pyogenic abscesses develop in benign hepatic
cysts following attempts at percutaneous
­drainage. Infection in these non-neoplastic cysts
may also occur in the absence of intervention.
Infection leads to an increase in size of the cysts
with thickening and inflammation of the cyst
wall. The commonest pathogens found in hepatic
abscesses are Staphylococcus aureus, Escherichia
coli, Hemophilus influenza and Pseudomonas.
Clinical features include fever, abdominal
pain, jaundice, shoulder pain and dyspnoea. The
symptoms may be of long duration, especially in
chronic abscesses. These children also have
weight loss and anemia. Diagnosis may be
delayed if there is a low index of suspicion. The
liver is enlarged and tender and a pleural effusion
may be present. 50% of patients have a positive
blood culture. Fungal cultures are positive in
liver abcesses occurring in children with leuke-
mia. Ultrasound scans are the mainstay in the
diagnosis of liver abscesses. CT scans may iden-
tify abscesses missed on ultrasound scan. A chest
X-ray may show elevation of the right hemi dia-
phragm and a right-sided pleural effusion. 70%
of liver abscesses are multiple. Anemia, leucocy-
tosis and a raised ESR are found in the majority
of cases. Broad spectrum antibiotics and image
guided percutaneous drainage is the treatment of
choice, especially in multiple abscesses.
Intravenous antibiotic therapy must be continued
for 4–6 weeks, followed by oral antibiotic ther-
apy for an additional 4–8 weeks. Drainage of the
abscess, as well as percutaneous catheter place-
ment, can be facilitated by laparoscopy. Open
drainage is advocated when there is associated
P. Ramachandran
intestinal pathology that needs intervention or
when the diagnosis is in doubt.
Neonatal liver abscesses occur in premature
rupture of membrane. Rarely, they may occur in
the neonates with omphalitis. They usually pres-
ent as miliary abscesses and hence are not ame-
nable to percutaneous drainage. Hence, long term
antibiotic therapy is the mainstay of treatment.
Septicemia in children with liver abscesses is
associated with a higher mortality. Large
abscesses may rupture into the subphrenic space,
the pleural cavity and the peritoneal cavity, in
which case these children require a drainage pro-
cedure and broad spectrum antibiotics.
81.2.1 Amoebic Liver Abscesses
Entamoeba histolytica infections are transmitted
from person to person due to poor sanitation in
overpopulated areas. It is endemic to Africa and
Asia. Liver involvement is the commonest extra
intestinal manifestation of amoebiasis. The tro-
phozoites from the colon enter the liver via the
portal vein and initiate abscess formation. Liver
involvement is infrequent in children. The right
lobe is more commonly affected. The clinical
features include fever with rigor, weight loss and
right upper quadrant pain. Dyspnoea is also seen
in the presence of a right sided pleural effusion.
The liver is enlarged and tender. A rapid increase
in distension is an indication that rupture of the
cyst is imminent.
Mild anemia, leucocytosis, raised ESR and
mildly elevated liver enzymes may be present.
The parasite is identified in the stool in only 30%
of cases. Serological tests like an indirect hemo-
globin assay is highly sensitive and specific,
although PCR has become available and this may
prove to be more accurate in the diagnosis. Other
investigations include a chest X-ray, which may
show an elevated diaphragm on the right side,
pleural effusion and right lower lobe infiltration.
An ultrasound scan accurately identifies the loca-
tion and size of these abscesses and image-aided
percutaneous aspiration to identify amoeba on
smear will confirm the diagnosis. CT scan offers
no advantage over ultrasound.
81  Hepatic Cysts and Abscesses
Most cases of amoebic liver abscesses
respond to Metronidazole. With intravenous
Metronidazole in doses of 35–50  mg/kg, clini-
cal remission is seen within 72  h in a majority
of cases. Radiological remission takes longer.
Closed aspiration is recommended if there is no
response to treatment after 4 days and in abcesses
larger than 5 cms or in those located in the left
lobe. Percutaneous aspiration is also advocated
in rapidly enlarging abscesses or when the diag-
nosis is in doubt and a pus culture is required.
Intraperitoneal rupture causes peritonitis and
shock. In such cases, urgent surgical exploration
is necessary. Rarely the abscess ruptures into the
pleural or pericardial cavity.
Rupture into the pleural cavity is marked by
severe dyspnea and tachypnea with clinical signs
of pleural effusion. Pleural collections can be
drained by a tube thoracostomy. Rupture into the
pericardial space is a serious complication lead-
ing to cardiac tamponade. The pericardial effu-
sion requires urgent decompression. Percutaneous
drainage of the liver abscesses will also drain the
pericardial effusion since there is a communica-
tion between both cavities. Peritoneal rupture
requires open surgical drainage and lavage.
Fortunately, the mortality of the disease and
the incidence of complications are on the decline
due to early diagnosis and prompt treatment.
81.3 Conclusion
Several diseases phenotypically present with
hepatic cysts. Hence, it is important to have an
algorithmic approach to investigate and manage
these cysts. Hepatic infections, which present as
abscesses in the liver, also have features which dis-
tinguish among the various disease causing patho-
gens. Appropriate identification of the pathogen is
important to institute effective treatment.
1121
Further Reading
Andrews WS (2006) Gall Bladder disease and hepatic
infection. In: O’Neil JA, Coran AG, Fonkalsrud EW
(eds) Pediatric surgery. Mosby Elsevier, Philadelphia,
pp 1635–1650
Cenk Buyukunal SW (2002) Hydatid disease. In: Howard
ER, Stringer MD, Colombani PM (eds) Surgery of the
Liver, Bile Ducts and pancreas in children. Arnold,
Hodder Headline Group, Euston, London, pp 355–362
Holcomb GW III, Andrew WS (2012) Gallbladder disease
and hepatic infections. In: Coran AG, Caldamone A
et al (eds) Pediatric surgery, 7th edn. Oxford Elsevier
Limited
Howard ER (2002) Cysts. In: Howard ER, Stringer MD,
Colombani PM (eds) Surgery of the liver, bile ducts
and pancreas in children. Arnold, Hodder Headline
Group, Euston, London, pp 239–246
Jain P, Shah H, Parelkar S (2008) Laparoscopic deroofing
of congenital solitary liver cyst in a neonate managed
at day 16 of life. J Laparoendosc Adv Surg Tech A
18(3):477–480. https://doi.org/10.1089/lap.2007.0186
Joseph VT (2002) Liver abscesses. In: Howard ER,
Stringer MD, Colombani PM (eds) Surgery of the
liver, bile ducts and pancreas in children. Arnold,
Hodder Headline Group, Euston, London, pp 355–362
Nordin AB, Fallon SC, Carter BA, Brandt ML (2013)
Congenital hepatic cyst with antenatal diagno-
sis: a case report and literature review. Pediatr
Surg Int 29(8):847–850. https://doi.org/10.1007/
s00383-­013-­3312-­x
Puri P, Höllwarth ME (eds) (2006) Pediatric surgery.
Springer, Berlin, Heidelberg
Shabbir Z, Javaid A (2018) Islah Ud Din. Infantile hepatic
hemangioma. J Pak Med Assoc 68(12):1846–1847
Stringer MD, Alizai NK (2005) Mesenchymal ham-
artoma of the liver: a systematic review. J Pediatr
Surg 40(11):1681–1690. https://doi.org/10.1016/j.
jpedsurg.2005.07.052
Sweeney WE, Avner ED (2001, July 19) Polycystic kidney
disease, autosomal recessive. In: Adam MP, Ardinger
HH, Pagon RA et al (eds) GeneReviews. University of
Washington, Seattle (WA)
Veigel MC, Prescott-Focht J, Rodriguez MG et al (2009)
Fibropolycystic liver disease in children. Pediatr
Radiol 39(4):317–421. https://doi.org/10.1007/
s00247-­008-­1070-­z
 Portal Hypertension
Mark D Stringer
82.1 Introduction
Portal hypertension (PHT) is a manifestation of a
wide variety of conditions, each with a different
etiology and natural history. It frequently pres-
ents with bleeding from esophageal varices
which is the commonest cause of serious gastro-
intestinal bleeding in children. Precise diagnosis,
a clear understanding of therapeutic options, and
a multidisciplinary approach are essential for
successful management.
82.2 Pathophysiology
and Definition
The portal vein transports blood to the liver from
the gastrointestinal tract and spleen (Fig.  82.1),
contributing two-thirds of the liver blood supply.
The right and left branches of the portal vein
undergo several divisions to supply each of the
eight liver segments, where they terminate in the
hepatic sinusoids.
M. D Stringer (*)
Department of Paediatric Surgery, Wellington
Children’s Hospital and Department of Paediatrics
and Child Health, Wellington School of Medicine,
University of Otago, Wellington, New Zealand
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_82
82
Portal venous pressure is the product of blood
flow and vascular resistance. Thus, PHT may
develop as a result of increased vascular resis-
tance (e.g. portal venous obstruction or hepatic
fibrosis) or increased blood flow (e.g. arteriopor-
tal fistula). In cirrhosis of the liver, there is both
an increase in splanchnic blood flow and an
increased resistance to portal blood flow within
the liver.
The normal portal venous pressure is
5–10  mmHg which exceeds inferior vena caval
pressure by a few millimetres. A rise in portal
pressure leads to splenomegaly and the develop-
ment of varices at sites of portosystemic anasto-
moses: the distal esophagus (esophageal varices);
the anal canal (anorectal varices); the falciform
ligament (umbilical varices); and in the abdomi-
nal wall and retroperitoneum. The junction
between mucosal and submucosal varices in the
distal 5 cms of the esophagus is a potential site of
rupture causing variceal bleeding. PHT is defined
by an increased hepatic venous pressure gradient
of more than 5 mmHg, i.e. the difference between
the wedged hepatic venous pressure (which
reflects portal venous pressure) and the free
hepatic venous pressure. Oesophageal varices
develop when the gradient exceeds 12  mmHg.
The risk of variceal bleeding is greatest in larger
varices with a higher internal pressure and wall
tension.
1123
1124
Fig. 82.1 The
portomesenteric venous
system in portal
hypertension
Portal vein
Superior
mesenteric vein
82.3 Classification and Etiology
The vast majority of children with PHT have one
of the following underlying causes:
–– Primary venous obstruction at a prehepatic
(e.g. portal vein obstruction), intrahepatic (e.g.
veno-occlusive disease) or posthepatic (e.g.
Budd-Chiari syndrome) level (Table 82.1),
–– chronic liver disease (e.g. hepatic fibrosis or
cirrhosis) (Fig. 82.2).
Chronic liver disease is the commonest overall
cause of PHT but isolated extrahepatic portal
vein occlusion (PVO) is the most frequent cause
of extrahepatic portal hypertension. Occasionally,
the picture is mixed as in the child with cirrhosis
complicated by portal vein thrombosis. Very
rarely, the primary cause of portal hypertension is
not venous obstruction but excessive arterial
inflow into a non-obstructed portal venous sys-
tem caused by a hepatic arterioportal fistula
(Ramachandran et al. 2014). The etiology of PHT
is key to determining appropriate treatment.
M. D Stringer
Left gastric vein
Splenic vein
Inferior
mesenteric vein
Many etiologic factors have been implicated
in the cause of PVO which accounts for up to
one-third of all children with bleeding oesopha-
geal varices (Table 82.2). Umbilical vein cathe-
terization in the newborn, with or without
infusion of irritant solutions, is a recognised but
uncommon cause of portal vein thrombosis
(Gharehbaghi et al. 2011). Thrombophilic disor-
ders such as deficiencies of protein C, S or anti-
thrombin III or gene mutations affecting Factor
V Leiden, prothrombin, or the enzyme methyl-
tetrahydrofolate reductase can predispose to
portal vein thrombosis but account for only a
small proportion of cases (Pinto et  al. 2004;
Bertocchini et al. 2014). However, it should be
noted that circulating levels of the natural anti-
coagulant proteins (C, S, and antithrombin III)
are commonly decreased as a result of PVO
(Mack et al. 2003).
In most children with isolated PVO, the etiol-
ogy is unknown. The portal vein is typically
replaced by multiple venous collaterals, the so-­
called portal vein cavernoma. Occasionally,
additional congenital anomalies such as con-
82  Portal Hypertension 1125

Table 82.1  Aetiology of portal hypertension in children

Anatomy Etiology
Primary venous Prehepatic Portal vein Congenital
obstruction occlusion Umbilical catheterization/sepsis in newborn
Portal pyemia (e.g. appendicitis)
Thrombophilic disorder
Tumours, pancreatitis, choledochal cyst, etc.
Intrahepatic Sinusoidal/ Veno-occlusive disease—Toxic sinusoidal injury (e.g.
venular occlusion chemotherapy, herbal toxins, bone marrow transplant)
Schistosomiasis
Hepatoportal sclerosis
Posthepatic Hepatic vein Budd-Chiari syndrome
occlusion
Chronic liver Liver fibrosis/ Biliary atresia, cystic fibrosis, chronic viral hepatitis,
disease cirrhosis autoimmune hepatitis, alpha-1-antitrypsin deficiency,
congenital hepatic fibrosis etc.
Very rarely, PHT occurs without intrinsic obstruction of the portal/hepatic venous systems (e.g. chronic constrictive
pericarditis or arterioportal fistula)

a b

Fig. 82.2  Cirrhosis and portal hypertension in a child. secondary to portal hypertension. Note the portal vein is
(a) Magnetic resonance scan/angiogram showing a patent. (b) Appearance of the cirrhotic liver at liver
shrunken cirrhotic liver with massive splenomegaly (Sp) transplantation

genital heart disease, intestinal malrotation,
duodenal atresia and craniofacial dysostosis are
present, supporting the suggestion that isolated
PVO is often a developmental malformation.
The portal venous system develops from paired
vitelline and umbilical veins, which drain the
yolk sac and placenta, respectively. These veins
intercommunicate around the embryonic duode-
num and supply the developing hepatic sinu-
soids. Selective involution of these embryonic
veins leaves the final arrangement of a single
postduodenal portal vein. Aberrations in this
process probably explain the origin of most por-
tal vein cavernomas.
1126
Table 82.2  Etiologic factors in portal vein occlusion/
thrombosis
General Factors
Developmental malformations
Septicemia
Thrombophilia
Local Factors
Umbilical sepsis/umbilical catheterization ± infusion of
irritant solutions in the newborn
Intra-abdominal sepsis and portal pyemia
Abdominal trauma (including surgical)
Structural lesions e.g. portal vein web
Cholangitis/choledochal cyst
Pancreatitis
Tumour infiltration/compression (including
lymphadenopathy)
Splenectomy
Post-liver transplant (local and general factors)
82.4 Clinical Features
PHT typically presents with acute gastrointesti-
nal bleeding and/or splenomegaly or with symp-
toms and signs of chronic liver disease. Variceal
bleeding results in hematemesis and/or melena
and may occur at any age. Bleeding is occasion-
ally precipitated by an upper respiratory tract
infection. Clinical examination usually reveals
splenomegaly. Rarely, this is caused by isolated
splenic vein obstruction rather than generalized
portal hypertension. Ascites usually denotes the
presence of chronic liver disease but may occur
transiently after a major variceal bleed in those
with extrahepatic portal hypertension.
Encephalopathy can complicate an episode of
bleeding in cirrhotics but is rare in children with
PVO (although may be detected at a subclinical
level [Srivastava et al. 2010]).
Portal hypertension may cause mucosal edema
in the small intestine leading to malabsorption,
protein loss, and failure to thrive. In established
M. D Stringer
portal hypertension dilated cutaneous collateral
veins carry blood away from the umbilicus
toward the tributaries of the vena cava (caput
medusae). Rarely, hepatopulmonary syndrome
(liver disease, hypoxemia, and intrapulmonary
vascular dilatation) may coexist with portal
hypertension, most often in children with cirrho-
sis but occasionally in those with PVO (Sari et al.
2012). Pulmonary arterial hypertension (porto-
pulmonary syndrome) is a similarly rare compli-
cation arising from vasoconstriction and
endothelial and smooth muscle proliferation in
the pulmonary arteries. It can occur as a compli-
cation of PHT with or without intrinsic liver dis-
ease (Tingo et al. 2017). Varices may develop at
sites other than the esophagus or stomach (ecto-
pic varices) such as the anorectum, at sites of pre-
vious intestinal anastomoses and around stomas.
In long-standing PVO, varices around the com-
mon bile duct can cause obstructive jaundice
(portal biliopathy).
Budd-Chiari syndrome due to hepatic vein
thrombosis is a rare condition in children. In ado-
lescent girls, onset may be precipitated by the
oral contraceptive pill. Similar clinical features
develop after hepatic vein occlusion from trauma,
malignancy, prothrombotic disorders or surgery
and in patients with retrohepatic inferior vena
caval obstruction. These include hepatomegaly,
intractable ascites, symptoms and signs of portal
hypertension, diarrhea, and progressive cachexia.
Jaundice is variable. Onset may be acute but is
more often chronic. The caudate lobe is fre-
quently spared and often hypertrophied because
of its independent venous drainage directly into
the inferior vena cava (rather than into the three
main hepatic veins) (Fig.  82.3). Caudate lobe
hypertrophy with caval compression or inferior
vena caval occlusion may cause lower limb
edema.
82  Portal Hypertension
a b
Fig. 82.3  Budd-Chiari syndrome in a 12-year-old boy.
(a) The magnetic resonance scan shows hepatomegaly
and vascular congestion with caudate lobe hypertrophy
82.5 Diagnosis and Investigation
Initial investigations are directed towards deter-
mining the cause and consequences of the PHT,
establishing if liver function or architecture is
impaired, and assessing the patency of the porto-
mesenteric venous system.
• Hematology—A full blood count may show
anemia, leucopenia and thrombocytopenia
from hypersplenism. The prothrombin time is
often prolonged in patients with chronic liver
disease. In patients with Budd-Chiari syn-
drome, an underlying myeloproliferative dis-
order or thrombophilia must be excluded.
Patients with a previous history or family his-
tory of venous thrombosis and those being
considered for meso-portal bypass should also
be screened for thrombophilic disorders.
• Biochemical liver function—In PVO, liver
function tests are typically normal although
plasma albumin may be reduced after a variceal
bleed. Chronic liver disease causes abnormal
liver function but standard biochemical LFTs
can be normal in well-compensated cirrhosis.
• Abdominal ultrasound scan—This confirms
non-specific features of portal hypertension
such as large collateral veins and splenomeg-
1127
(C). (b) Appearance of congested hepatomegaly at liver
transplantation
aly. The hepatic echotexture is abnormal in
chronic liver disease. In PVO, the normal por-
tal vein is usually replaced by a leash of col-
lateral veins. Colour Doppler flow studies
demonstrate the direction and velocity of flow
in the portal vein, hepatic veins and vena cava.
• Gastrointestinal endoscopy—This can be used
to evaluate gastro-esophageal and anorectal
varices and mucosal features of portal hyper-
tension. Esophageal varices are graded accord-
ing to severity. Large varices may show ‘red
signs’ of recent or impending variceal hemor-
rhage (Fig. 82.4a). Portal gastropathy is char-
acterised by mucosal hyperemia (Fig. 82.4b).
• Computerized tomography and magnetic reso-
nance imaging—Both imaging modalities are
useful in detecting focal liver lesions associ-
ated with portal hypertension. Magnetic reso-
nance angiography provides detailed
information about the patency and calibre of
veins within the portomesenteric system
(Fig. 82.5). This is essential when considering
meso-portal bypass (Cárdenas et  al. 2012) or
portosystemic shunt surgery and in the preop-
erative workup of patients with an occluded
portal vein before liver transplantation. Hepatic
venography is useful in evaluating Budd-
Chiari syndrome.
1128
a b
Fig. 82.4 (a) Esophageal varices with red signs. (b) Portal gastropathy in the gastric antrum (the pylorus is on the
right)
Fig. 82.5  Coronal magnetic resonance scan after gado-
linium enhancement demonstrating a portal vein caver-
noma in a child with extrahepatic portal hypertension. At
the porta hepatis the main portal vein is replaced by a
leash of collateral vessels rather than dividing into right
and left branches
• Percutaneous liver biopsy—If there are no
contraindications, a percutaneous biopsy is
undertaken to diagnose or exclude underlying
liver disease. In extrahepatic PVO, the liver
architecture is essentially normal.
M. D Stringer
82.6 Treatment
and Complications
If acute portal vein thrombosis is diagnosed
promptly (e.g. after umbilical venous catheteriza-
tion), thrombolytic therapy and/or anticoagulation
can sometimes restore normal patency, but this is an
uncommon scenario. More typically, the child pres-
ents with splenomegaly or variceal bleeding from
chronic portal hypertension due to PVO or chronic
liver disease. Initial treatment is focused on the
emergency management of active variceal bleeding
and the prevention of recurrent variceal bleeding.
With appropriate resuscitation, the mortality from a
first variceal bleed is less than 1% (Shneider et al.
2016). Subsequently, in those with good liver func-
tion (e.g. PVO or congenital hepatic fibrosis), con-
trol of portal hypertension is the priority, but in the
child with cirrhosis, management is also dictated by
the nature and severity of underlying liver disease.
In children with known esophageal varices second-
ary to chronic liver disease but who have never bled,
primary prophylaxis can be considered.
82.6.1 Emergency Management
of Variceal Bleeding
Bleeding from esophageal varices is life-­
threatening. Resuscitation, investigation and ini-
82  Portal Hypertension 1129

tial treatment are outlined in Table  82.3.

Octreotide, a long-acting analogue of somatosta-
tin with a plasma half-life of more than 1 hour,
reduces splanchnic blood flow and portal pres-
sure with minimal short-term side effects.
Balloon tamponade is rarely required to control
active variceal bleeding  but, if necessary, a

Table 82.3  Emergency management of bleeding esoph-

ageal varices
Resuscitation
•  Airway (must be secure)
•  Breathing (give oxygen if shocked)
• Circulation: insert two intravenous cannulae (22G or
larger) and commence intravenous fluids (10 mls/kg
crystalloid boluses)
Investigation
• Full blood count, clotting, urea, creatinine,
electrolytes, liver function tests
• Blood cultures and cross-match (at least two units of Fig. 82.6  Sengstaken tube in situ in a young child with
packed red blood cells) severe variceal bleeding (prior to endoscopic therapy).
•  Monitor and maintain blood glucose Note that only the gastric balloon has been filled with con-
• Accurate monitoring of (a) fluid balance and (b) trast and that the child has an endotracheal tube in place
cardiorespiratory status
•  Watch for encephalopathy Sengstaken-type tube can be inserted after secur-
Treatment ing the airway by endotracheal intubation; only
• Transfuse packed red cells slowly aiming for Hb of
8–10 g/dl (avoid overtransfusion)
the gastric balloon is inflated, and correct posi-
• Give vitamin K 300 μg/kg (maximum 10 mg) slowly tioning must be verified radiographically
i.v. and correct coagulopathy with fresh frozen (Fig. 82.6). Moderate traction is applied by tap-
plasma and platelets ing the tube to the side of the face; excessive trac-
• Start octreotide infusion: bolus dose of 1 μg/kg i.v. tion may cause catastrophic balloon displacement.
(maximum 50 μg) over 5 mins followed by infusion
The balloon is deflated after 12–24 h at the time
at 1–3 μg/kg/hr. (maximum 50 μg/hr) via a dedicated
line. Continue infusion until 24 hr, after bleeding of endoscopy.
ceases and then wean off slowly over 24 hrs
•  Keep nil by mouth initially
•  Give gastric protection: Omeprazole i.v. and oral 82.6.2 Endoscopic Treatment
Sucralfate
of Esophageal Varices
•  Start i.v. antibiotics if any evidence of sepsis
• Ensure appropriately sized pediatric Sengstaken-type
tube is available to provide balloon tamponade if 82.6.2.1 Injection Sclerotherapy
necessary This is performed with a flexible fiberoptic gas-
• Consider prophylaxis against encephalopathy if poor troscope under general anaesthesia with an endo-
liver function
tracheal tube in place (Stringer 2005). A variety
• Urgent upper GI endoscopy within 24 hr. to confirm
the source of bleeding and to begin treatment of of injection techniques and sclerosants have been
varices by banding or sclerotherapy, if appropriate used. An intravariceal injection technique using
5% ethanolamine oleate is popular (Fig. 82.7a).
Modified from Stringer MD and Howard ER.  Surgical Varices are initially injected every few weeks and
Disorders of the Liver and Bile Ducts and Portal then at monthly intervals until sclerosis is com-
Hypertension. In: Kelly DA (ed) Diseases of the Liver and
Biliary System in Children. Blackwell Publishing Ltd.,
plete. Patients are often given oral Sucralfate and
Oxford, 2004, pp345 Omeprazole for two weeks after each injection
session to reduce complications from ulceration.
1130 M. D Stringer

Fig. 82.7 (a) Endoscopic injection sclerotherapy (intra- transparent cylinder attached to the end of the gastroscope
variceal technique). (b) Endoscopic variceal ligation. (c) holding the bands limits the view
Endoscopic view during variceal ligation (banding). The
82  Portal Hypertension
c
Fig. 82.7 (continued)
Endoscopic review is undertaken after 6 months
and then annually; only large recurrent varices
require further treatment.
Endoscopic injection sclerotherapy (EIS) is a
highly effective treatment for esophageal varices
(Stringer 2006; Maksoud-Filho et  al. 2009).
However, numerous complications have been
reported. Transient retrosternal discomfort and
fever are common after EIS and usually resolve
within 48  hrs. Children with a pyrexia >38  °C
should be treated with intravenous antibiotics
pending the results of blood cultures because
they may have transient bacteremia. Antibiotic
prophylaxis is given to patients at risk of bacterial
endocarditis, those with cirrhosis and ascites, and
the immunosuppressed. Major complications of
sclerotherapy include esophageal stricturing
(which responds to esophageal dilatation), recur-
rent esophageal varices, and intercurrent bleed-
ing occurring between treatment episodes (due to
a persistent varix or an injection ulcer). Some
children experience intermittent dysphagia and
heartburn secondary to esophageal dysmotility
and gastro-esophageal reflux. Dissemination of
the injected sclerosant can rarely cause a distant
complication. Complication rates have fallen in
recent years, probably as a result of injecting
smaller volumes of sclerosant.
82.6.2.2 Variceal Ligation (Banding)
The varix is aspirated into a transparent cylinder
fitted to the end of a flexible gastroscope and an
1131
elastic band is released by a trip wire passing
through the biopsy channel. This strangles the
varix which thromboses and sloughs (Fig. 82.7b
and c) (Stringer 2005). Treatment begins with
ligation of the most distal varix in the esophagus
just above the cardia. Up to four bands are applied
to the varices at each session; the treatment is
repeated every few weeks, and then monthly,
until the varices have been obliterated.
Endoscopic variceal ligation (EVL) is the opti-
mum endoscopic method of treating bleeding
esophageal varices and for preventing rebleeding
in children. Compared to EIS, EVL achieves
more rapid eradication with fewer treatment ses-
sions and a significantly lower complication rate
(Zargar et  al. 2002; Pimenta et  al. 2017).
Esophageal ulcers caused by banding are more
superficial and resolve quicker than those induced
by sclerotherapy. However, EVL poses technical
challenges in infants on account of their size rela-
tive to that of the equipment.
Most gastric varices are fundal and directly
contiguous with lower esophageal varices; these
tend to be eradicated by treatment of the esopha-
geal varices. Isolated gastric varices are less com-
mon and more difficult to control using
endoscopic techniques but injection with tissue
adhesive (N-butyl-2-cyanoacrylate) or balloon-­
occluded retrograde transvenous obliteration
may be successful if definitive treatment of portal
hypertension is not possible (Mansoor-Ul-Haq
et al. 2020; Deswal et al. 2020).
82.6.3 Primary Prophylaxis
of Variceal Bleeding
This is a contentious issue with no clear answer
on the role of pharmacologic or endoscopic inter-
ventions (Shneider 2017; Gana et al. 2019; Gana
et al. 2020).
• Non-selective beta-blockers such as proprano-
lol reduce portal pressure by causing splanch-
nic vasoconstriction and decreasing cardiac
output. These agents are effective in prevent-
ing variceal bleeding and rebleeding in adults
with cirrhosis but there are no randomized
1132
controlled trials in primary prophylaxis in
children and their use in this setting is not cur-
rently recommended (Shneider et al. 2016).
• Prophylactic endoscopic therapy is controver-
sial. Prophylactic EIS does not offer an overall
survival advantage, but EVS may be beneficial
in preventing bleeding in children with large
esophageal varices related to liver disease
(Shneider et  al. 2016; Bernabeu et  al. 2020;
Gattini et al. 2020).
However, there is expert agreement that chil-
dren with extrahepatic PVO should be offered
meso-portal bypass surgery (see below) for pri-
mary and secondary prophylaxis of variceal
bleeding and other complications of PHT pro-
vided several criteria are met: (1) appropriate sur-
gical expertise is available, (2) preoperative and
intraoperative assessment shows favorable anat-
omy, and (3) there is appropriate multidisci-
plinary care pre- and post-operatively (Shneider
et al. 2016).
82.6.4 Surgery for Portal
Hypertension
Surgery and endoscopic therapy are complemen-
tary procedures in the management of portal
hypertension. The indications for surgery have
changed over the years as the relative merits of
portosystemic shunt surgery, endoscopic therapy,
and meso-portal bypass have become clearer.
Current indications for surgery are:
• Primary treatment of PVO
When anatomically feasible, the mesenterico-­
left portal (Rex) bypass is the optimum treatment
of portal hypertension in children with extrahe-
patic portal hypertension secondary to PVO. This
applies to both primary and secondary prophy-
laxis of variceal bleeding (Shneider et al. 2016).
In these patients, the liver is intrinsically healthy.
First performed in 1989 for acquired PVO in a
young adult (Chen et al. 1992), the procedure was Fig. 82.8 Diagram illustrating the standard type of
modified by Jean de Ville de Goyet to treat PVO meso-portal (Rex) bypass. The bypass graft is typically
complicating liver transplantation (de Ville de autologous internal jugular vein
M. D Stringer
Goyet et al. 1992) and subsequently championed
by him as a treatment for isolated PVO (de Ville
de Goyet et  al. 1998; di Francesco et  al. 2014).
The procedure is contraindicated in patients with
chronic liver disease since the site of venous
obstruction in these patients is intrahepatic. In the
standard procedure, an autologous internal jugu-
lar vein graft is interposed between the superior
mesenteric vein and the intrahepatic left portal
vein located within the umbilical fissure (Rex
recess) of the liver adjacent to the falciform liga-
ment (Figs. 82.8 and 82.9). The portal vein occlu-
sion is bypassed, hepatic portal blood flow is
restored and portal hypertension is corrected. The
operation requires the presence of an adequate
calibre, patent intrahepatic left portal vein with
good run-off into the liver. Ideally, the splenic and
superior mesenteric veins should be patent and in
continuity; this is usually determined preopera-
tively using a combination of magnetic resonance
angiography, ultrasound and/or retrograde hepatic
venography (Cárdenas et  al. 2012; Bertocchini
et al. 2014). Both internal jugular veins should be
patent on ultrasound. This bypass is an ideal
option for children with isolated PVO since it
restores normal portal venous physiology, meta-
bolic function, somatic growth, and, in some, neu-
rocognitive function (Stringer 2007; Lautz et  al.
2009). Alternative tributaries of the portomesen-
teric venous system (e.g. coronary vein, inferior
mesenteric vein, splenic vein) can be used to pro-
Umbilical segment
of left portal vein
Internal jugular
vein bypass graft
Portal vein Splenic vein
cavernoma
Superior
mesenteric vein
82  Portal Hypertension 1133

a b

Fig. 82.9  Operative views of a completed meso-portal left portal vein in the Rex recess (after division of an over-
bypass (Rex shunt) for extrahepatic PVO in a child using lying liver bridge in this case). (b) The bottom end of the
autologous internal jugular vein. (a) The internal jugular internal jugular vein bypass graft anastomosed to the
vein bypass graft (IJV) is anastomosed to the intrahepatic superior mesenteric vein (SMV)

vide inflow to the intrahepatic left portal vein if

the superior mesenteric vein is occluded (Ha et al.
2015), but venous collateral communications with
the selected vein must allow adequate decompres- Bypass graft
Intrahepatic PV
sion of the entire portomesenteric venous system.
Recently, a portal bypass to the right portal vein
has been described (Long et al. 2018), but experi-
ence of this technique is limited. Some form of
meso-­portal bypass is not feasible in up to 60% of SMV
children with PVO because the intrahepatic veins
are thrombosed or too abnormal (e.g. after portal
vein thrombosis complicating umbilical vein
catheterization (Gibelli et al. 2011; Guérin et al.
Fig. 82.10  Postoperative colour Doppler ultrasound scan
2013; Bertocchini et  al. 2014). Alternatives to of Rex bypass graft confirming good patency and flow
autologous jugular vein as the bypass conduit velocity. SMV superior mesenteric vein, PV portal vein
(e.g. composite saphenous vein grafts, recana-
lized umbilical vein, synthetic materials, etc.)
have generally yielded less favourable long-term ful bypass/shunt surgery, growth velocity and
patency results. hypersplenism both improve (the latter reflected
Patency of a Rex bypass or portosystemic in an increased platelet count and a reduction in
shunt can be assessed directly by colour Doppler splenomegaly).
ultrasound imaging (Fig. 82.10), magnetic reso- Thrombosis of a meso-portal bypass or porto-
nance angiography, or conventional angiography. systemic shunt is a major complication, likely to
It may also be assessed indirectly by ultrasound manifest as recurrent variceal bleeding. With
from directional flow in portomesenteric and sys- appropriately selected patients in experienced
temic veins and spleen size, and by endoscopic centres, patency rates after the standard meso-­
evidence of regression of varices. After success- portal Rex bypass using autologous jugular vein
1134 M. D Stringer

are typically greater than 90% between one and that it achieves gastrosplenic variceal decompres-
five years later (Sharif et  al. 2010; Lautz et  al. sion whilst maintaining some hepatopetal portal
2013; Zhang et al. 2018). Failure rates have his- perfusion (Shilyansky et  al. 1999; Botha et  al.
torically been slightly higher with portosystemic 2004). If a portosystemic shunt is considered,
shunts. Stenosis of the bypass or shunt is another then the type of shunt selected is determined by
potential complication; this may be corrected by the pattern of portal venous occlusion (Stringer
percutaneous interventional radiology. et  al. 1994; Bertocchini et  al. 2014) i.e. which
Encephalopathy is a well-recognised complica- veins are patent, together with available surgical
tion of portosystemic shunt surgery in patients expertise.
with cirrhosis; it occurs infrequently after shunt
surgery for PVO (Valayer and Branchereau • Bleeding gastric or ectopic varices that can-
1998), although can be detected more often at a not be controlled endoscopically.
subclinical level (Srivastava et  al. 2010).
Portopulmonary syndrome (Valayer and Potential options include meso-portal bypass,
Branchereau 1998) and liver nodules (adenoma portosystemic shunt, or liver transplantation,
or focal nodular hyperplasia) are potential late depending on the aetiology of portal
but rare complications of portosystemic shunt hypertension.
surgery (Guérin et al. 2009).
• Massive splenomegaly causing severe hyper-
• Uncontrolled bleeding from esophageal vari- splenism or abdominal pain.
ces (not responding to at least two endoscopic
treatment sessions). All patients with PHT have some degree of
hypersplenism manifest as splenomegaly with
This is an unusual situation in PVO but if vari- leukopenia and thrombocytopenia. In most cases,
ceal bleeding cannot be controlled endoscopi- no specific treatment is needed. When severe, a
cally and a meso-portal bypass is not feasible, a meso-portal bypass or portosystemic shunt pro-
portosystemic shunt is an alternative option. A cedure may be necessary; splenectomy alone
surgical portosystemic shunt or TIPS procedure does not address the PHT itself, risks compro-
(see below) may also be used to control variceal mising future portosystemic shunt options and is
bleeding in patients with chronic liver disease seldom an appropriate option unless there is
provided synthetic liver function is well-­ widespread portomesenteric thrombosis preclud-
preserved, as in some children with hepatic fibro- ing other options. There are some reports of suc-
sis or cystic fibrosis. However, this is often a cessful treatment with partial splenic embolization
palliative procedure in this circumstance and which controls hypersplenism and may reduce
liver transplantation is often the definitive solu- the severity of PHT by reducing arterial supply to
tion. Many types of portosystemic shunt have the portal venous system (Nio et  al. 2003;
been performed, but the distal spleno-renal and Koconis et  al. 2007; Vittorio et  al. 2019).
mesocaval shunts are both reliable options in However, post-embolization syndrome and recur-
children (Fig.  82.11). The mesocaval shunt rent splenomegaly have limited the uptake of this
should be performed with an interposed segment approach.
of autologous jugular vein for optimum results
(Gauthier et al. 1989). It is a non-selective porto- • Lack of access to expert endoscopy.
systemic shunt and, as such, carries a risk of
encephalopathy. The distal spleno-renal (Warren) This is now a rare indication for surgery in
shunt is considered to be a ‘selective’ shunt in portal hypertension.
82  Portal Hypertension
a b
Spleen
Kidney
c d
Fig. 82.11  Portosystemic shunt operations for portal
hypertension. (a) Distal splenorenal, (b) Proximal spleno-
renal, (c) Mesocaval and (d) Side-to-side splenorenal
shunt (modified from Stringer MD and Howard
82.6.5 Other Interventions for Portal
Hypertension
A transjugular intrahepatic portosystemic stent
shunt (TIPS) involves the percutaneous insertion
of an expandable stent between the hepatic and
portal veins in the liver and is performed by an
interventional radiologist (Fig. 82.12). It creates
a non-selective portocaval shunt. TIPS patency
rates in children have steadily improved such that
it is no longer regarded only as a bridge to liver
transplantation (see above) but may be a viable
alternative to a surgical portosystemic shunt in
selected cases (Ling 2020). However, the non-­
selective nature of the TIPS portosystemic shunt
and the lack of long-term data argue for caution
when considering this option over a selective
shunt such as the distal splenorenal shunt.
1135
ER.  Surgical Disorders of the Liver and Bile Ducts and
Portal Hypertension. In: Kelly DA (ed) Diseases of the
Liver and Biliary System in Children. Blackwell
Publishing Ltd., Oxford, 2004)
There are various interventional and surgical
treatments for PHT that are essentially historical
and have little or no place in modern manage-
ment. They were frequently beset by high failure
rates and did not provide a durable solution. They
include esophageal transection with or without
splenectomy and gastric devascularisation
(Uchiyama et  al. 1994) and percutaneous
­transhepatic stenting of the occluded portal vein
(Cwikiel et al. 2003).
82.6.6 Liver Transplantation
Liver transplantation is the treatment of choice for
most children with variceal bleeding complicat-
ing end-stage chronic liver disease. Previous por-
tosystemic shunting does not compromise survival
1136
Fig. 82.12  Transjugular portosystemic shunt (TIPS) in a
child with Budd-Chiari syndrome. In the left venogram
image, a catheter is seen in the occluded hepatic veins
(HV) and a patent portal vein (PV) is visible. In the subse-
after liver transplantation although operative mor-
bidity is less in those who have had a distal sple-
norenal or mesocaval shunt, both of which avoid
complicating the surgical field around the porta
hepatis. TIPS is occasionally indicated for treat-
ing refractory variceal bleeding or ascites in chil-
dren with chronic liver disease awaiting liver
transplantation but it may precipitate hepatic
encephalopathy and is only a bridge to transplan-
tation in this circumstance. Hepatopulmonary and
portopulmonary syndrome are also usually indi-
cations for liver transplantation in PHT (if a meso-
portal bypass is not appropriate and feasible).
82.6.7 Surgery for Budd-Chiari
Syndrome
Rarely, this has a radiologically or surgically treat-
able cause such as a caval web which can be stented
M. D Stringer
quent view the stent is clearly seen extending between the
inferior vena cava just below the right atrium and the por-
tal vein
or resected. Transvenous balloon venoplasty and
anticoagulation are other options in selected cases
(Redkar et al. 2020). Some children with hepatic
vein thrombosis can be successfully managed by
medical therapy directed at controlling ascites and
preventing progressive venous thrombosis.
Portosystemic shunt surgery may be required for
variceal bleeding, deteriorating liver function, and
intractable ascites. Portosystemic shunting con-
verts the portal vein into a venous outflow tract.
TIPS is a less invasive but possibly less durable
alternative to portosystemic shunt surgery although
some centers have reported reasonable short-term
results (Sharma et al. 2016; Allaire et al. 2020). In
chronic Budd-­Chiari syndrome liver transplanta-
tion may be required to treat refractory symptoms
secondary to liver congestion and/or cirrhosis
(Nobre et al. 2017)
Prophylaxis and treatment of veno-occlusive
disease (sinusoidal obstruction) in children
82  Portal Hypertension
Fig. 82.13  Arteriogram demonstrating an intrahepatic
arterio-portal fistula, which was causing portal hyperten-
sion. Contrast passes directly from the hepatic artery into
the intrahepatic portal vein
undergoing bone marrow transplant involves
medical management (Dignan et al. 2013).
82.6.8 Arterioportal Fistula
Selective embolization or liver resection can be
successful if the arterioportal communication is
relatively localized in the liver (Fig. 82.13), but in
complex fistulas, an end-to-side portocaval shunt
(Sutcliffe et  al. 2008) or even liver transplanta-
tion may be required. The latter should be done
before the portal vein becomes too ‘arterialized’.
82.7 Conclusions and Future
Directions
Portal hypertension is a manifestation of numer-
ous different conditions in and around the liver,
most commonly intrinsic chronic liver disease or
extrahepatic portal vein occlusion. The cause of
portal hypertension must be precisely and fully
elucidated in order to select the most appropriate
therapy from the wide variety of available man-
agement options. Bleeding from varices is the
main cause of morbidity and mortality in chil-
dren with portal hypertension. There is a need for
well-designed, adequately powered, randomized
1137
multicenter clinical trials to assess the benefits
and harmful effects of endoscopic treatment ver-
sus beta-blockers for primary prophylaxis of
bleeding from esophageal varices in children
with portal hypertension secondary to intrinsic
liver disease.
Acknowledgements This chapter has been adapted by
permission from Springer Nature: Springer-Verlag,
Berlin. Pediatric Surgery. Diagnosis and Management.
Portal hypertension, Stringer MD, 2009.
References
Allaire M, Walter A, Sutter O, Nahon P, Ganne-Carrié
N, Amathieu R, Nault JC (2020) TIPS for manage-
ment of portal-hypertension-related complications in
patients with cirrhosis. Clin Res Hepatol Gastroenterol
44(3):249–263
Bernabeu JQ, Juampérez J, Mercadal-Hally M et al (2020)
Endoscopic variceal ligation as primary prophy-
laxis for upper GI bleeding in children. Gastrointest
Endosc 92(2):269–275. https://doi.org/10.1016/j.
gie.2020.02.035
Bertocchini A, Falappa P, Grimaldi C, Bolla G, Monti L,
de Ville de Goyet J (2014) Intrahepatic portal venous
systems in children with noncirrhotic prehepatic por-
tal hypertension: anatomy and clinical relevance. J
Pediatr Surg 49:1268–1275
Botha JF, Campos BD, Grant WJ, Horslen SP, Sudan
DL, Shaw BW Jr, Langnas AN (2004) Portosystemic
shunts in children: a 15-year experience. J Am Coll
Surg 199:179–185
Cárdenas AM, Epelman M, Darge K, Rand EB, Anupindi
SA (2012) Pre- and postoperative imaging of the rex
shunt in children: what radiologists should know. AJR
Am J Roentgenol 198(5):1032–1037
Chen VT, Wei J, Liu YC (1992) A new procedure for man-
agement of extrahepatic portal obstruction. Proximal
splenic-left intrahepatic portal shunt. Arch Surg
127:1358–1360
Cwikiel W, Keussen I, Larsson L, Solvig J, Kullendorff
CM (2003) Interventional treatment of children with
portal hypertension secondary to portal vein occlu-
sion. Eur J Pediatr Surg 13:312–318
Deswal S, Madhusudhan KS, Sharma S, Malik R (2020)
Balloon-occluded retrograde Transvenous obliteration
(BRTO): a treatment option in children with gastric
varices. Indian J Pediatr 87(12):1078–1079
di Francesco F, Grimaldi C, de Ville de Goyet J (2014)
Meso-Rex bypass--a procedure to cure prehepatic por-
tal hypertension: the insight and the inside. J Am Coll
Surg 218:e23–e36
Dignan FL, Wynn RF, Hadzic N et  al (2013) BCSH/
BSBMT guideline: diagnosis and management of
veno-occlusive disease (sinusoidal obstruction syn-
1138
drome) following haematopoietic stem cell transplan-
tation. Br J Haematol 163:444–457
Gana JC, Cifuentes LI, Gattini D, Villarroel Del Pino LA,
Peña A, Torres-Robles R (2019) Band ligation versus
beta-blockers for primary prophylaxis of oesophageal
variceal bleeding in children with chronic liver disease
or portal vein thrombosis. Cochrane Database Syst
Rev 9(9):CD010546
Gana JC, Cifuentes LI, Gattini D, Torres-Robles R (2020)
Band ligation versus sclerotherapy for primary pro-
phylaxis of oesophageal variceal bleeding in children
with chronic liver disease or portal vein thrombosis.
Cochrane Database Syst Rev 11:CD011803
Gattini D, Cifuentes LI, Torres-Robles R, Gana JC (2020
Mar 5) Sclerotherapy versus sham or no interven-
tion for primary prophylaxis of oesophageal variceal
bleeding in children with chronic liver disease or
portal vein thrombosis. Cochrane Database Syst Rev
3:CD011573
Gauthier F, De Dreuzy O, Valayer J, Montupet P (1989
Oct) H-type shunt with an autologous venous graft for
treatment of portal hypertension in children. J Pediatr
Surg 24(10):1041–1043
Gharehbaghi MM, Nemati M, Hosseinpour SS, Taei R,
Ghargharechi R (2011) Umbilical vascular catheter
associated portal vein thrombosis detected by ultra-
sound. Indian J Pediatr 78:161–164
Gibelli NE, Tannuri AC, Pinho-Apezzato ML, Maksoud-­
Filho JG, Tannuri U (2011 Jan) Extrahepatic por-
tal vein thrombosis after umbilical catheterization:
Is it a good choice for rex shunt? J Pediatr Surg
46(1):214–216
Guérin F, Porras J, Fabre M, Guettier C, Pariente D,
Bernard O, Gauthier F (2009) Liver nodules after por-
tal systemic shunt surgery for extrahepatic portal vein
obstruction in children. J Pediatr Surg 44:1337–1343
Guérin F, Bidault V, Gonzales E, Franchi-Abella S, De
Lambert G, Branchereau S (2013) Meso-rex bypass
for extrahepatic portal vein obstruction in children. Br
J Surg 100(12):1606–1613
Ha TY, Kim KM, Ko GY, Oh SH, Kwon TW, Cho YP, Lee
SG (2015) Variant meso-rex bypass with transposition
of abdominal autogenous vein for the management of
idiopathic extrahepatic portal vein obstruction: a retro-
spective observational study. BMC Surg 15:116
Koconis KG, Singh H, Soares G (2007) Partial splenic
embolization in the treatment of patients with portal
hypertension: a review of the english language litera-
ture. J Vasc Interv Radiol 18:463–481
Lautz TB, Keys LA, Melvin JC, Ito J, Superina RA
(2013) Advantages of the meso-Rex bypass com-
pared with portosystemic shunts in the management
of extrahepatic portal vein obstruction in children. J
Am Coll Surg. 216(1):83–9. https://doi.org/10.1016/j.
jamcollsurg.2012.09.013
Lautz TB, Sundaram SS, Whitington PF, Keys L,
Superina RA (2009) Growth impairment in children
with extrahepatic portal vein obstruction is improved
by mesenterico-left portal vein bypass. J Pediatr Surg
44:2067–2070
M. D Stringer
Ling SC (2020) TIPS for kids: are we at the tipping point?
J Pediatr Gastroenterol Nutr 70(5):539–540
Long L, Jinshan Z, Zhen C et  al (2018) Portal-to-right
portal vein bypass for extrahepatic portal vein obstruc-
tion. J Pediatr Surg 53:1403–1407
Mack CL, Superina RA, Whitington PF (2003) Surgical
restoration of portal flow corrects procoagulant and
anticoagulant deficiencies associated with extrahe-
patic portal vein thrombosis. J Pediatr 142:197–199
Maksoud-Filho JG, Gonçalves ME, Cardoso SR, Gibelli
NE, Tannuri U (2009) Long-term follow-up of chil-
dren with extrahepatic portal vein obstruction: impact
of an endoscopic sclerotherapy program on bleeding
episodes, hepatic function, hypersplenism, and mor-
tality. J Pediatr Surg 44(10):1877–1883
Mansoor-Ul-Haq M, Latif A, Asad M, Aziz MF (2020)
Treatment of bleeding gastric varices by endoscopic
cyanoacrylate injection: a developing-country per-
spective. Cureus 12(2):e7062
Nio M, Hayashi Y, Sano N, Ishii T, Sasaki H, Ohi R (2003)
Long-term efficacy of partial splenic embolization in
children. J Pediatr Surg 38:1760–1762
Nobre S, Khanna R, Bab N, Kyrana E, Height S, Karani
J, Kane P, Heaton N, Dhawan A (2017) Primary Budd-­
Chiari syndrome in children: King's college hospital
experience. J Pediatr Gastroenterol Nutr 65:93–96
Pimenta JR, Ferreira AR, Fagundes ED, Bittencourt PF,
Moura AM, Carvalho SD (2017) Evaluation of endo-
scopic secondary prophylaxis in children and ado-
lescents with esophageal varices. Arq Gastroenterol
54(1):21–26
Pinto RB, Silveira TR, Bandinelli E, Röhsig L (2004)
Portal vein thrombosis in children and adolescents: the
low prevalence of hereditary thrombophilic disorders.
J Pediatr Surg 39:1356–1361
Ramachandran P, Shanmugam NP, Vij M, Rela M (2014)
Surgical management of hepatic arterioportal fistula in
a neonate. Pediatr Surg Int 30:557–559
Redkar R, Raj V, Tewari S, Sharma R, Joshi S (2020)
Balloon venoplasty as primary modality of treatment
in children with Budd-Chiari syndrome. J Pediatr Surg
55(10):2183–2186
Sari S, Oguz D, Sucak T, Dalgic B, Atasever T (2012)
Hepatopulmonary syndrome in children with cirrhotic
and non-cirrhotic portal hypertension: a single-center
experience. Dig Dis Sci 57:175–181
Sharif K, McKiernan P, de Ville de Goyet J (2010)
Mesoportal bypass for extrahepatic portal vein
obstruction in children: close to a cure for most! J
Pediatr Surg 45:272–276
Sharma VK, Ranade PR, Marar S, Nabi F, Nagral A
(2016) Long-term clinical outcome of Budd-Chiari
syndrome in children after radiological intervention.
Eur J Gastroenterol Hepatol 28:567–575
Shilyansky J, Roberts EA, Superina RA (1999) Distal
splenorenal shunts for the treatment of severe throm-
bocytopenia from portal hypertension in children. J
Gastrointest Surg 3(2):167–172
Shneider BL (2017) Tacit consensus. J Pediatr
Gastroenterol Nutr 64:497
82  Portal Hypertension
Shneider BL, de Ville de Goyet J, Leung DH et al (2016)
Primary prophylaxis of variceal bleeding in children
and the role of MesoRex bypass: summary of the
Baveno VI pediatric satellite symposium. Hepatology
63:1368–1380
Srivastava A, Yadav SK, Lal R, Yachha SK, Thomas MA,
Saraswat VA, Gupta RK (2010 Dec) Effect of surgi-
cal portosystemic shunt on prevalence of minimal
hepatic encephalopathy in children with extrahe-
patic portal venous obstruction: assessment by mag-
netic resonance imaging and psychometry. J Pediatr
Gastroenterol Nutr 51(6):766–772
Stringer MD (2005) Esophageal varices. In: Najmaldin
AS, Rothenberg S, Crabbe DCG, Beasley S (eds)
Operative endoscopy and endoscopic surgery in infants
and children. Hodder Arnold, London, pp 141–148
Stringer MD (2006) Portal hypertension. In: Stringer
MD, Oldham KT, Mouriquand PDE (eds) Pediatric
surgery and urology: Long term outcomes, 2nd edn.
Cambridge University Press, New York, pp 491–509
Stringer MD (2007) Improved body mass index
after mesenterico-­ portal bypass. Pediatr Surg Int
23:539–543
Stringer MD, Heaton ND, Karani J, Olliff S, Howard ER
(1994) Patterns of portal vein occlusion and their aeti-
ological significance. Br J Surg 81:1328–1331
Sutcliffe R, Mieli-Vergani G, Dhawan A, Corbally M,
Karani J, Heaton N (2008) A novel treatment of con-
genital hepatoportal arteriovenous fistula. J Pediatr
Surg 43:571–573
Tingo J, Rosenzweig EB, Lobritto S, Krishnan US (2017)
Portopulmonary hypertension in children: a rare but
potentially lethal and under-recognized disease. Pulm
Circ 7:712–718
1139
Uchiyama M, Iwafuchi M, Ohsawa Y, Naito S, Naito M,
Yagi M, Tsukada K (1994) Long-term results after
nonshunt operations for esophageal varices in chil-
dren. J Pediatr Surg 29(11):1429–1433
Valayer J, Branchereau S (1998) Portal hypertension:
portosystemic shunts. In: Stringer MD, Oldham KT,
Mouriquand PDE et  al (eds) Pediatric surgery and
urology: Long-term outcomes. Saunders, Philadelphia,
PA, pp 439–446
de Ville de Goyet J, Clapuyt P, Otte JB (1992) Extrahilar
mesenterico-left portal shunt to relieve extrahepatic
portal hypertension after partial liver transplant.
Transplantation 53:231–232
de Ville de Goyet J, Alberti D, Clapuyt P et al (1998) Direct
bypassing of extrahepatic portal venous obstruction in
children: a new technique for combined hepatic portal
revascularization and treatment of extrahepatic portal
hypertension. J Pediatr Surg 33:597–601
Vittorio J, Orellana K, Martinez M, Ovchinsky N,
Schlossberg P, Griesemer A, Lobritto S (2019) Partial
splenic embolization is a safe and effective alternative
in the Management of Portal Hypertension in children.
J Pediatr Gastroenterol Nutr 68(6):793–798
Zargar SA, Javid G, Khan BA, Yattoo GN, Shah AH,
Gulzar GM et al (2002) Endoscopic ligation compared
with sclerotherapy for bleeding esophageal varices in
children with extrahepatic portal venous obstruction.
Hepatology 36:666–672
Zhang JS, Li L, Cheng W (2018) Surgical treatment
for rebleeding caused by bypass failure after rex
shunt: re-rex shunt or Warren shunt? Pediatr Surg Int
34(5):521–527
 Gallbladder Disease
Charlene Dekonenko, Shawn D. St. Peter,
and George W. Holcomb III
83.1 Introduction
Over the last several decades, the incidence of
gallbladder disease in children has increased,
although it is still not seen nearly as often as in
adults. While the development of cholelithiasis is
the primary cause of gallbladder disease in both
adults and children, there are some differences in
the underlying disease processes contributing to
the gallbladder pathology. There is a much higher
incidence of pigmented gallstones due to hemo-
lytic disease in children.(Holcomb et  al. 1980)
Moreover, acute and chronic cholecystitis with
severe inflammation and/or scarring of the gall-
bladder and surrounding tissues are far less com-
mon in children. However, the increase in
gallbladder disease in children is likely due to the
rise in non-hemolytic cholelithiasis as a result of
the Western diet (similar to the pathogenesis seen
in adults), and also due to increased recognition
of biliary dyskinesia as a cause of gallbladder
disease in this age group (Walker et  al. 2013;
Khoo et al. 2014; Lacher et al. 2013).
Regardless of the underlying disease process,
children with gallbladder disease present with
similar symptoms. This primarily consists of
abdominal pain in the right upper quadrant or
C. Dekonenko · S. D. St. Peter
G. W. Holcomb III (*)
Children’s Mercy Hospital, Kansas City, MO, USA
e-mail: sspeter@cmh.edu; gholcomb@cmh.edu
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_83
83
mid-upper abdomen but can be vague and it often
occurs after fat-laden meals. In addition, some
patients report associated nausea and vomiting.
This constellation of symptoms is referred to as
biliary colic and can be seen in children with cho-
lelithiasis or biliary dyskinesia, which often
makes it difficult to determine the etiology of
their symptoms.
83.2 Etiology
As mentioned, gallbladder disease in children can
arise from a range of underlying conditions. In
most cases, gallbladder disease is related to the
development of cholelithiasis. Gallstones in chil-
dren are usually classified as being either hemo-
lytic or non-hemolytic in etiology. Hemolytic
disease results in consumption of red blood cells
leading to increased liver metabolism of bilirubin
which precipitates stone formation. Non-­hemolytic
disease includes a variety of causes for stone devel-
opment or symptomatic gallbladder disease with-
out stone formation. Stone formation is a complex
interaction of several components centered around
the lithogenic potential of a given individual.
83.2.1 Gallstone Formation
The major chemical components of bile, which
contribute to its lithogenic potential, are bile
1141
1142
salts, phospholipids, cholesterol, bilirubin, and
electrolyte/water balance. The phospholipid
component is mostly lecithin which, along with
bile salts, serves as detergents in the bile. With
polar and non-polar portions to these molecules,
they form lecithin–bile acid–cholesterol micelles
that keep the cholesterol soluble within the
hydrophobic center of the micelle. An imbalance
in the concentration of these substances is almost
always due to an increase in cholesterol ­secretion,
which results in cholesterol crystal precipitation.
These crystals serve as the nidus for further pre-
cipitation resulting in macroscopically detectable
gallstones. Thus, excess cholesterol causing the
supersaturation of bile leads to cholesterol stones.
This pathway accounts for the majority of adult
gallstones but is less common in children.
Cholesterol gallstones are extremely rare in pre-
pubertal children. However, obesity is a risk fac-
tor for cholesterol stone formation and with the
onset of obesity now occurring as early as child-
hood, cholesterol stone formation in children is
becoming more frequent (Kaechele et  al. 2006;
Liem et al. 2020).
The physiologic components contributing to
stone formation include gallbladder emptying,
bile duct emptying, as well as the integrity and
function of the biliary epithelium. Epithelial
mucin production accelerates stone formation.
Epithelial inflammation, which can result from
cholesterol supersaturation, alters the electrolyte/
water balance and leads to the epithelial release
of pro-crystallizing proteins (Moody et al. 1989).
Furthermore, the bacterial colonization of biliary
epithelium can cause deconjugation of bilirubin
leading to stone formation.
83.3 Hemolytic Disease
The other common form of cholelithiasis in chil-
dren is pigmented stones that form as a result of
hemolytic diseases such as sickle cell anemia,
hereditary spherocytosis, and thalassemias,
which account for about 15% (Walker et  al.
2013). Red blood cells are composed of a plasma
membrane, the hemoglobin moiety, and cytoplas-
mic enzymes. These 3 basic components provide
C. Dekonenko et al.
Table 83.1  The major hereditary hemolytic diseases
Membrane defects; hemoglobin defects; enzyme defects
Spherocytosis; sickle cell disease; glucose-6-phosphate
deficiency
Elliptocytosis; α thalassemia; pyruvate kinase deficiency
Pyropoikilocytosis; β thalassemia
Hydrocytosis
Xerocytosis
a functional outline for the major hereditary
hemolytic diseases (Table  83.1). Their constant
turnover requires the reticuloendothelial system
to retrieve dying and dead red blood cells leading
to the breakdown of the hemoglobin moiety to
bilirubin. A nearly insoluble molecule, bilirubin
requires conjugation by glucuronyl transferase to
produce bilirubin diglucuronide, the molecule
measured as “direct” bilirubin, which is more
soluble. The enzymatic process of conjugation is
saturable. Thus, the hemolytic states listed in
Table  83.1 may cause an abnormal level of
unconjugated (indirect) bilirubin in the bile. This
results in calcium bilirubinate formation, which
polymerizes with bilirubin to form pigmented
gallstones. This interaction is not linear but is
affected by the physiologic function of the gall-
bladder, the biliary system, and the other vari-
ables of bile lithogenesis described previously.
Fifty percent of patients with sickle cell ane-
mia develop gallstones by 20  years of age
(Lachman et  al. 1979). The abnormal form of
hemoglobin in sickle cell anemia allows red
blood cells to adhere to one another creating cell
masses that, along with the abnormal red blood
cells, are filtered by the spleen. This high turn-
over results in the development of gallstones at a
very early age. These patients have the risk of
going into a sickle cell crisis, which can be fatal.
Dehydration, pain, and systemic inflammation
can increase this risk, and therefore, improved
operative outcomes are dependent on adequate
hydration and blood transfusion to an acceptable
hemoglobin level of 10 mg/dl, sometimes requir-
ing preoperative admission to achieve these goals
(Rhodes et  al. 2014). A recent retrospective
review reported decreased morbidity for elective
cholecystectomy in patients with sickle cell ane-
mia compared to children undergoing an emer-
83  Gallbladder Disease
gent operation (Goodwin et al. 2017). Thus, it is
prudent to perform cholecystectomy, even for
asymptomatic cholelithiasis in children with
sickle cell disease, as cholecystitis, cholangitis,
and pancreatitis can be fatal in these patients.
The incidence of cholelithiasis with hereditary
spherocytosis is also around 50% and slightly
more common in girls than boys. If a patient is
going to undergo splenectomy, an ultrasound
(US) should be performed to evaluate for
­cholelithiasis prior to the splenectomy and the
gallbladder should be removed at the same time
as splenectomy if gallstones are identified.
However, if gallstones are not identified, a chole-
cystectomy is not indicated (Sandler et al. 1999).
In the absence of stones, the splenectomy should
decrease the possibility of future pigment stone
development due to decreased hemolysis follow-
ing splenectomy.
Previously, thalassemia patients were also at
high risk for cholelithiasis. However, the inci-
dence has decreased due to more aggressive
transfusion management, which prevents the pro-
duction and release of native red cells with the
defective hemoglobin.
83.4 Non-Hemolytic Disease
Conjugated bile salts are deconjugated by intesti-
nal bacterial and resorbed, mostly in the terminal
ileum. Through the portal circulation, they return
to the hepatocytes for biliary excretion. Patients
with short gut syndrome, ileal resection, or ileal
disease (Crohn disease) are at risk for stone
development due to the decreased bile salt pool.
Also, patients with bowel dysmotility or dysfunc-
tion may have altered bacterial flora which also
affects the enterohepatic circulation.
The most common condition associated with
the development of non-hemolytic cholelithiasis
in neonates and infants is the use of total parenteral
nutrition (TPN). Up to 40% of children receiving
long-term TPN will develop gallstones (King et al.
1987; Matos et  al. 1987). While many patients
requiring long-term TPN have gastrointestinal dis-
ease, the complete picture of TPN-­associated cho-
lestasis, liver disease, and gallstone formation is
1143
not completely understood. Decreased bile flow
and gallbladder emptying from a lack of enteral
stimulation has been postulated to be an important
contribution to TPN-­ associated gallstones.
However, in one study, the use of cholecystokinin
(CCK) to prevent stone development in TPN-
dependent children had no effect, implying that
the role of CCK-mediated bile flow may be less
important than previously thought in gallstone for-
mation for these patients. TPN has a primary litho-
genic effect on bile causing increased bilirubin and
calcium concentrations. These effects have been
shown to be prevented with glutamine supplemen-
tation, suggesting there are intermediary steps that
still need to be clarified. Septicemia, dehydration,
and chronic furosemide therapy have also been
implicated as contributing factors to lithogenesis
in TPN-dependent patients (Matos et  al. 1987;
Schweizer et al. 2000).
Cystic fibrosis, the phenotype of defective epi-
thelial chloride channels, results in decreased
transport of water and chloride, which increases
the viscosity of bile contributing to stone forma-
tion. Also, it can lead to obstruction of biliary
ductules leading to liver failure.
Finally, cholelithiasis has been reported in
children undergoing cardiac bypass, cardiac
transplantation, and in patients who have previ-
ously undergone extracorporeal membrane oxy-
genation as a newborn (Schweizer et al. 2000).
83.4.1 Acalculous Gallbladder
Disease
The gallbladder can be a source of symptoms in
patients without gallstones. Acute inflammatory
attacks of the gallbladder, called acute acalculous
cholecystitis, can result from bile stasis, ischemia
(or both), and occur more commonly in associa-
tion with severe illness such as sepsis, burns, or
trauma (McChesney et  al. 2003). Symptoms of
biliary colic are common and the diagnosis is
made with an US, showing gallbladder wall
thickening, edema, and pericholecystic fluid in
the absence of gallstones. This entity is much less
common in children than adults, but can be seen
in critically ill children.
1144
Biliary dyskinesia, or dysfunctional emptying
of the gallbladder, is the most common gallblad-
der condition that does not involve stones, and its
recognition is partially responsible for the
increasing incidence of gallbladder disease in
children (Walker et  al. 2013; Khoo et  al. 2014;
Lacher et al. 2013). However, it is unclear if bili-
ary dyskinesia is a primary entity or part of other
abdominal pain syndromes such as dyspepsia or
irritable bowel syndrome (Friesen et  al. 2011;
Rau et al. 2006). While the etiology of the symp-
toms is felt to be gallbladder distension second-
ary to poor emptying, bile stasis can promote
sludge, microscopic bile crystallization, and sub-
sequent mucosal irritation. On pathologic exami-
nation, chronic cholecystitis is often documented
on histologic examination of the gallbladder
specimen. Laboratory tests and an abdominal US
will be normal in biliary dyskinesia. Thus, the
diagnosis is usually based on the gallbladder
ejection fraction calculated following a CCK-­
stimulated hepatobiliary iminodiacetic acid
(HIDA) scan. A gallbladder ejection fraction of
less than 35% is considered to be diagnostic for
biliary dyskinesia (Srinath et al. 2014). The treat-
ment is laparoscopic cholecystectomy but data on
long-term symptom resolution for these patients
are mixed (Lai et al. 2017; St Peter et al. 2008;
Knott et al. 2013). No reliable preoperative fac-
tors for symptom relief following cholecystec-
tomy have been identified so it is imperative to
emphasize during preoperative counseling that
symptoms may not completely resolve following
the cholecystectomy (Carney et  al. 2004).
Recently, reports of patients with symptoms of
biliary dyskinesia with an abnormally high gall-
bladder ejection fraction have been published. In
these small series, symptom relief following cho-
lecystectomy appears to be similar to those with
standard biliary dyskinesia and should therefore
be treated similarly (Dekonenko et  al. 2019).
Overlapping with biliary dyskinesia is sphincter
of Oddi hypertension as a cause of chronic acal-
culous symptomatic gallbladder disease.
However, this condition is quite rare in children.
Hydrops is characterized by massive disten-
tion of the gallbladder in the absence of stones,
infection, or congenital anomalies. It has been
C. Dekonenko et al.
most frequently reported in association with
Kawasaki disease and is usually due to transient
obstruction of the cystic duct or to increased
mucus secretion by the gallbladder resulting in
poor emptying. With additional gallbladder dis-
tention, further angulation of the cystic duct may
increase the obstruction.
Gallbladder polyps are unusual in children.
Surgical recommendations have followed the
adult literature, which suggests that cholecystec-
tomy should be performed if there are symptoms
or if the polyp is ≥1  cm in size (Stringer et  al.
2003).
83.5 Diagnosis
83.5.1 Clinical Features
The diagnosis of gallstones in young children
may be delayed due to vague and non-specific
symptoms such as abdominal pain, nausea, and
emesis. Abdominal pain may vary from the typi-
cal right upper abdominal location due to poor
localization of the pain by young children. Older
children and teenagers present with symptoms
and physical findings similar to those in adults.
Right subcostal pain, with or without radiation to
the subscapular area or central back, is the most
common complaint. The pain typically fluctuates
with intensity over a background level of pain
due to gallbladder contractions against the cystic
duct obstruction. Nausea and vomiting com-
monly occur following the onset of pain. A his-
tory of precipitating fatty food ingestion can
often be elicited.
Patients with symptoms of biliary colic will
usually experience resolution of symptoms
within hours and almost always resolve within a
day. After the acute attack abates, patients will
often describe a dull ache or soreness in the right
upper quadrant for the next several days. These
patients usually experience tenderness with deep
palpation. Murphy sign, more indicative of acute
cholecystitis, is the hesitation or inability to com-
plete a deep inhalation when steady, gentle pres-
sure is applied in a superior and posterior
direction under the right costal margin. Patients
83  Gallbladder Disease
with acute cholecystitis present similarly, but will
usually appear more systemically ill. Moreover,
they are more likely to have a fever and demon-
strate leukocytosis. In these patients, the diagno-
sis is confirmed by inflammatory changes on
imaging studies (see below).
83.5.2 Radiologic Diagnosis
Stones can be radio-opaque in up to 20% of
patients with hemolytic disorders and in 15% of
patients with cholesterol stones (Fig.  83.1a).
Thus, abdominal radiographs are not usually a
useful diagnostic or screening tool for gallstones
but may be considered if symptoms were more
concerning for constipation or another etiology.
The anatomic appearance of the gallbladder and
its contents is best evaluated by US, which has a
sensitivity and specificity >95% for detecting
gallstones, although its accuracy is lower in chil-
dren than adults (Fig.  83.1b) (Cooperberg and
Burhenne 1980; Tsai et al. 2013). It also allows
the evaluation of the hepatic ducts and common
bile for stones or obstruction as well as the evalu-
ation of other abnormalities in the liver or pan-
creas. Though a computed tomography (CT) scan
can define the gallbladder inflammation and
a b
Fig. 83.1  In a small number of patients, gallstones may be radiopaque (a). Ultrasound is a very accurate imaging study
for gallbladder disease (b). Note the acoustic shadowing that develops behind the gallstones on this study
1145
stones, delineation of the common bile duct is
suboptimal with a routine abdominal CT and
exposes children to unnecessary radiation (Liem
et al. 2020). If the ultrasound reveals inflamma-
tory changes, as evidenced by gallbladder wall
thickening, pericholecystic fluid, or tenderness
elicited by probe placement directly over the
gallbladder (sonographic Murphy sign), the diag-
nosis of acute cholecystitis can be made. Stones
without inflammatory changes should be associ-
ated with symptoms for the diagnosis of symp-
tomatic cholelithiasis.
For a patient with symptoms and exam find-
ings consistent with acute cholecystitis but with-
out stones on US, cholescintigraphy (HIDA scan)
should be performed. 99mTc-labeled iminodiace-
tic acid (IDA) or an analog is injected peripher-
ally and is taken up by hepatocytes. The analog is
processed by the same mechanism as bilirubin
and is excreted in bile. If the biliary tree is illumi-
nated without filling the gallbladder, then the
diagnosis of acute cholecystitis is confirmed.
Occasional false positives occur in critically ill
fasting patients and intravenous morphine may
be useful to enhance the visualization of the gall-
bladder. The morphine constricts the sphincter of
Oddi, which enhances filling of the gallbladder.
When the gallbladder fills with radioactive sub-
1146
strate, emptying can then be quantified by calcu-
lating the ejection fraction of the gallbladder with
cholecystokinin-assisted or lipomul challenge.
As previously mentioned, the normal ejection
fraction approximates 75% and patients with an
ejection fraction under 35% are considered to
have biliary dyskinesia.
83.6 Management
The gallbladder acts as a reservoir to provide a
bolus of bile for digestion, especially with large
fatty meals. Because the need for a functioning
gallbladder is marginal, it can be removed with
few long-term consequences. Historically, gall-
stones have been treated with non-operative ther-
apies including oral dissolution agents, direct
instillation of dissolution agents, and extracorpo-
real shockwave lithotripsy. These measures have
been abandoned due to failure, recurrence, and
high cost. Thus, once a patient presents with
symptoms, the risk of complications from their
gallbladder disease is generally higher than the
risks of laparoscopic cholecystectomy. Patients
with symptomatic gallbladder disease who are
older than 3  years should undergo laparoscopic
cholecystectomy. The treatment of younger
patients, particularly infants, should be individu-
alized. Infant gallstones secondary to prolonged
TPN have been reported to dissolve spontane-
ously (St-Vil et  al. 1992). Therefore, in the
absence of complications, these young patients
should be observed for 6–12  months following
the cessation of the TPN and the initiation of
enteral alimentation.
When compared to the open operation, the
laparoscopic approach provides less discomfort,
reduces the length of postoperative hospitaliza-
tion, improves cosmesis, and results in a much
faster return to routine activities such as work,
school, play, or participation in athletic activity.
Although most children can be discharged
the day of surgery or certainly within 24 hours,
as previously mentioned, patients with sickle
cell disease require special preoperative care to
prevent postoperative complications. Several
authors have reported favorable results with
C. Dekonenko et al.
c­holecystectomy in this patient population and
have emphasized the need for preoperative trans-
fusion (Bhattacharyya et  al. 1993; Haberkern
et  al. 1997; Sandoval et  al. 2002). The laparo-
scopic approach is not more hazardous for these
patients and is therefore preferred (Wales et  al.
2001). Because gallbladder sludge is frequently
documented in patients with sickle cell anemia,
elective cholecystectomy has been recommended
when evidence suggests the presence of sludge,
with or without stones (Lachman et  al. 1979).
In one study of 35 patients with sickle cell dis-
ease and biliary sludge, 23 (65.7%) went on to
develop gallstones.
Patients presenting with an acute episode of
cholecystitis and signs of inflammation on labo-
ratory or radiologic studies can be managed with
either semi-urgent laparoscopic cholecystectomy
or antibiotics followed by interval cholecystec-
tomy. In adult studies, no difference in operative
complications between early and delayed chole-
cystectomy has been found, but delayed chole-
cystectomy has been associated with more
complications from the disease (relapse, choled-
ocholithiasis, pancreatitis). Early intervention
has also been shown to be much more cost-­
effective than interval cholecystectomy.
Because gallbladder polyps are an unusual
finding in children, the natural history of these
polyps is unclear. Due to the inability to assure
life-long follow-up combined with the extremely
poor prognosis when gallbladder cancer devel-
ops, a laparoscopic cholecystectomy is a reason-
able option for symptomatic children with
gallbladder polyps or for patients with a polyp
≥1 cm.
83.6.1 Concomitant Splenectomy
If a patient is going to undergo splenectomy due
to hematologic disease, an US should be per-
formed to evaluate for the presence of gallstones
prior to surgery as it is relatively straightforward
to remove the gallbladder at the same time if
stones are identified. A splenectomy should not
be performed by the open approach except in
extraordinary circumstances, and the cholecys-
83  Gallbladder Disease
tectomy can be easily performed at the time of
laparoscopic splenectomy. If gallstones are not
identified on preoperative US, then a cholecys-
tectomy is not indicated as removal of the spleen
will eliminate the source of pigmented stones. In
a study of 17 patients with hereditary spherocyto-
sis but no cholelithiasis, none developed
­gallstones at a mean follow-up of 15 years, fur-
ther emphasizing that prophylactic cholecystec-
tomy at the time of splenectomy is not necessary
(Sandler et al. 1999).
83.6.2 Choledocholithiasis
A variety of management strategies exist for
patients who present with choledocholithiasis as
suggested by symptoms such as jaundice, dark
urine, and acholic stools, or by US findings such
as dilated ducts or common duct stones. If endo-
scopic retrograde cholangiopancreatography
(ERCP) is available, then sphincterotomy and
stone extraction (if stones are found) can be per-
formed either preoperatively or postoperatively
(Mah et al. 2004; Cuschieri et al. 1999; Newman
et al. 1997; Zargar et al. 2003). Alternatively, one
can proceed with laparoscopic cholecystectomy
with cholangiography and attempted extraction
of the stones if seen on cholangiography
(Cuschieri et  al. 1999). If stones are found and
cannot be extracted, then either laparoscopic or
open choledochal exploration is performed, or
postoperative ERCP and sphincterotomy are
attempted.
The approach is primarily influenced by the
surgeon’s experience with laparoscopic chole-
dochal exploration, but also by the availability of
an endoscopist experienced in ERCP in children.
However, many pediatric gastroenterologists are
not trained in this technique and many children’s
hospitals utilize an adult gastroenterologist from
another hospital. For the majority of pediatric
surgeons, the best option may be to perform a
preoperative ERCP with sphincterotomy and
stone extraction if stones are located. If success-
ful, the surgeon can then proceed with laparo-
scopic cholecystectomy. However, if stones are
1147
found and cannot be extracted at ERCP and
sphincterotomy, then the surgeon will know
whether or not laparoscopic or open choledochal
exploration is indicated at the time of the chole-
cystectomy. This latter approach is preferred at
our institution (Fig. 83.2).
If common duct stones are found on intra-­
operative cholangiography, flushing the common
duct with saline through the cholangiocatheter
may allow them to pass. This procedure can be
augmented by the intravenous injection of 1 mg
of glucagon to relax the sphincter of Oddi and
can be repeated as needed. If glucagon is not
available, naloxone and nitroglycerin have also
been used successfully. Also, the cholangiocath-
eter can be gently passed into the common duct
under fluoroscopy and used to push small stones
through the sphincter of Oddi. A choledocho-
scope can also be passed through the cystic duct
and into the common duct to push the stones
through the sphincter. If a choledochoscope is
unavailable, a standard ureteroscope is the same
size (7 mm diameter) and is efficacious as well. If
a stone is not amenable to any of these tech-
niques, it is likely to be stuck in the duct and will
require basket retrieval. Currently, this is gener-
ally safer if approached by ERCP, which can be
performed postoperatively. All of the aforemen-
tioned laparoscopic techniques can be performed
through the cystic duct.
Intra-operative cholangiography at the time of
the laparoscopic cholecystectomy is an area of
debate when treating adult patients. However, in
children, where choledocholithiasis is far less
common, there is no evidence for its routine use,
and many surgeons do not support routinely
increasing the operative time, cost, and risk of the
procedure for a large population of patients in
whom it is not necessary (Wu et  al. 2005).
Contrary to individual, non-comparative series,
review of complications leading to litigation has
suggested that routine cholangiography does not
reduce complications (McLean 2006). Therefore,
in children, cholangiography should be used
selectively in patients in whom the anatomy is
not clear or in whom there is the suspicion of
choledocholithiasis.
1148 C. Dekonenko et al.

Fig. 83.2 This
algorithm depicts our Suspected Choledocholithiasis
current management for
suspected
choledocholithiasis.
Preoperative ERCP is
preferred so that
sphincterotomy and
ERCP
stone removal can be
performed prior to the
laparoscopic
cholecystectomy. With
this management
strategy, the surgeon will
know whether or not Stone(s) Confirmed No Stones
laparoscopic or open
choledochal exploration
will be required at the
time of the laparoscopic
cholecystectomy

Endoscopic Sphincterotomy
&
Stone Removal

Not
Successful Successful

Laparoscopic or Open
Cholecystectomy and Laparoscopic
Common Duct Exploration Cholecystectomy

83.6.3 Gallstone Pancreatitis
Children, like adults, who present with pancreati-
tis and who are found to have sludge or stones in
the gallbladder or common duct are considered to
have gallstone pancreatitis. The majority of cases
are mild, and the stone usually passes within
48  hours without the need for preoperative
ERCP.  Clinical improvement is indicated by an
acute and dramatic decrease in the patient’s
symptoms when the stone passes after which the
abnormal laboratory values usually decline
sharply over the next several days. Laparoscopic
cholecystectomy can be safely performed when
the pain is improved and the labs are trending
down without waiting until they normalize. These
patients should undergo cholecystectomy prior to
discharge due to the high risk of recurrent pan-
creatitis if these patients are scheduled for a later
elective operation.
In the unusual scenario where the pancreatitis
is severe with systemic manifestations and pan-
creatic complications (such as a pseudocyst),
ERCP should be initially performed with sphinc-
terotomy. In this situation, laparoscopic chole-
cystectomy can then be performed later when the
patient has recovered, and even at the time of
intervention for the pseudocyst. We have had suc-
83  Gallbladder Disease 1149

cess with concomitant laparoscopic cyst-­ general, particularly in children, if a 5 mm clip is
gastrostomy and cholecystectomy in this not of sufficient size for the cystic duct, then the
situation. anatomy should be re-examined starting from the
The most common etiology for pancreatitis in gallbladder to assure this structure is, indeed, the
children is idiopathic. However, if the idiopathic cystic duct. The other 2 instruments are posi-
pancreatitis recurs a second time, these patients tioned on the right side of the abdomen in varying
should be evaluated with an ERCP to look for an configurations for retracting the fundus of the
anatomic cause. If none is found, a sphincterot- gallbladder cephalad and the infundibulum later-
omy or cholecystectomy should be considered ally. These instruments can be introduced with-
due to the high possibility of microlithiasis or out a port via a stab incision if the patient is not
biliary sludge as the underlying cause. obese as these 2 instruments are not exchanged
Ursodeoxycholic acid therapy has been reported during the operation (Fig. 83.3). This technique
to be successful in preventing recurrence by allows 3 or 5  mm instruments to be introduced
decreasing the viscosity of the bile, but this man- through extremely small incisions in lean or
agement strategy requires ongoing therapy and younger patients (Ostlie and Holcomb 2003). We
does not hold much promise for children. position one of these ports in the midline between
Therefore, children with a second episode of the 2 midline ports so the surgeon does not reach
idiopathic pancreatitis are candidates for sphinc- across the patient, and the other in the right mid
terotomy or cholecystectomy. abdomen to retract the fundus cephalad. Adequate
cephalad retraction of the fundus and lateral
retraction of the infundibulum should allow good
83.7 Operative Considerations visualization of the triangle of Calot for dissec-
tion without the need for changing patient posi-
83.7.1 Laparoscopic tion (Fig. 83.4). Because the duodenum is fixed,
Cholecystectomy moving the table has little impact on improving
exposure. Blunt dissection of the cystic duct
The movement toward minimally invasive sur- should start at the level of the gallbladder to
gery began with the laparoscopic approach for assure accurate identification of the cystic duct
choleystectomy (Holcomb 1993; Reynolds 2001; regardless of how short or anomalous its course.
Dubois et  al. 1989; Reddick and Olsen 1989; It is important to visualize the critical view of
McKernan 1991; Holcomb et  al. 1991). While
multiple variations are described and utilized, the
standard technique is as follows: The patient is
placed supine on the operating table and two
video monitors are positioned at the head of the
table. Historically, an orogastric tube has been
inserted for decompression of the stomach but
this not necessary, nor is a urinary catheter
(Holcomb et al. 1999). The surgeon stands on the
patient’s left side for the four-­ port technique.
Generally, a 10 mm port in the umbilicus is the
only large port necessary in order to remove the
gallbladder, but a 5 mm telescope can be used to
avoid switching scope sizes later. A 5 mm port,
inserted in the epigastrum just to the right of mid- Fig. 83.3  This photograph shows placement of the ports
line, becomes the main operating site for the sur- for laparoscopic cholecystectomy in a 9-year-old girl with
biliary dyskinesia. The two right-sided 3 mm instruments
geon, allowing for the endoscopic clips to be have been placed directly through the abdominal wall
applied to the cystic duct and cystic artery. In without the use of cannulas
1150
safety to correctly identify the anatomy. This
critical view is bounded by the CBD medially,
the cystic duct inferiorly, the gallbladder later-
ally, and the liver superiorly (Strasberg et  al.
1995; Vettoretto et al. 2011; Strasberg and Brunt
2010). The cystic duct should be ligated with
endoscopic clips and divided between the clips so
as to avoid CBD injury (Fig.  83.5). The cystic
artery is ligated and divided in a similar fashion.
However, if the artery is taken distally enough,
usually as it enters the gallbladder, sometimes
after dividing into anterior and posterior branches,
Fig. 83.4  In order to create a 90° angle between the cys-
tic duct and common duct, it is important to laterally
retract the infundibulum. Lateral retraction of the infun-
dibulum aids in accurate identification of the cystic duct
(solid arrow) and common bile duct (asterisk). Note the
cystic artery (dotted arrow) in this photograph
a b
Fig. 83.5  The cystic duct is ligated so that two clips will
remain on the cystic duct stump (a). In a similar fashion,
the cystic artery is ligated and divided (b). Note the two
C. Dekonenko et al.
it can be divided with cautery. This is done by
grasping with the Maryland dissecting instru-
ment to cauterize, then dividing it away from the
CBD with monopolar cautery. Following ligation
of the cystic artery and duct, the gallbladder is
dissected from the liver bed with cautery. Prior to
complete mobilization from the liver, the area of
dissection should be inspected to ensure hemo-
stasis, and then the gallbladder is completely
detached. The specimen can be extracted through
the umbilical incision without the use of an endo-
scopic bag if there is little to no inflammation.
83.7.2 Single-Site Laparoscopic
Cholecystectomy
Shortly after surgeons began performing opera-
tions by the single incision approach, cholecys-
tectomy was an operation for which it was
commonly utilized (Garey et al. 2011). For this
approach, all the instrumentation is placed
through the umbilicus, resulting in an incision
approximately 2  cm in length. In the United
States (U.S.), a premanufactured port is often uti-
lized. The two main devices used are the SILS
Port (Covidien, Norwalk CT) or the TriPort
(Olympus America, Center Valley PA). The SILS
Port is a foam port with three working channels.
A 3 mm or 5 mm instrument can then be placed
along the left side of the port (Fig.  83.6). It is
clips (arrow) remaining on the cystic duct stump. The
common bile duct is marked with an asterisk in both
photographs
83  Gallbladder Disease 1151

Table 83.2  Operative data comparing patients undergo-

ing laparoscopic cholecystectomy via a single-incision
approach vs a four-port technique
Single
Incision 4-port
(N = 30) (N = 30) p-value
Operative time 68.6 ± 22.1 56.1 ± 22.1 0.03
(min)
Surgical 2.7 ± 1.0 1.9 ± 0.8 0.005
difficulty (1:
Easy to 5:
Difficult)
From: Ostlie DJ, Adibe OO, Juang D, et al. Single incision
versus standard 4-port laparoscopic cholecystectomy: A
prospective randomized trial. J Pediatr Surg
2013;48:209–14

Fig. 83.6  In the U.S., a premanufactured port can be uti-

postoperative analgesia as well as accruing more
lized for a single-site umbilical laparoscopic cholecystec-
tomy. A SILS Port (Covidien, Norwalk, CT) has three significant hospital charges. Since the main
advantage purported by many surgeons was
working channels and a fourth instrument can be inserted
along the side of the port superior cosmetic results, the patients from this
trial were followed in the short and long term
helpful to have a long telescope so that the assis- with a validated scar assessment tool. The results
tant holding the telescope is standing out of the showed a marginal advantage in early follow-up,
way of the operating surgeon. Many surgeons and also a very small effect size at a minimum
place a single port in the umbilicus with instru- 18 months after the operation (Gasior et al. 2014).
ments inserted through the fascia surrounding the Multiple randomized trials and comparative
port. Sometimes, low profile, 5  mm individual series have been performed in adults, largely con-
ports are utilized. Regardless of the technique firming these findings with longer operative times
and orientation of the instruments through the with single incision and little other differences.
umbilicus, the principles of the procedure are the Several meta-analyses have come to the same
same as for the traditional four-port laparoscopic conclusions. In a Cochrane review of 9 random-
cholecystectomy. It can be difficult to reinsert ized trials with 427 single-site patients compared
these ports so it is best to assure hemostasis and to 428 four-port patients, those undergoing the
irrigate/suction, if necessary, prior to extracting single-site approach had longer operating times
the gallbladder. with low quality evidence to support any benefit
After initial case series were reported, the while the safety profile remains unproven
single-­site approach garnered a lot enthusiasm. (Gurusamy et al. 2014). Interestingly, this meta-­
The largest prospective, randomized trial in chil- analysis also concluded no cosmetic benefit.
dren enrolled 60 children to either a single-­ While the current data suggest no advantages for
incision laparoscopic cholecystectomy or the the single-site approach, the takeaway message is
standard four-port technique (Ostlie et al. 2013). that any advantage potentially gained from the
The results showed that children who underwent single-site approach is small if it exists. Therefore,
a single-incision laparoscopic cholecystectomy the surgeon should not labor through a difficult
had longer mean operative times when compared operation. If at any point there is technical diffi-
to the standard four-port technique as well as culty with the single-site approach, additional
greater surgical difficulty (Table  83.2). instruments should be placed as there is little to
Furthermore, patients undergoing a single-­ no downside. Regardless of the technique used to
incision laparoscopic cholecystectomy described remove the gallbladder laparoscopically, in the
having more pain and required higher doses of absence of other problems, these patients can
1152
safely be discharged after the operation with a
low rate of readmission (Vaughan et  al. 2014;
Dalton et al. 2015; Gould et al. 2016).
83.8 Complications
The most dreaded and severe complication from
laparoscopic cholecystectomy is injury to the
CBD. The rate of ductal injury after laparoscopic
cholecystectomy has decreased as experience
increased, but the number of adults patients who
require a ductal reconstructive operation remains
about 1 per 1000 undergoing a laparoscopic cho-
lecystectomy (Dolan et al. 2005). It would be rea-
sonable to expect that complications in children
will remain less than adults given the high per-
centage of adult patients who require an opera-
tion due to severe inflammation obliterating the
planes of dissection. Very few pediatric patients
present in this fashion. Regardless, any surgeon
performing the operation should be aware of the
complications that can occur and how to manage
them.
When ductal injury is recognized at the time
of the operation, conversion with open repair
should be performed with a low threshold for
performing a hepaticojejunostomy (Lillemoe
2006). Obstructive ductal injuries and delayed
strictures detected after the operation can be tem-
porized with percutaneous transhepatic drainage
of the ductal system after which the patient
should be referred to a center with a large experi-
ence reconstructing these injuries. In this setting,
Roux-en-Y hepaticojejunostomy has a 98% suc-
cess rate (Lillemoe et al. 2000).
Bile leaks after laparoscopic cholecystectomy
are usually from the cystic duct stump or from
the gallbladder fossa (ducts of Luschka). The
sphincter of Oddi maintains a pressure gradient
of about 10  mmHg in the biliary system. Thus,
almost all postoperative leaks resolve promptly
with sphincterotomy and/or stent placement
which should be performed at the time of the
ERCP that is used for delineating the location of
the leak (Lillemoe 2006; Lillemoe et  al. 2000;
Agarwal et al. 2006).
C. Dekonenko et al.
Infectious complications are uncommon unless
gallstones are spilled. Therefore, spilled stones
should be removed to reduce these complications.
The most common complications following
cholecystectomy in children occur with hemo-
lytic disease and are related to the manifestations
of the underlying disease. In a report of 364 cases
from a national sickle cell disease study group,
the total complication rate was 39% with sickle
cell events representing 19%; intra-operative or
recovery room problems, 11%; transfusion com-
plications, 10%; postoperative surgical events,
4%; and death in 1% (Haberkern et  al. 1997).
Acute chest syndrome can be seen in up to 20%
of sickle cell patients undergoing abdominal sur-
gery. In addition, the laparoscopic approach does
not seem to decrease the incidence of this com-
plication. Meticulous attention to perioperative
management, transfusion guidelines, and pulmo-
nary care should help reduce the incidence of
these complications in sickle cell patients.
83.9 Conclusion
The pediatric surgeon should be well versed in
the disease processes, diagnostic evaluation,
management options, and outcomes in children
with gallbladder disease. The surgical approach
should be consistent with the surgeon’s personal
comfort and experience.
References
Agarwal N, Sharma BC, Garg S et al (2006) Endoscopic
management of postoperative bile leaks. Hepatobiliary
Pancreat Dis Int 5:273–277
Bhattacharyya N et al (1993) Perioperative management
for cholecystectomy in sickle cell disease. J Pediatr
Surg 28:72–75
Carney DE, Evan R, Kokoska ER et al (2004) Predictors
of successful outcome after cholecystectomy for bili-
ary dyskinesia. J Pediatr Surg 39:813–816
Cooperberg PL, Burhenne HJ (1980) Real-time ultraso-
nography. Diagnostic technique of choice in calculous
gallbladder disease. N Engl J Med 302:1277–1279
Cuschieri A, Lezoche E, Morino M et al (1999) E.a.E.S.
multicenter prospective randomized trial comparing
two-stage vs. single-stage management of patients
83  Gallbladder Disease
with gallstone disease and ductal calculi. Surg Endosc
29:952–957
Dalton BG, Gonzalez KW, Knott EM et al (2015) Same
day discharge after laparoscopic cholecystectomy in
children. J Surg Res 195:418–412
Dekonenko C, Sujka JA, Dorman RM, Oyetunji TA, St
Peter SD (2019 Sep) Self-reported outcomes after
cholecystectomy for pediatric hyperkinetic biliary
dyskinesia. J Surg Res 25(246):73–77
Dolan JP, Diggs BS, Sheppard BC et al (2005) Ten-year
trend in the national volume of bile duct injuries
requiring operative repair. Surg Endosc 19:967–993
Dubois F, Berthelot G, Levard H (1989) Cholecystectomy
by coelioscopy. Presse Med 18:980–982
Friesen CA, Neilan N, Daniel JF et  al (2011) Mast cell
activation and clinical outcome in pediatric choleli-
thiasis and biliary dyskinesia. BMC Res Notes 4:322
Garey CL, Laituri CA, Ostlie DJ, Snyder CL, Andrews
WS, Holcomb GW 3rd, St Peter SD (2011 May)
Single-incision laparoscopic surgery in children:
initial single-center experience. J Pediatr Surg
46(5):904–907
Gasior AC, Knott EM, Holcomb GW 3rd, Ostlie DJ, St
Peter SD (2014 Jan) Patient and parental scar assess-
ment after single incision versus standard 3-port
laparoscopic appendectomy: Long-term follow-up
from a prospective randomized trial. J Pediatr Surg
49(1):120–122
Goodwin EF, Partain PI, Lebensburger JD et  al (2017)
Elective cholecystectomy reduces morbidity of chole-
lithiasis in pediatric sickle cell disease. Pediatr Blood
Cancer 64:113–120
Gould JL, Poola AS, St Peter SD et al (2016) Same day
discharge protocol implementation trends in laparo-
scopic cholecystectomy in pediatric patients. J Pediatr
Surg 51:1936–1938
Gurusamy KS, Vaughan J, Rossi M, Davidson BR (2014,
Feb 20) Fewer-than-four ports versus four ports for
laparoscopic cholecystectomy Cochrane Database
Syst Rev (2):CD007109
Haberkern CM, Neumayr LD, Orringer EP et  al (1997)
Cholecystectomy in sickle cell anemia patients: peri-
operative outcome of 364 cases from the national pre-
operative transfusion study. Preoperative transfusion in
sickle cell disease study group. Blood 89:1533–1542
Holcomb GW III (1993) Laparoscopic cholecystectomy.
Semin Pediatr Surg 2:159–167
Holcomb GW III, Sharp KW, Olsen DO (1991)
Laparoscopic cholecystectomy in the pediatric patient.
J Pediatr Surg 26:1186–1190
Holcomb GW III, Morgan WM III, Neblett WW III et al
(1999) Laparoscopic cholecystectomy in children: les-
sons learned from the first 100 patients. J Pediatr Surg
34:1236–1240
Holcomb GW Jr, O’Neill JA Jr, Holcomb GW III (1980)
Cholecystitis, cholelithiasis and common duct stenosis
in children and adolescents. Ann Surg 191:626–635
Kaechele V, Wabitsch M, Thiere D et al (2006) Prevalence
of gallbladder stone disease in obese children and ado-
lescents: influence of the degree of obesity, sex, and
1153
pubertal development. J Pediatr Gastroenterol Nutr
42:66–70
Khoo AK, Cartwright R, Berry S et  al (2014)
Cholecystectomy in English children: evidence of an
epidemic (1997-2012). J Pediatr Surg 49:284–288
King DR, Ginn-Pease ME, Lloyd TV et  al (1987)
Parenteral nutrition with associated cholelithiasis:
another iatrogenic disease of infants and children. J
Pediatr Surg 22:593–596
Knott EM, Fike FB, Gasior AC, Cusick R, Brownie E, St
Peter SD, Azarow KS (2013 Dec) Multi-institutional
analysis of long-term symptom resolution after chole-
cystectomy for biliary dyskinesia in children. Pediatr
Surg Int 29(12):1243–1247
Lacher M, Yannam GR, Muensterer OJ et  al (2013)
Laparoscopic cholecystectomy for biliary dyskine-
sia in children: frequency increasing. J Pediatr Surg
48:1716–1721
Lachman BS, Lazerson J, Starshak RJ et  al (1979) The
prevalence of cholelithiasis in sickle cell disease
as diagnosed by ultrasound and cholecystography.
Pediatrics 64:601–603
Lai SW, Rothenberg SS, Kay SM et al (2017) Outcomes
of laparoscopic cholecystectomy for biliary dyski-
nesia in children. J Laparoendosc Adv Surg Tech A
27:845–850
Liem NT, Benedict LA, Holcomb GW III (2020)
Choldedochal cyst and gallbladder disease. In:
Holcomb and Ashcraft’s pediatric surgery, 7th edn.
Elsevier Inc., St. Louis, pp 695–708
Lillemoe KD (2006) Evaluation of suspected bile duct
injuries. Surg Endosc 20:1638–1643
Lillemoe KD, Melton GB, Cameron JL et  al (2000)
Postoperative bile duct strictures: management and
outcome in the 1990s. Ann Surg 232:430–441
Mah D, Wales P, Njere I et  al (2004) Management of
suspected common bile duct stones in children: role
of selective intraoperative cholangiogram and endo-
scopic retrograde cholangiopancreatography. J Pediatr
Surg 39:808–812
Matos C, Avni EF, VanGansbeke D et al (1987) Total par-
enteral nutrition (TPN) and gallbladder diseases in
neonates. J Ultrasound Med 6:243–248
McChesney JA, Northup PG, Bickston SJ (2003) Acute
acalculous cholecystitis associated with systemic
sepsis and visceral arterial hypoperfusion: a case
series and review of pathophysiology. Dig Dis Sci
48:1960–1967
McKernan JB (1991) Laparoscopic cholecystectomy. Am
Surg 57:309–312
McLean TR (2006) Risk management observations from
litigation involving laparoscopic cholecystectomy.
Arch Surg 141:643–648
Moody FG, Haley-Russell D, Li YF et  al (1989) The
effects of lithogenic bile on gallbladder epithelium.
Ann Surg 210:406–415
Newman KD, Powell DM, Holcomb GW III (1997) The
management of choledocholithiasis in children in the
era of laparoscopic cholecystectomy. J Pediatr Surg
32:1116–1119
1154
Ostlie DJ, Holcomb GW III (2003) The use of stab inci-
sions for instrument access in laparoscopic operations.
J Pediatr Surg 38:1837–1840
Ostlie DJ, Juang OO, Iqbal CW et al (2013) Single inci-
sion versus standard 4-port laparoscopic cholecystec-
tomy: a prospective randomized trial. J Pediatr Surg
48:209–214
Rau B, Friesen CA, Daniel JF et  al (2006) Gallbladder
wall inflammatory cells in pediatric patients with
biliary dyskinesia and cholelithiasis: a pilot study. J
Pediatr Surg 41:1545–1548
Reddick EJ, Olsen DO (1989) Laparoscopic laser chole-
cystectomy. A comparison with mini-lap cholecystec-
tomy. Surg Endosc 3:131–113
Reynolds W Jr (2001) The first laparoscopic cholecystec-
tomy. JSLS 5:89–94
Rhodes MM, Bates DG, Andrews T et  al (2014)
Abdominal pain in children with sickle cell disease. J
Clin Gastroenterol 48:99–105
Sandler A, Winkel G, Kimura K et al (1999) The role of
prophylactic cholecystectomy during splenectomy in
children with hereditary spherocytosis. J Pediatr Surg
34:1077–1078
Sandoval C, Stringel G, Ozkaynak MF et  al (2002)
Perioperative management of children with sickle
cell disease undergoing laparoscopic surgery. JSLS
6:29–33
Schweizer P, Lenz MP, Kirschner HJ (2000) Pathogenesis
and symptomatology of cholelithiasis in childhood. A
prospective study. Dig Surg 17:459–467
Srinath AI, Youk AO, Bielefeldt K (2014) Biliary dyski-
nesia and symptomatic gallstone disease in children:
two sides of the same coin? Dig Dis Sci 59:1307–1315
St Peter SD, Keckler SJ, Nair A, Andrews WS, Sharp
RJ, Snyder CL, Ostlie DJ, Holcomb GW (2008 Feb)
Laparoscopic cholecystectomy in the pediatric popula-
tion. J Laparoendosc Adv Surg Tech A 18(1):127–130
C. Dekonenko et al.
Strasberg SM, Brunt LM (2010) Rationale and use of the
critical view of safety in laparoscopic cholecystec-
tomy. J Am Coll Surg 211:132–138
Strasberg SM, Hertl M, Soper NJ (1995) An analysis of
the problem of biliary injury during laparoscopic cho-
lecystectomy. J Am Coll Surg 180:101–125
Stringer MD, Ceylan H, Ward K et al (2003) Gallbladder
polyps in children – classification and management. J
Pediatr Surg 38:1680–1684
St-Vil D, Yazbeck S, Luks FL et al (1992) Cholelithiasis
in newborns and infants. J Pediatr Surg 27:1305–1307
Tsai J, Sulkowski JP, Cooper JN et al (2013) Sensitivity
and predictive value of ultrasound in pediatric chole-
cystitis. J Surg Res 184:378–382
Vaughan J, Nagendran M, Cooper J, Davidson BR,
Gurusamy KS (2014 Jan 24) Anaesthetic regimens
for day-procedure laparoscopic cholecystectomy.
Cochrane Database Syst Rev (1):CD009784
Vettoretto N, Saronni C, Harbi A et al (2011) Critical view
of safety during laparoscopic cholecystectomy. JSLS
15:322–325
Wales PW, Carver E, Crawford MW et  al (2001) Acute
chest syndrome after abdominal surgery in children
with sickle cell disease: is a laparoscopic approach
better? J Pediatr Surg 36:718–721
Walker SK, Maki AC, Cannon RM et al (2013) Etiology
and incidence of pediatric gallbladder disease. Surgery
154:927–933
Wu SC, Chen FC, Lo CJ (2005) Selective intraoperative
cholangiography and single-stage management of
common bile duct stone in laparoscopic cholecystec-
tomy. World J Surg 29:1402–1408
Zargar SA, Javid G, Khan BA et  al (2003) Endoscopic
sphincterotomy in the management of bile duct stones
in children. Am J Gastroenterol 98:586–589
 Pancreatic Disorders
84
Elke Zani-Ruttenstock and Augusto Zani

84.1 Introduction 84.2 Historical Overview

Pancreatic disorders can be congenital or acquired The pancreas was first identified by Herophilus
and can either affect the exocrine or the endo- (335–280 BC), a Greek anatomist and surgeon
crine component of the gland. Disorders that (Howard and Hess 2002), and named pancreas a
require surgical attention are classified in few hundred years later by Rufus of Ephesus,
Table 84.1. This chapter will give an overview of another Greek anatomist. Etymologically, the
these conditions, except for pancreatic trauma word “pancreas,” comes from the Greek πᾶν
that is discussed in a dedicated chapter. (pân, “all”) and κρέας (kréas, “flesh”), literally
meaning all-flesh, due to its fleshy consistency.
Table 84.1  Pancreatic diseases affecting children and The actual importance of the pancreas was only
adolescents recognized in 1889, when Oskar Minkowski,
Diseases of the pancreas German physician and physiologist, discovered
Structural anomalies Annular pancreas that removing the pancreas from a dog resulted in
Pancreas divisum diabetes. Insulin was later isolated from pancre-
Pancreaticobiliary malunion atic islets by the Canadian pediatric surgeon,
Genetic disorders Congenital hyperinsulinism Frederick Banting, and his research assistant,
Inflammatory conditions Acute pancreatitis
Charles Herbert Best, in 1921 (Karamanou et al.
Chronic pancreatitis
Cysts
2016).
Neoplasms
Trauma
84.3 Embryology

The pancreas starts developing at approximately

E. Zani-Ruttenstock · A. Zani (*)
Division of General and Thoracic Surgery, The
Hospital for Sick Children, Toronto, Canada
Department of Surgery, University of Toronto,
Toronto, Canada
e-mail: elke.ruttenstock@sickkids.ca; augusto.zani@
sickkids.ca
5 weeks of gestation as two outpouchings of the
duodenal endodermal lining, just distal to the
developing stomach (Sadler 2019). The out-
pouchings are the ventral and dorsal pancreatic
buds, and each one joins with the foregut through
a duct. By the end of the sixth week, the dorsal
bud drains through the dorsal duct of Santorini
into the foregut and the ventral bud drains through

© Springer Nature Switzerland AG 2023 1155

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_84
1156
a Formation b Rotation
Stomach
Common
bile duct
Ventral
Gallbladder Dorsal pancreas
pancreas
Dorsal
pancreatic
duct
Fig. 84.1  The three main stages of pancreas development: formation (a), rotation (b), and fusion (c)
the ventral duct of Wirsung into the hepatic diver-
ticulum (Fig. 84.1a). During the seventh week of
gestation, the ventral bud rotates behind the duo-
denum and positions itself adjacent to the inferior
part of the dorsal bud (Fig.  84.1b). During the
eighth week of gestation, the two buds fuse with
the dorsal bud, resulting in a single gland and a
combined ductal system (Fig. 84.1c): the ventral
duct of Wirsung and the distal part of the dorsal
duct of Santorini fuse, forming the main pancre-
atic duct. The main pancreatic duct joins the
common bile duct (CBD) just prior to the ampulla
of Vater, where both ducts perforate the medial
side of the second part of the duodenum at the
major duodenal papilla. The proximal part of the
dorsal duct of Santorini usually regresses, but if it
persists, it becomes the accessory pancreatic
duct. In the final anatomic arrangement, the neck,
body, and tail of the definitive pancreas originate
from the dorsal bud, whereas the head and unci-
nate process form the ventral bud.
84.4 Structural Pancreatic
Anomalies
Disorders of the pancreas can originate from con-
genital structural anomalies that may require sur-
gical interventions. The most common structural
anomalies are annular pancreas, pancreas divi-
sum, and pancreaticobiliary malunion. In these
anomalies, the exocrine and endocrine functions
of the pancreas are typically not affected, since
E. Zani-Ruttenstock and A. Zani
c Fusion
Common
bile duct
Accessory
pancreatic Main pancreatic
duct
duct
Ventral Dorsal pancreas
pancreas
Ventral
pancreatic
duct
the organ continues to produce normal levels of
digestive enzymes and insulin.
84.4.1 Annular Pancreas
Annular pancreas is caused by the incorrect
migration of the developing ventral pancreatic
bud leading to an annular fusion with the dorsal
pancreatic bud around the second part of the duo-
denum. Among several theories explaining the
development of annular pancreas, the most com-
monly accepted one is Lecco’s (Lecco 1910):
adherence of the ventral bud to the duodenal wall
prior to rotation results in its persistence and
encirclement of the duodenum. The exact preva-
lence of annular pancreas, estimated to be <1%
of the general population, is actually unknown as
many individuals remain asymptomatic through-
out their life. Approximately one-third of patients
with annular pancreas develop symptoms. The
commonest presentation in children is that asso-
ciated with duodenal atresia. Annular pancreas
can also be seen in combination with Down’s
syndrome, intestinal malrotation, esophageal
atresia, and congenital heart defects (Zyromski
et al. 2008).
84.4.1.1 Diagnosis and Differential
Diagnosis
Annular pancreas is rarely diagnosed on antena-
tal ultrasound where the fetus presents with poly-
hydramnios, double bubble sign (dilated stomach
84  Pancreatic Disorders
and first part of the duodenum), and a periduode-
nal hyperechogenic band (Zyromski et al. 2008;
Dankovcik et al. 2008). Postnatally, annular pan-
creas usually presents with symptoms associated
with duodenal atresia, which causes a partial or
complete obstruction of the duodenum. The baby
with complete duodenal obstruction classically
presents in the first few hours of life with the
onset of bilious or non-bilious vomiting, depend-
ing on whether the obstruction is pre- or
post-­
­ ampullary (i.e., proximal or distal to the
ampulla of Vater). A plain abdominal X-ray may
demonstrate the typical “double bubble” of duo-
denal atresia with the absence of distal gas.
Annular pancreas can also be detected on postna-
tal abdominal ultrasonography (Hameed et  al.
2020). Cases of incomplete duodenal obstruction
often present with subtle symptoms, may have a
delayed diagnosis, and may need an upper gas-
trointestinal contrast study.
84.4.1.2 Management
Once the diagnosis of duodenal obstruction has
been confirmed, decompression of the stomach
using a nasogastric tube and fluid resuscitation
are started. After correction of the electrolyte
imbalance, the patient undergoes surgery, where
the diagnosis of annual pancreas is made or con-
firmed. The pancreas is left in situ and a duodenal
bypass, in the form of duodeno-duodenostomy or
duodeno-jejunostomy, is performed with an open
or a laparoscopic approach (Zani et al. 2017).
84.4.2 Pancreas Divisum
This is the most common congenital anomaly of
the pancreas and is estimated to occur in approxi-
mately 5–10% of the population (Bülow et  al.
2014; Dimitriou et al. 2018; Hafezi et al. 2017).
It is thought to result from failure of the fusion of
the ventral and dorsal pancreatic ducts that nor-
mally occurs when the ventral and dorsal pancre-
atic buds fuse (Kozu et al. 1995). This leads to the
dominance of Santorini’s duct, which drains the
1157
majority of pancreatic secretions. As a result, the
relatively small size of the Santorini’s papilla
(also known as minor papilla) generates a signifi-
cantly higher intrapancreatic ductal pressure.
84.4.2.1 Diagnosis and Differential
Diagnosis
Most patients are asymptomatic (90–95%).
However, the restricted drainage of the pancre-
atic secretions through the minor papilla, and the
increased intrapancreatic duct pressure may
cause pancreatic duct dilatation and pancreatitis.
The diagnosis of pancreas divisum is made with
magnetic resonance cholangiopancreatography
(MRCP), which is currently considered the gold
standard. More recently, secretin-enhanced
MRCP has been suggested to enhance the detec-
tion of congenital pancreaticobiliary malforma-
tions (Rustagi and Njei 2014). Pancreas divisum
can also be detected with endoscopic retrograde
cholangiopancreatography (ERCP), which also
enables simultaneous therapeutic interventions
(Quest and Lombard 2000).
84.4.2.2 Management
Pancreas divisum is managed conservatively in
asymptomatic patients or when incidentally dis-
covered. In symptomatic patients, the goal of
treatment is to establish adequate drainage of the
main pancreatic duct in order to relieve symptoms
and to prevent the development of chronic pancre-
atitis and pancreatic insufficiency. The first line of
treatment is using ERCP with papillotomy of the
minor papilla, with or without stent implantation
(Michailidis et al. 2017). Typically, this yields a
good response and is associated with a low com-
plication rate. In the case of failed endoscopic
treatment or chronic pancreatitis and local com-
plications, such as CBD stricture or duct stenosis,
surgery is considered. Different surgical treat-
ments have been described, including sphinctero-
plasty, longitudinal pancreaticojejunostomy
(Puestow procedure), duodenopancreatectomy,
and duodenum-­preserving pancreatic head resec-
tion (Ferri et al. 2019).
1158
84.4.3 Pancreaticobiliary Malunion
Pancreaticobiliary malunion is defined as an
anomalous junction of the pancreatic duct and
the CBD proximal to the ampulla of Vater. As the
junction is located proximal to the duodenal mus-
culature, it is not controlled by the contractile
function of the sphincter of Oddi. This results in
a common channel that allows free intraductal
mixing of both pancreatic and biliary secretions
(Dabbas and Davenport 2009). Pancreaticobiliary
malunion is an intrinsic part of the choledochal
malformation complex typically consisting of a
long common pancreaticobiliary channel, an
abnormal angle of insertion for the CBD, and a
variable degree of proximal bile duct stenosis
(Zani-Ruttenstock and Davenport 2016). For
diagnosis, differential diagnosis and manage-
ment of pancreaticobiliary malunion, we there-
fore refer to the chapter on choledocal cysts.
84.5 Congenital Hyperinsulinism
Congenital hyperinsulinism (CHI) represents a
group of disorders characterized by excessive
insulin secretion from pancreatic β-cells, result-
ing in severe hypoglycemia (Pierro and Nah
2011; Laje and Scott Adzick 2020). In the past,
the same condition was named in several ways:
nesidioblastosis, islet cell adenomatosis, beta
cell dysregulation syndrome, or persistent
hyperinsulinemic hypoglycemia of infancy. CHI
is a rare disorder, with an incidence of 1  in
50,000 live births in most countries, but it can
be as frequent as 1 in 2,500 in countries with a
high degree of consanguinity (Flanagan et  al.
2011).
84.5.1 Differential Diagnosis
CHI can present as a transient or a persistent phe-
nomenon. Transient hyperinsulinism is mainly
caused by nongenetic factors, such as perinatal
stress, or it occurs in babies who are either small
for gestational age or macrosomic. The latter
may have transient CHI due to mono-allelic inac-
E. Zani-Ruttenstock and A. Zani
tivating mutation of hepatocyte nuclear factor 4α
(HNF4A) and may develop maturity onset diabe-
tes of the young type 1 (MODY1) later in life.
For this reason, they should be followed up after
CHI resolution.
Most cases of persistent hyperinsulinism have
recognized genetic etiologies. The genetic forms
are caused by recessive or dominant mutations of
regulatory genes for glucose metabolism and
insulin secretion. The genes most commonly
involved are ABCC8 and KCNJ11, which encode
for the subunits SUR1 and KIR6.2 of the ATP-­
sensitive K+ (K-ATP) channel. Mutated genes
cause the K-ATP channels to remain permanently
closed and to generate a nonregulated, persistent
insulin release, that leads to constant
hypoglycemia.
Recessive inactivating mutations in ABCC8 or
KCNJ11 cause continuous β-cell membrane
depolarization and subsequent calcium influx,
which results in unregulated insulin secretion.
Dominant inactivating mutations in the same
genes cause a milder form of CHI.
Other genes implicated in rarer forms of CHI
are those that regulate insulin secretion from the
pancreatic β-cells, such as GLUD1 (encoding the
enzyme glutamate dehydrogenase), GCK (encod-
ing glucokinase), HADH (encoding the L-3-­
hydroxyacyl-coenzyme A dehydrogenase),
UCP2 (encoding the uncoupling protein 2), and
SLC16A1 (encoding the monocarboxylate trans-
porter 1).
CHI can be classified into three histological
variants: focal, diffuse, and atypical (Laje and
Adzick 2020).
Focal CHI involves nodules of adenomatous
islet cell hyperplasia with normal islets surround-
ing these areas. The size of these focal lesions
varies from a few millimeters to >1 cm and can
be found on the surface or deep within the
pancreas.
Diffuse CHI is characterized by β-cell nucleo-
megaly, which is found throughout the entire
pancreas in most cases.
In addition to the two major forms, there are
rare atypical cases of CHI, where the pancreas
presents a diffuse CHI pattern in one large area,
but the rest of the organ is histologically normal.
84  Pancreatic Disorders
84.5.2 Diagnosis
The inappropriate insulin secretion that is seen in
CHI results in increased glucose consumption,
inhibition of glycogenolysis and gluconeogene-
sis, and ultimately hypoglycemia. This puts the
brain at severe risk of injury, due to the suppres-
sion of ketone body production, i.e., its alterna-
tive energy substrate, secondary to insulin-induced
inhibition of lipolysis and free fatty acid produc-
tion. Most commonly, neonates present with
hypoglycemia and poor feeding shortly after
birth. Some infants have lethargy or irritability,
whereas others have a more severe presentation
with apnea, coma, and seizures (Laje and Adzick
2020). It is therefore imperative that glycemic
levels are checked in all neonates immediately
after birth. Older patients can present with typical
clinical features of hypoglycemia such as pallor,
sweat, and tachycardia.
The diagnostic criteria for CHI include the
following:
• Fasting and/or postprandial hypoglycemia (<
2.5–3 mmol/L or <50mg/dL)
• Inappropriate plasma insulin and c-peptide
levels concomitant to hypoglycemia
• Increased blood glucose levels (>1.7 mmol/L
or 30 mg/dL) within 30–40 minutes following
glucagon (direct insulin antagonist) adminis-
tration. This is important for the differential
diagnosis with glycogen storage disease
• Inappropriately low plasma and urine ketone
bodies and low plasma free fatty acids for fast-
ing hypoglycemia
To distinguish between focal and diffuse sub-
types, a genetic analysis is performed. Typically,
children with diffuse CHI have homozygous
recessive or compound heterozygous mutations
in ABCC8 and KCNJ11 genes. When genetic
testing confirms diffuse CHI, no further imaging
studies are necessary.
Focal CHI is associated with loss of hetero-
zygosity for paternally inherited mutations in
K-ATP genes. Infants without characteristic
genetic mutations of diffuse CHI undergo 18
F-L-3,4-dihydroxyphenylalanine positron emis-
1159
sion tomography combined with computed
tomography scan (18 F-DOPA-PET-CT scan).
This imaging technique allows for an accurate
differentiation between focal and diffuse forms,
and in the case of focal CHI, it helps to locate
lesions that are deep in the parenchyma and
hardly visible macroscopically (Zani et  al.
2011).
84.5.3 Management
In the case of CHI, it is crucial to make a prompt
diagnosis and institute immediate treatment,
since delay in management may cause irrevers-
ible brain damage and long-term neurodevelop-
mental disabilities. It is critical to have a central
venous catheter inserted to monitor blood glu-
cose levels and provide a reliable route for intra-
venous glucose administration. Infants with CHI
have a fasting tolerance of less than 1 hour, so
they require continuous intravenous glucose infu-
sion at > 8 mg/kg/minute to maintain normal gly-
cemic levels (normal: 4–6 mg/kg/minute). The
goal of therapy is to achieve normal blood glu-
cose levels (>63 mg/dl or >3.5 mmol/L) and to
inhibit inappropriate insulin secretion.
Therapeutic strategies can be medical, surgical,
or combined.
84.5.4 Medical Treatment
In the emergency setting, such as in the case of
seizures, the infant should receive an intrave-
nous bolus of 2ml/kg of 10% glucose, followed
by a glucose infusion at >6–8 mg/kg/minute. In
the absence of intravenous access, glucagon
can be administered via intramuscular injec-
tion to stimulate glycogenolysis and glucose
release from the liver. Patients will also be
started on a continuous or frequent feeding
regime.
Drug treatment includes diazoxide with chlo-
rothiazide, nifedepine, octreotide, and glucagon.
Diazoxide (K-ATP agonist) is an inhibitor of
insulin secretion and it is used in all types of CHI
as first-line therapy. Diazoxide binds to the SUR1
1160 E. Zani-Ruttenstock and A. Zani

subunit of the K-ATP channel, which activates 84.5.5 Surgical Treatment

intact K-ATP channels and reduces insulin secre-
tion. To be effective, both subunits of the channel In the case of focal CHI, a partial pancreatectomy
must be structurally and functionally normal. with clear margins will be curative. Intraoperative
However, patients with diffuse high-resolution ultrasound can help in localizing
CHI secondary to ABCC8 or KCNJ11 muta- focal lesions and can reduce the number of frozen
tion and those with focal lesions are unresponsive section biopsies. Furthermore, it can be useful to
to diazoxide, as their K-ATP channels are not identify the pancreatic duct.
structurally normal. Since the most common Small and superficial lesions in the body or tail
cause of CHI is a mutation in the SUR1/KIR6.2 of the pancreas are treated by simple resection,
gene complex, about two-thirds of CHI patients whereas deep, periductal lesions are treated by dis-
do not respond to diazoxide. To identify nonre- tal pancreatectomy. Small and superficial lesions
sponders, patients undergo a diazoxide response in the head of the pancreas can also be treated by
test, which consists of a five-day administration simple resection, whereas deep lesions are difficult
of the drug followed by a fasting test in the to resect without causing damage to the CBD or
absence of glucose infusion or other medications. pancreatic duct. To ensure a complete resection of
Patients who maintain a plasma glucose level these more challenging lesions, the majority of the
>70 mg/dL for ≥12 hours are considered “diazox- pancreatic head will be removed, and a Roux-en-Y
ide responsive.” These patients are subsequently pancreatico-jejunostomy will be fashioned to
managed by a regimen of frequent feedings and allow drainage of the pancreatic body and tail
long-term diazoxide. Conversely, diazoxide-­ (Laje et al. 2012). In the case of extension of the
resistant patients generally require surgery; prior focal pancreatic head lesion into the duodenal
to surgery, to maintain euglycemia, they can use wall, a pancreaticoduodenectomy (i.e., Whipple
octreotide, glucagon, and nifedipine as an procedure) is the treatment of choice.
alternative. Laparoscopic surgery is especially suitable
Side effects: diazoxide can causes severe for focal lesions in the pancreatic body or tail.
sodium and water retention that may result in In cases of diffuse CHI that fail to respond to
congestive heart failure, especially in neonates. medical management, an open or laparoscopic
For this reason and for its synergistic effect on near-total pancreatectomy (95%) is considered as
suppressing insulin secretion, chlorothiazide, a the gold standard (Fig. 84.2). A near-total pancre-
thiazide diuretic is used in combination with
diazoxide.
Octreotide is a long-acting analogue of soma-
tostatin, which naturally inhibits the insulin
release from the pancreatic β-cells. Octreotide
activates the somatostatin receptor 5 (SSTR5),
thus inhibiting calcium mobilization and acetyl-
choline activity and decreasing insulin gene pro-
moter activity, which reduces insulin biosynthesis
and insulin secretion.
Glucagon is a natural insulin antagonist that is
mainly used to promptly reverse severe hypogly-
cemic episodes.
Nifedipine is a calcium-channel blocker that
inhibits insulin secretion by inactivation of
voltage-­gated calcium-channels. However, it is Fig. 84.2  Near-total pancreatectomy: the head, body,
and tail of the pancreas including the uncinate process are
only effective in a minority of patients, and dos- resected, and only a small portion (5%) of pancreatic tis-
age is often limited by hypotension. sue is left in situ
84  Pancreatic Disorders 1161

atectomy consists of resection of the tail, body,
uncinate process, and most of the pancreatic
head, leaving a rim of pancreatic tissue surround-
ing the CBD along the duodenum. A near-total
pancreatectomy can be performed with an open
or laparoscopic approach.
84.5.6 Postoperative Management
Postoperatively, an intravenous glucose infusion
is restarted at a low rate as surgical stress induces
hepatic glycogenolysis (Laje and Adzick 2020).
In the case of persistently high plasma glucose
levels, an intravenous insulin infusion is also
started. Upon return of bowel function, enteral
feeds are resumed and advanced gradually, with
subsequent weaning of the intravenous glucose
infusion. Postoperative pain after neonatal pan-
createctomy is controlled with an epidural cathe-
ter or intravenous narcotics. Postoperative
complications specific to pancreas surgery, such
as pancreatic leak, lymphatic leak, and CBD
injury are reported with an incidence of <1%
(Laje and Adzick 2020).
(Keim et al. 2003; Rebours et al. 2009), this condi-
tion is far less common in children than in adults.
Recognizing the limited literature, especially for
the more complex forms, the “International Study
Group of Pediatric Pancreatitis: In Search for a
Cure” (INSPPIRE) consortium formed as a multi-
center approach to systematically characterize
pancreatitis in children. In 2012, the consortium
reported a consensus statement on the definitions
of childhood-onset acute pancreatitis, acute recur-
rent pancreatitis, and chronic pancreatitis
(Morinville et al. 2012) (Table 84.2).
84.6.1 Acute Pancreatitis
The most common causes of acute pancreatitis in
children are summarized in Table  84.3 (Diesen
2020a, b). Although acute pancreatitis is rare in
childhood, it appears that the incidence of pediat-
ric pancreatitis has increased over the last two
decades (Diesen 2020a, b). This is likely due to a
combination of factors, including the increased
incidence of obesity in the pediatric population,
drug-associated pancreatitis, and improvements
in its diagnosis (Poddar et al. 2017).

84.5.7 Postoperative Outcome

Table 84.2  Definition of pancreatitis in children (modi-
The outcome for patients with focal CHI is excel- fied from Morinville et al. 2012)
lent, with >95% of cases cured after surgical Acute Chronic
pancreatitis Acute recurrent pancreatitis
excision. Conversely, following the near-total
(AP) pancreatitis (ARP) (CP)
pancreatectomy, 50% of patients with diffuse One of the three • AT least two • Imaging
CHI have persistent hypoglycemia, 25% are criteria: distinct attacks of findings of
cured and have normoglycemia, and 25% develop • Abdominal AP CP
diabetes that require insulin (Laje and Adzick pain N.B.: Attacks must AND one of
suggestive of be separated by the following
2020). Moreover, children who undergo near-­ AP (acute either ≥1-month three:
total pancreatectomy for diffuse CHI commonly onset in the pain-free interval or • Abdominal
have neurobehavioral deficits (Lord et al. 2015). epigastrium) pain resolution and pain
• Serum lipase normalization of suggestive of
or amylase at serum amylase and pancreatitis
least three lipase before the • Exocrine
84.6 Pancreatitis times the subsequent attack pancreatic
upper limit of insufficiency
Pancreatitis is defined as an inflammatory process normal • Endocrine
• Imaging pancreatic
secondary to the intraductal activation and secre- findings insufficiency
tion of pancreatic enzymes, which results in the characteristic
digestion of the parenchyma. With an estimated of AP (any
incidence of 3–13 cases per 100,000 children modality)
1162 E. Zani-Ruttenstock and A. Zani

Table 84.3  Causes of acute pancreatitis in the pediatric may be less well localized, and particularly in
population (modified from Diesen 2020a, b)
younger children, nausea and vomiting are often
Causes of pancreatitis the predominant features. To determine the cause
Trauma Blunt/penetrating of pancreatitis, it is important to elucidate a his-
Non-accidental trauma
Motor vehicle collision tory of recent abdominal trauma or surgery,
Iatrogenic recent infections or sick contacts, use of new
Abdominal surgery medications, toxin exposure, history of blood
Endoscopy/ERCP disorders, gallstones, and family history of medi-
Choledocholithiasis Cholesterol stones—obesity
Hemolytic
cal problems involving abdominal pain, surgery,
stones—spherocytosis or pancreatitis (Table  84.3). Physical examina-
 • alpha-thalassemia tion may elicit a range of signs and symptoms,
 • sickle cell disease including low-grade pyrexia, jaundice, diarrhea,
Parenteral nutrition
irritability, lethargy, epigastric tenderness, and
Congenital anomalies Annular pancreas
Pancreas divisum generalized peritonitis. In severe forms of pan-
Pancreaticobiliary malunion creatitis, the child may present with hypovolemic
Choledochal cyst shock or multi-organ failure with dyspnea, oligu-
Systemic and genetic Hypertriglyceridemia ria, and mental status change. In cases of necro-
diseases Hypercalcemia
Cystic fibrosis tizing or hemorrhagic pancreatitis, patients may
Crohn’s disease have abdominal discoloration at the umbilicus
Ulcerative colitis (Cullen sign) or flank (Grey Turner sign) due to
Systemic lupus internal bleeding (Diesen 2020a, b).
erythematosus
Familial pancreatitis The key laboratory tests for the diagnosis of
Shwachman-Diamond pancreatitis are serum amylase and lipase.
syndrome Amylase rises quicker than lipase, whose levels
Infections Mumps remain elevated a few days longer. Complete
Rubella
Coxsackie B
blood count and electrolyte levels are indicative
HIV of general clinical status and useful to determine
Parasites the need for resuscitation. Serum triglyceride lev-
Drugs Valproic acid els may indicate that the pancreatitis is the result
l-asparaginase
of hyperlipidemia, whereas liver function tests
Steroids
Metronidazole are important to rule out biliary obstruction in
Macrodantin patients with gallstones.
Azathioprine A plain abdominal X-ray (AXR) is often the
Tetracycline
first imaging study obtained, especially in
Toxin exposure
patients with abdominal pain of unknown etiol-
Liver transplant
Autoimmune pancreatitis ogy. However, this study will not show any spe-
Idiopathic cific signs of acute pancreatitis. Similarly, a chest
X-ray may demonstrate pulmonary edema or
pleural effusion, which may be associated with
84.6.1.1 Diagnosis and Differential pancreatitis. Abdominal ultrasonography can
Diagnosis detect some signs of acute pancreatitis, such as
The diagnosis of acute pancreatitis is based on diffuse or focal enlargement and altered echo-
history and physical examination and confirmed genicity, and can be useful to detect dilation of
by the results of both laboratory and radiological the pancreatic duct and the presence of stones.
investigations. Nonetheless, it is not uncommon that the pan-
Acute pancreatitis classically presents with creas appears normal despite the presence of
epigastric pain that may radiate to the upper inflammatory changes (Restrepo et al. 2016). An
quadrants and/or the back. Sometimes, the pain abdominal CT scan with contrast provides a more
84  Pancreatic Disorders
Fig. 84.3  Abdominal CT scan with double contrast of a
12-year-old girl presenting with acute necrotizing pancre-
atitis secondary to gallstones. The scan demonstrates a
diffusely swollen pancreas with 75% necrosis
accurate picture of the severity of pancreatic
damage (enlargement, peripancreatic stranding,
pancreatic necrosis, and accumulation of peri-
pancreatic fluid) and a better visualization of the
ductal anatomy and surrounding vasculature
(Restrepo et al. 2016) (Fig. 84.3). The MRCP is
useful to study anomalies of the pancreatic ducts
(e.g., strictures, stones, tumors) and CBD, and it
is particularly indicated in patients with recurrent
pancreatitis of unknown etiology. In cases of
acute pancreatitis, an ERCP is typically not indi-
cated, also because it has an associated risk
(5–10%) of post-procedural pancreatitis.
However, an ERCP can be useful in patients with
pancreatic duct disruption secondary to trauma
for diagnosis and management (stent placement),
in cases of idiopathic recurrent pancreatitis, and
in patients with pancreatitis associated with chol-
angitis/cholelithiasis, if the anatomy is not well
characterized on MRCP.
84.6.1.2 Management
Children with acute pancreatitis require very
close monitoring throughout their admission and
sometimes may even need transfer to a higher
level of care (e.g., intensive care unit). As chil-
dren with acute pancreatitis may deteriorate rap-
idly, early signs of multi-organ failure should be
treated aggressively with ventilation support and
active fluid resuscitation as needed. Pain control
1163
is usually achieved with nonsteroidal anti-­
inflammatory agents and narcotics. Antibiotics
are typically not needed, unless the patient has
evidence of infected pancreatic necrosis
(Villatoro et  al. 2010). In contrast to previous
teachings, enteral nutrition should be provided
from early on, regardless of the feeding route,
i.e., oral, gastric, or post-pyloric (Márta et  al.
2016; Poropat et al. 2015). Conversely, parenteral
nutrition should be considered only if the patient
is unable to tolerate feeds. Nasogastric decom-
pression is needed only if the child is vomiting.
If the acute pancreatitis is secondary to a
stone, an MRCP or an ERCP may be needed to
identify or remove the obstruction. If there is
strong suspicion of ductal obstruction on ultra-
sound, the ERCP is the next logical step to
remove the stone, dilate, and/or stent a duct stric-
ture, biopsy a pancreatic mass if identified, per-
form a sphincterotomy. If the patient has
gallstones, once the abdominal pain has resolved,
a cholecystectomy prior to discharge from the
hospital is warranted to prevent early recurrence
of pancreatitis.
84.6.2 Complications of Acute
Pancreatitis
84.6.2.1 Pancreatic Pseudocyst
A pseudocyst results from the accumulation of
leaked pancreatic enzymes enclosed within an
inflammatory non-epithelial lining (Fig.  84.4).
Pseudocysts initially appear as peripancreatic
fluid collection 10–14 days after the onset of
acute pancreatitis and should be suspected if ele-
vated amylase levels reappear after having set-
tled. Most pseudocysts develop soon after the
insult and resolve spontaneously in <6 weeks.
Spontaneous resolution is seen more commonly
in pseudocysts secondary to trauma. If a pseudo-
cyst persists, maturation of the pseudocyst wall is
essential before definitive treatment is performed,
typically 6–8 weeks after the acute episode.
Most patients with asymptomatic pseudocysts
respond well to conservative management, as
spontaneous resolution is common if no stones
are present. Patients with symptomatic pseudo-
1164
Fig. 84.4  3D reconstruction of abdominal MRI with
contrast, showing irregular dilation of the distal pancreatic
duct communicating with multiple peripancreatic collec-
tions of walled off necrosis in a 15-year-old male, who
developed acute pancreatitis secondary to blunt trauma
cysts usually develop abdominal pain, abdominal
distention, early satiety, and/or vomiting espe-
cially if pressing on the stomach, jaundice, gas-
trointestinal bleeding, weight loss, or intestinal
obstruction (Diesen 2020a, b). These patients are
typically managed conservatively for the first
weeks, while waiting for the pseudocyst wall to
mature. This is critical to minimize the risk of
leakage of the cyst content into the peritoneum
upon drainage. If the patient is still symptomatic
after the cyst wall has formed, a drainage proce-
dure is performed by endoscopic drainage or sur-
gical cystgastrostomy, cystduodenostomy, or
cystjejunostomy with an open or laparoscopic
approach (Diesen 2020a, b).
84.6.2.2 Necrotizing Pancreatitis
Patients with necrotizing pancreatitis are typi-
cally managed with supportive care unless there
is evidence of pancreatic infection or ductal
obstruction (Diesen 2020a, b). Pancreatic infec-
tion can be suspected on CT scan when air is
detected in the area of necrosis (Fig.  84.3).
Antibiotics are the first line of treatment, fol-
lowed by drain insertion to obtain source control
and, lastly, by surgical debridement if the patient
does not respond. This should be first attempted
with a minimally invasive approach (percutane-
ous, endoscopic, laparoscopic, or retroperito-
neal). If the patient does not respond or has rapid
E. Zani-Ruttenstock and A. Zani
clinical deterioration or uncontrolled sepsis, an
open debridement and necrosectomy may be nec-
essary. These patients are critically ill and may
need repeated washouts and debridement.
84.6.2.3 Pancreatic Fistula
A pancreatic fistula is usually a postoperative
complication and can be associated with a low or
high output. Most low-output fistulae close spon-
taneously. Therefore, treatment is directed at
maintaining adequate nutrition, reducing secre-
tions with pharmacological agents, such as soma-
tostatin analogues, and ensuring that the fistula
tract does not become prematurely obstructed.
Persistent fistulae may require surgical
intervention.
84.6.2.4 Pancreatic Hemorrhage
Treatment of pancreatic hemorrhage includes
fluid resuscitation and correction of associated
coagulopathy and/or thrombocytopenia. Severe
hemorrhage may require embolization or
surgery.
Although the majority of cases of acute pan-
creatitis are single episodes and self-resolving,
some children may develop acute recurrent pan-
creatitis, and others may develop irreversible
damage of the pancreas leading to chronic
pancreatitis.
84.6.3 Acute Recurrent Pancreatitis
In their seminal study, the INSPPIRE consortium
proposed this term as a more accurate alternative
to the commonly used “recurrent acute pancreati-
tis” (Morinville et  al. 2012). The diagnosis is
made based on the definition reported in
Table  84.2. Children presenting with recurrent
pain and a pseudocyst are not considered as hav-
ing acute recurrent pancreatitis, as the pseudocyst
is to be seen as a complication of the previous
episode of acute pancreatitis. The management of
acute recurrent pancreatitis in children is nonspe-
cific and partially overlaps that of acute pancre-
atitis. Nonetheless, patients are studied in detail
to identify the possible causes of the recurrent
attacks, which are mainly ascribed to choledo-
84  Pancreatic Disorders
cholithiasis, congenital anomalies, or systemic/
genetic diseases (Testoni 2014).
84.6.4 Chronic Pancreatitis
Chronic pancreatitis is defined as irreversible
structural damage to the pancreas resulting in
focal or diffuse destruction, sclerosis, or pancre-
atic duct abnormalities with or without insuffi-
ciency (Diesen 2020a, b; Morinville et al. 2012).
It may be the result of acute or acute recurrent
pancreatitis that finally leads to irreversible struc-
tural damage. Like patients with acute recurrent
pancreatitis, the etiology of chronic pancreatitis
is often found in congenital anomalies or sys-
temic/genetic diseases.
84.6.4.1 Diagnosis and Differential
Diagnosis
For the diagnosis of chronic pancreatitis, imaging
findings are crucial and are to be associated with
either abdominal pain, exocrine, or endocrine
pancreatic insufficiency (Table 84.2).
The abdominal pain and associated symptoms
are similar to those described for patients with
acute pancreatitis. Advanced cases can also pres-
ent with neurological symptoms and/or malnutri-
a b
Fig. 84.5 (a & b) Abdominal ERCP with contrast in a patient with chronic pancreatitis due to missense mutation of
N34S (SPINK 1 gene). The pancreatic duct is dilated up to 10 mm. (Courtesy of Prof. M. E. Höllwarth)
1165
tion. Those with endocrine insufficiency have
diabetes, whereas those with exocrine insuffi-
ciency experience maldigestion and malabsorp-
tion with diarrhea, steatorrhea, abdominal
distention, fat-soluble vitamin (A, D, E, and K)
deficiency, anorexia, and weight loss (Diesen
2020a, b). Laboratory tests are critical to establish
the general patient status and organ function.
Imaging findings of chronic pancreatitis in
children are similar to those of adults and include
parenchymal changes (fibrosis, atrophy), duct
changes (dilation, irregularity, strictures, periduc-
tal fibrosis, debris or calcifications), and calcifica-
tions (Fig. 84.5a, b). Abdominal ultrasonography
is the first imaging modality used in children with
suspected chronic pancreatitis, but better defini-
tion of chronic changes is obtained with cross-
sectional imaging (CT and MRCP). Recently,
functional imaging (MRI with secretin stimula-
tion) of the volume of fluid secreted by the pan-
creas as an estimate of exocrine function has been
found to be accurate and also reproducible in chil-
dren (Trout et  al. 2018). ERCP is helpful in
assessing and treating children with chronic pan-
creatitis and obstructing lesions. Specifically,
ERCP allows for duct dilatation and/or stenting,
mass biopsy, and performing a sphincterotomy
(Sacco Casamassima et al. 2017).
1166
84.6.4.2 Management
Supportive care, with a focus on adequate pain
management, is crucial while investigating the
etiology of the episodes. Moreover, patients with
chronic pancreatitis often have endocrine and/or
exocrine insufficiency that needs correction.
Early diagnosis and treatment of diabetes are
important to minimize the risk of complications.
Similarly, pancreatic enzymes and/or vitamin
supplementation may be needed.
First-line treatment for children with chronic
pancreatitis is represented by therapeutic endos-
copy (Sacco Casamassima et al. 2017). Intractable
pain is the main indication for surgery, which is
performed if the endoscopic management of
chronic pain and complications (pseudocysts,
strictures, and obstructing stones) has failed.
Surgical options consist of pancreatic drainage
procedures, resection, or a combination of both.
The most common surgery performed in children
with chronic pancreatitis is a longitudinal pancre-
aticojejunostomy, also known as Puestow proce-
dure (Sacco Casamassima et al. 2017). If only a
focal area of the pancreas is diseased, a resection
of the affected area with or without a drainage
procedure can be contemplated. If the diseased
portion of the pancreas is localized in the tail, a
distal pancreatectomy may be the treatment of
choice. If the affected area is in the head of the
pancreas, a lateral pancreaticojejunostomy, also
known as Frey procedure is a good option. Some
patients with extensive disease are treated with a
Whipple procedure.
The outcome of children and adolescents with
chronic pancreatitis often depends on their under-
lying diagnosis. In fact, those with systemic and
genetic diseases, such as cystic fibrosis, tend to
do worse than the others. Overall, the success
rate of treating pediatric patients with chronic
pancreatitis is similar to that of adults and ranges
between 65 and 100% (Diesen 2020a, b).
84.7 Pancreatic Cysts
Cystic lesions of the pancreas can be divided into
congenital and acquired. Congenital cysts can
present as a single lesion, which is typically
E. Zani-Ruttenstock and A. Zani
Fig. 84.6  Abdominal ultrasonography showing a con-
genital pancreatic cyst in a newborn with sedimentation
on the dorsal part of the cyst. (Courtesy of Prof. M.  E.
Höllwarth)
located in the tail and body of the pancreas, with-
out communicating with the pancreatic duct
(Fig.  84.6). Enteric duplication cysts within the
pancreas have been reported in children, but they
are extremely rare (Hunter et  al. 2008; Awais
et  al. 2019). Some children with genetic disor-
ders, such as polycystic kidney disease or von
Hippel-Lindau syndrome, may have multiple
congenital pancreatic cysts that are located
throughout the entire organ. Acquired pancreatic
cysts are usually secondary to trauma or pancre-
atitis and present as pseudocysts, as described
above. Less commonly, acquired cysts develop
following pancreatic duct obstruction (retention
cysts) or infection (hydatid cyst).
84.7.1 Diagnosis and Differential
Diagnosis
Congenital cysts are often diagnosed incidentally
or as part of the screening process routinely con-
ducted in patients with polycystic genetic disor-
ders. In some cases, congenital cysts may be as
large as to cause compression symptoms on the
surrounding structures. In the latter cases,
patients may experience nausea, early satiety,
abdominal pain, and jaundice. These symptoms
are often observed also in patients with acquired
84  Pancreatic Disorders
cysts. Children with enteric duplication cysts of Table 84.4  Classification of primary pancreatic tumors
the pancreas typically present with signs and Exocrine
symptoms of pancreatitis.
Abdominal ultrasonography is the first imag-
ing study, which confirms the clinical suspicion
or incidentally detects the pancreatic cyst. Cross-­
sectional imaging (CT and MRI) are useful to
provide detailed information about the size and
structure of the cyst and its nature and anatomical
relations.
84.7.2 Management
Small incidental, asymptomatic congenital cysts
can be left in situ and monitored over time, pro-
vided that the malignant nature has been ruled
out. Patients presenting with symptoms due to
the presence of single or multiple congenital
cysts benefit from surgical excision that can also
be performed laparoscopically when located in
the tail and the body. Symptomatic cysts located
in the head of the pancreas may require extensive
pancreatic resection. Acquired pancreatic cysts
are treated according to the nature of the cyst.
The treatment strategy of post-traumatic or post-­
pancreatitis pseudocysts is described above.
Patients with hydatid cysts of the pancreas are
treated surgically, with conservative techniques
rather than radical intervention, especially if the
cyst is located in the pancreatic head (Akbulut
et  al. 2014). Symptomatic enteric duplication
cysts should be resected (Hunter et  al. 2008;
Awais et al. 2019).
84.8 Pancreatic Tumors
Pancreatic tumors are rare in the pediatric popu-
lation; thus even high-volume centers treat a very
small number of patients with this neoplasm
every year (Perez et al. 2009). For this reason, the
diagnosis of pancreatic tumors is often
challenging.
Pancreatic childhood tumors have tradition-
ally been classified according to the World Health
Organization classification of pancreatic neo-
plasms of adults (Johnson and Spitz 2000). In
1167
Benign
pancreas • Serous cystadenoma (microcystic/
oligocystic)
• Mucinous cystadenoma
• Mature cystic teratoma
Borderline malignant
• Solid pseudopapillary neoplasm
(SPN; previously known as Frantz
tumor)
Malignant
• Pancreatoblastoma
• Mucinous cystadenocarcinoma
• Ductal adenocarcinoma
• Acinar cell carcinoma
Endocrine Benign
pancreas • Sporadic insulinoma
• Sporadic gastrinoma
• VIPomas
Borderline malignant to malignant
• Multiple endocrine neoplasia (MEN
1) associated gastrinoma and
insulinoma
this chapter, we will give an overview of primary
tumors of the exocrine and endocrine pancreas
(Table 84.4).
84.8.1 Diagnosis and Differential
Diagnosis
As often occurs in the adult population, children
with pancreatic tumors are typically asymptom-
atic. Moreover, in the early phases, children usu-
ally do not present with signs and symptoms of
obstructive jaundice, and when symptoms are
present, they are aspecific, such as vague abdom-
inal pain in the epigastrium. However, children
with endocrinologically active tumors, such as
insulinoma or gastrinoma, typically present with
signs and symptoms related to the effects of hor-
mone hypersecretion.
On physical examination, patients may present
with a palpable abdominal mass in the epigas-
trium. Some patients present with a benign abdo-
men and are diagnosed incidentally with abdominal
ultrasonography. Regardless of the presentation,
ultrasonography is the first-line imaging modality,
with a good accuracy for the initial delineation of
pancreatic masses. Nonetheless, cross-sectional
1168
imaging is required to determine the anatomic
relationship of a pancreatic mass and its surround-
ing structures and is critical to determine the
potential for resectability and the type of surgical
approach. Most patients initially undergo a CT
scan as it provides very precise multiplanar infor-
mation. However, some types of pancreatic
tumors, e.g., insulinoma, are better studied with an
abdominal MRI scan. Moreover, metabolically
active primary malignancies of the pancreas, such
as insulinomas, can be functionally characterized
using PET-CT scan in children and adolescents
(Bohl et al. 2018). Lastly, endoscopic ultrasonog-
raphy has also become increasingly used for the
diagnosis of pancreatic neoplasms, as it combines
the advantages of accurate visualization of the
mass with the possibility of simultaneous fine nee-
dle aspiration or biopsy (Cofer 2020).
84.8.2 Treatment
Benign tumors of the exocrine pancreas include
serous cystadenomas, mucinous cystadenomas,
and mature cystic teratomas (Table 84.4).
Serous cystadenomas are characterized by the
presence of multiple small cysts (microcystic) in
the tail or a few larger cysts (oligocystic) in the
head of the pancreas (Cofer 2020). Malignant
transformation of these masses has not been
reported in children; thus, neither type requires
surgical excision unless they grow large and
become symptomatic.
Conversely, mucinous cystadenomas are
masses that localize in the tail, can grow big, and
are premalignant lesions. Therefore, mucinous
cystadenomas require surgical excision as they
can transform into mucinous cystadenocarci-
noma. If complete excision is achieved, progno-
sis is excellent.
Mature cystic teratoma is an extragonadal
germ cell tumor that consists of either one large
unilocular or multilocular cysts. Surgical resec-
tion is indicated due to its premalignant nature.
Borderline malignant tumors of the exocrine
pancreas are represented only by one type—that
is the solid pseudopapillary neoplasm, once was
known as Frantz tumor. This tumor, which is typ-
E. Zani-Ruttenstock and A. Zani
ical of adolescent girls and young women, pres-
ents as a large, round, solitary mass that can arise
from any region of the pancreas. Given its nature,
surgical excision is indicated and associated with
excellent prognosis.
Malignant tumors of the exocrine pancreas
include pancreatoblastoma, mucinous cystadeno-
carcinoma, ductal adenocarcinoma, and acinar
cell carcinoma.
Of these, pancreatoblastoma is the most com-
mon malignant pancreatic tumor occurring in
children, more commonly in boys. Patients are
typically young and between 1 and 8 years of age
and often have Beckwith-Wiedemann syndrome
or familial adenomatous polyposis syndrome.
Pancreatoblastoma usually presents as a single,
large mass that can be found anywhere in the
pancreas.
The mucinous cystadenocarcinoma is very
rare and presents similar features to the mucinous
cystadenomas, making the differential diagnosis
between the two difficult. Surgical excision is
mandatory.
The ductal adenocarcinoma of the pancreas is
another very rare tumor in the pediatric age
group. Patients with such neoplasm often have
associated Peutz-Jeghers syndrome, adenoma-
tous familial polyposis syndromes, or hereditary
pancreatic cancer syndrome (Lüttges et al. 2004).
Given the location of these masses that locate
often in the head of the pancreas, patients may
present with signs and symptoms of obstructive
jaundice. Complete surgical excision is the main-
stay of treatment. However, this is only techni-
cally feasible in a minority of patients, who are
then managed with palliative gastrointestinal and
biliary bypass procedures (Cofer 2020).
The acinar cell carcinoma is also a rare epi-
thelial tumor with less than 30 cases reported in
the pediatric literature (Morikawa et  al. 2016).
This neoplasm usually presents as a single large
mass (6–20 cm) that can be located anywhere in
the pancreatic parenchyma. Complete surgical
excision has been associated with good progno-
sis, as metastatic disease has not been reported
(Morikawa et al. 2016).
Benign tumors of the endocrine pancreas
include insulinoma, gastrinoma, and VIPomas.
84  Pancreatic Disorders
The incidence of these tumors is very low in the
pediatric population, and they are usually seen in
adolescent and young adults. When insulinoma
and gastrinoma are associated with multiple
endocrine neoplasia type 1 (MEN 1), they are
considered premalignant (borderline malignant
to malignant).
Insulinoma is a neoplasm that arises from pan-
creatic beta islet cells that hypersecrete insulin
and lead to severe hypoglycemia. The latter is the
typical first sign of the tumor, which is usually
very small and does not produce mass effects or
obstruction symptoms. Once the diagnosis is sus-
pected with laboratory tests, the tumor is studied
with multiphase contrast-enhanced thin slice
cross-sectional imaging. However, insulinomas
that are small and are not detected with CT or
MRI scan are better evaluated with PET/CT scan-
ning (with 18-fluoro-DOPA or with gallium-68
DOTATATE tracers), arterial stimulation/venous
sampling, or transhepatic portal venous sampling
(Peranteau et al. 2013). Surgical excision is indi-
cated even in benign lesions due to their meta-
bolic activity. Small lesions benefit from
pancreas-sparing techniques, whereas larger
masses or those suspected of being malignant are
treated with classic pancreatic resection. If symp-
toms recur after surgery, the patient should be
evaluated for additional lesions (Peranteau et al.
2013).
Gastrinoma is neoplasm that arises from G
cells in the pancreas that hypersecrete gastrin and
lead to hypergastrinemia. G cells are normally
present in the pyloric antrum but can also be
found in the duodenum and the pancreas, the two
organs where gastrinomas can develop. Patients
suffering from these tumors present with compli-
cations of excessive gastric acid production and
occasionally with a clinical syndrome known as
Zollinger-Ellison syndrome (Ellison and Johnson
2009). Once high serum gastrin levels are
detected by laboratory tests, cross-sectional
imaging combined with nuclear scanning (octreo-
tide scan also known as somatostatin receptor
scintigraphy) or endoscopic ultrasonography is
useful to localize the tumor. This is critical for
planning surgical excision, which at present
remains the first-line treatment for gastrinomas.
1169
In sporadic lesions, metastases may be found in
the regional nodes; hence, sampling during surgi-
cal enucleation of the primary lesion is manda-
tory. In gastrinomas associated with MEN-I
syndrome, metastases to nodes and liver are com-
mon; hence, surgery is more aggressive, and
recurrences are to be expected. Prognosis is
highly dependent on the presence and location of
metastases.
84.9 Conclusion
Although uncommon in children, surgical pan-
creatic disorders also affect the pediatric popula-
tion, causing dysfunction of the exocrine or
endocrine component of the gland. Some disor-
ders are the result of structural or genetic condi-
tions and can be detected soon after birth, whereas
others are acquired and present later in life.
Regardless of their presenting symptomatology,
these conditions should be recognized as soon as
possible, since they can lead to long-term
sequelae if left untreated. For some pancreatic
disorders, such as structural anomalies, diagnosis
and treatment have remained fairly similar over
decades. Conversely, aspects of the management
of conditions such as CHI and pancreatitis have
significantly changed. In the latter cases, advance-
ments in genetic testing and imaging studies are
redefining the paradigm of care for children suf-
fering from these conditions.
References
Akbulut S, Yavuz R, Sogutcu N, Kaya B, Hatipoglu
S, Senol A, Demircan F (2014) Hydatid cyst of the
pancreas: report of an undiagnosed case of pancre-
atic hydatid cyst and brief literature review. World J
Gastrointest Surg 6:190–200
Awais M, Rehman A, Baloch NU (2019) Gastric-type
enteric duplication cyst in the tail of a Bifid Pancreas
in Association with complete intestinal malrotation. J
Coll Physicians Surg Pak 29:489–490
Bohl CE, Federico SM, Robinson GW, Bahrami A,
Shulkin BL (2018) FDG-PET CT in the evaluation
of primary and secondary pancreatic malignancies.
Pediatr Blood Cancer 65:e27115
Bülow R, Simon P, Thiel R, Thamm P, Messner P, Lerch
MM, Mayerle J, Völzke H, Hosten N, Kühn JP (2014)
1170
Anatomic variants of the pancreatic duct and their
clinical relevance: an MR-guided study in the general
population. Eur Radiol 24:3142–3149
Cofer BR (2020). Pancreatic tumors. Pediatric sur-
gery NaT.  American Pediatric Surgical Association
APSA Webapp. www.pedsurglibrary.com/
apsa/view/Pediatric-­S urgery-­N aT/829508/all/
Pancreatic_Tumors.)
Dabbas N, Davenport M (2009) Congenital choledochal
malformation: not just a problem for children. Ann R
Coll Surg Engl 91:100–105
Dankovcik R, Jirasek JE, Kucera E, Feyereisl J, Radonak
J, Dudas M (2008) Prenatal diagnosis of annular
pancreas: reliability of the double bubble sign with
periduodenal hyperechogenic band. Fetal Diagn Ther
24:483–490
Diesen DL (2020a) Acute pancreatitis. Pediatric sur-
gery NaT. American Pediatric Surgical Association.
APSA Webapp. www.pedsurglibrary.com/apsa/view/
Pediatric-­Surgery-­NaT/829507/all/Acute_Pancreatitis
Diesen DL (2020b) Chronic pancreatitis. Pediatric
surgery NaT. American Pediatric Surgical
Association. APSA Webapp. www.pedsurglibrary.
com/apsa/view/Pediatric-­Surgery-­NaT/829601/all/
Chronic_Pancreatitis
Dimitriou I, Katsourakis A, Nikolaidou E, Noussios G
(2018) The main anatomical variations of the pancre-
atic duct system: review of the literature and its impor-
tance in surgical practice. J Clin Med Res 10:370–375
Ellison EC, Johnson JA (2009) The Zollinger-Ellison
syndrome: a comprehensive review of historical, sci-
entific, and clinical considerations. Curr Probl Surg
46:13–106
Ferri V, Vicente E, Quijano Y, Ielpo B, Duran H, Diaz E,
Fabra I, Caruso R (2019) Diagnosis and treatment of
pancreas divisum: a literature review. Hepatobiliary
Pancreat Dis Int 18:332–336
Flanagan SE, Kapoor RR, Hussain K (2011) Genetics of
congenital hyperinsulinemic hypoglycemia. Semin
Pediatr Surg 20:13–17
Hafezi M, Mayschak B, Probst P, Büchler MW, Hackert T,
Mehrabi A (2017) A systematic review and quantita-
tive analysis of different therapies for pancreas divi-
sum. Am J Surg 214:525–537
Hameed S, Caro-Domínguez P, Daneman A, Zani-­
Ruttenstock E, Zani A, Navarro OM (2020) The role
of sonography in differentiating congenital intrin-
sic duodenal anomalies from midgut malrotation:
emphasizing the new signs of duodenal and gastric
wall thickening and hyperechogenicity. Pediatr Radiol
50:673–683
Howard JM, Hess W (2002) The early surgeon-anatomists.
In: Howard JM, Hess W (eds) History of the pancreas:
mysteries of a hidden organ, 1st edn. Springer Science
& Business Media, New York
Hunter CJ, Connelly ME, Ghaffari N, Anselmo D,
Gonzalez I, Shin C (2008) Enteric duplication cysts of
the pancreas: a report of two cases and review of the
literature. Pediatr Surg Int 24:227–233
E. Zani-Ruttenstock and A. Zani
Johnson PR, Spitz L (2000) Cysts and tumors of the pan-
creas. Semin Pediatr Surg 9:209–215
Karamanou M, Protogerou A, Tsoucalas G, Androutsos
G, Poulakou-Rebelakou E (2016) Milestones in the
history of diabetes mellitus: the main contributors.
World J Diabetes 7:1–7
Keim V, Witt H, Bauer N, Bodeker H, Rosendahl J, Teich
N, Mossner J (2003) The course of genetically deter-
mined chronic pancreatitis. JOP 4:146–154
Kozu T, Suda K, Toki F (1995) Pancreatic development
and anatomical variation. Gastrointest Endosc Clin N
Am 5:1–30
Laje P, Scott Adzick N (2020) Congenital hyperinsu-
linism. Pediatric surgery NaT, American Pediatric
Surgical Association. APSA Webapp. www.pedsurgli-
brary.com/apsa/view/Pediatric-­SurgeryNaT/829506/
all/Congenital_Hyperinsulinism
Laje P, Stanley CA, Palladino AA, Becker SA, Adzick
NS (2012) Pancreatic head resection and Roux-en-Y
pancreaticojejunostomy for the treatment of the focal
form of congenital hyperinsulinism. J Pediatr Surg
47:130–135
Lecco TM (1910) Zur morphologie des pankreas annu-
lare. Sitzungsb Akad Wissensch 119:391–406
Lord K, Radcliffe J, Gallagher PR, Adzick NS, Stanley
CA, De León DD (2015) High risk of diabetes and
neurobehavioral deficits in individuals with surgically
treated hyperinsulinism. J Clin Endocrinol Metab
100:4133–4139
Lüttges J, Stigge C, Pacena M, Klöppel G (2004) Rare
ductal adenocarcinoma of the pancreas in patients
younger than age 40 years. Cancer 100:173–182
Márta K, Farkas N, Szabó I, Illés A, Vincze Á, Pár G,
Sarlós P, Bajor J, Szűcs Á, Czimmer J, Mosztbacher
D, Párniczky A, Szemes K, Pécsi D, Hegyi P (2016)
Meta-analysis of early nutrition: the benefits of enteral
feeding compared to a Nil Per Os diet not only in
severe, but also in mild and moderate acute pancreati-
tis. Int J Mol Sci 17:1691
Michailidis L, Aslam B, Grigorian A, Mardini H (2017)
The efficacy of endoscopic therapy for pancreas divi-
sum: a meta-analysis. Ann Gastroenterol 30:550–558
Morikawa M, Nakazawa T, Kato S, Fujimoto D, Koneri K,
Murakami M, Hirono Y, Maeda H, Goi T, Katayama
K, Imamura Y, Tanizawa A, Yamaguchi A (2016)
Acinar cell carcinoma of the pancreas in childhood.
Int Cancer Conf J 5:140–145
Morinville VD, Husain SZ, Bai H, Barth B, Alhosh R,
Durie PR, Freedman SD, Himes R, Lowe ME, Pohl
J, Werlin S, Wilschanski M, Uc A, INSPPIRE Group
(2012) Definitions of pediatric pancreatitis and survey
of present clinical practices. J Pediatr Gastroenterol
Nutr 55:261–265
Peranteau WH, Palladino AA, Bhatti TR, Becker SA,
States LJ, Stanley CA, Adzick NS (2013) The surgical
management of insulinomas in children. J Pediatr Surg
48(12):2517–2524
Perez EA, Gutierrez JC, Koniaris LG, Neville HL,
Thompson WR, Sola JE (2009) Malignant ­pancreatic
84  Pancreatic Disorders
tumors: incidence and outcome in 58 pediatric
patients. J Pediatr Surg 44:197–203
Pierro A, Nah SA (2011) Surgical management of con-
genital hyperinsulinism of infancy. Semin Pediatr
Surg 20:50–53
Poddar U, Yachha SK, Borkar V, Srivastava A, Kumar
S (2017) A report of 320 cases of childhood pancre-
atitis: increasing incidence, etiologic categorization,
dynamics, severity assessment, and outcome. Pancreas
46:110–115
Poropat G, Giljaca V, Hauser G, Štimac D (2015) Enteral
nutrition formulations for acute pancreatitis. Cochrane
Database Syst Rev 3:CD010605
Quest L, Lombard M (2000) Pancreas divisum: opinion
divisa. Gut 47:317–319
Rebours V, Boutron-Ruault MC, Schnee M, Férec
C, Le Maréchal C, Hentic O, Maire F, Hammel
P, Ruszniewski P, Lévy P (2009) The natural his-
tory of hereditary pancreatitis: a national series. Gut
58:97–103
Restrepo R, Hagerott HE, Kulkarni S, Yasrebi M, Lee EY
(2016) Acute pancreatitis in pediatric patients: demo-
graphics, etiology, and diagnostic imaging. AJR Am J
Roentgenol 206:632–644
Rustagi T, Njei B (2014) Magnetic resonance cholan-
giopancreatography in the diagnosis of pancreas divi-
sum: a systemic review and meta-analysis. Pancreas
43:823–828
Sacco Casamassima MG, Goldstein SD, Yang J, Gause
CD, Abdullah F, Meoded A, Makary MA, Colombani
PM (2017) The impact of surgical strategies on out-
comes for pediatric chronic pancreatitis. Pediatr Surg
Int 33:75–83
Sadler TW (2019) Langman’s medical embryology, 14th
edn. Wolters Kluwer, Philadelphia
1171
Testoni PA (2014) Acute recurrent pancreatitis: etio-
pathogenesis, diagnosis and treatment. World J
Gastroenterol 20:16891–16901
Trout AT, Serai SD, Fei L, Sun Q, Abu-El-Haija M (2018)
Prospective assessment of normal pancreatic secretory
function measured by MRI in a cohort of healthy chil-
dren. Am J Gastroenterol 113:1385
Villatoro E, Mulla M, Larvin M (2010) Antibiotic therapy
for prophylaxis against infection of pancreatic necro-
sis in acute pancreatitis. Cochrane Database Syst Rev
2010:CD002941
Zani A, Nah SA, Ron O, Totonelli G, Ismail D, Smith
VV, Ashworth M, Blankenstein O, Mohnike W,
De Coppi P, Eaton S, Hussain K, Pierro A (2011)
The predictive value of preoperative fluorine-18-­
L-3,4-­dihydroxyphenylalanine positron emission
tomography-­computed tomography scans in children
with congenital hyperinsulinism of infancy. J Pediatr
Surg 46:204–208
Zani A, Yeh JB, King SK, Chiu PP, Wales PW (2017)
Duodeno-duodenostomy or duodeno-jejunostomy for
duodenal atresia: is one repair better than the other?
Pediatr Surg Int 33:245–248
Zani-Ruttenstock E, Davenport M (2016) Biliary atre-
sia and choledocal malformations. In: Guandalini S,
Dhawan A, Branski D (eds) Pediatric gastroenter-
ology, hepatology and nutrition, 1st edn. Springer,
London
Zyromski NJ, Sandoval JA, Pitt HA, Ladd AP, Fogel EL,
Mattar WE, Sandrasegaran K, Amrhein DW, Rescorla
FJ, Howard TJ, Lillemoe KD, Grosfeld JL (2008)
Annular pancreas: dramatic differences between chil-
dren and adults. J Am Coll Surg 206:1019–1025
 Splenic Disorders
Takashi Doi and Thom E. Lobe
85.1 Introduction
The spleen is an important peripheral immune
organ, which has many functions, such as regu-
lating the circulating blood volume, blood filtra-
tion, production of a variety of immunoglobulins
and opsonins, and regulation of the endocrine
system. The primary immunologic function of
the spleen is to filter out virulent pathogens and
antigens. It has the ability to store blood and con-
tract in response to acute changes in blood vol-
ume and it becomes the focus of surgical attention
in a variety of circumstances that we will discuss
in detail in this chapter.
85.2 Anatomy and Physiology
The spleen serves as a hemopoietic organ during
the hepatosplenothymic phase (beginning in the
second month). The spleen mainly produces cells
of the erythroid lineage in the fetus and lympho-
cytes in the newborn. Thymic and B lymphocytes
migrate to the spleen and populate the white pulp
T. Doi (*)
Department of Pediatric Surgery, Kansai Medical
University, Osaka, Japan
e-mail: doitak@hirakata.kmu.ac.jp
T. E. Lobe
Pediatric Surgery, University of Illinois,
Chicago, IL, USA
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_85
85
associated with the trabeculae (the sinuses con-
tain erythroblastic tissue, or red pulp). The spleen
contains phagocytes that remove “exhausted”
erythrocytes and foreign bodies from the blood,
and it is a major reservoir for blood. Absence of
the spleen predisposes to certain infections.
The organ is located in the upper left part of
the abdomen, behind the stomach and just below
the diaphragm. In normal adult individuals, this
organ measures about 125 × 75 × 50 mm (5 × 3 ×
2 in.) in size, with an average weight of 150 g. It
consists of masses of lymphoid tissue of granular
appearance located around fine terminal branches
of veins and arteries.
These vessels are connected by modified cap-
illaries called splenic sinuses.
The splenic artery, the largest branch of the
celiac artery, is remarkable for the tortuosity of
its course as it passes horizontally to the left side,
behind the stomach and the omental bursa of the
peritoneum, and along the upper border of the
pancreas. Some surgeons identify and ligate the
splenic vessel in this location before splenectomy
to minimize the risk of hemorrhage. Preoperative
embolization of the splenic artery has also been
described. The artery is accompanied by the
splenic vein, which lies below it; the artery
crosses in front of the upper part of the left kid-
ney and, on arriving near the spleen, divides into
branches, some of which enter the splenic hilum
between the two layers of the phrenicosplenic
ligament to be distributed to the tissues of the
1173
1174
Fig. 85.1  Splenic vascular anatomy
spleen; some go to the pancreas, while others
pass to the greater curvature of the stomach
between the layers of the gastrosplenic ligament
(Fig. 85.1).
The short gastric arteries are a group of short
arteries arising from the terminal splenic artery
and the left gastroepiploic artery. They pass from
left to right, between the layers of the gastro-
splenic ligament, and are distributed to the greater
curvature of the stomach, anastomosing with
branches of the left gastric and left gastroepiploic
arteries.
Approximately 10% of people have one or
more accessory spleens. They may form near the
hilum of the main spleen, the junction at which
the splenic vessels enter and leave the organ.
There are several peritoneal ligaments that
support the spleen: the gastrosplenic ligament
connects the stomach to spleen, the splenorenal
ligament connects the spleen to the kidney, and
the phrenicocolic ligament connects the left colic
flexure to the thoracic diaphragm—the middle of
it connects to the spleen.
The splenic red pulp is composed of “sinuses”
(or “sinusoids”), which are filled with blood, and
“splenic cords” of reticular fibers and “marginal
zone” bordering on white pulp. Its function is
T. Doi and T. E. Lobe
mechanical filtration  – to remove unwanted
materials from the blood, including senescent red
blood cells.
White pulp, on the other hand, is composed of
nodules called Malpighian corpuscles. These are
composed of “lymphoid follicles” (or “folli-
cles”), rich in B lymphocytes, and “periarteriolar
lymphoid sheaths,” rich in T lymphocytes. Its
function is to help fight infections.
Other functions of the spleen are less promi-
nent including the production of opsonins, pro-
perdin, and tuftsin.
The creation of red blood cells occurs in the
spleen up until the fifth month of gestation; the
spleen has important hematopoietic functions.
After birth, no significant hematopoietic function
is left in the spleen except in some hematologic
disorders, e.g., myelodysplastic syndrome and
hemoglobinopathies.
85.3 Pathology
The spleen is the largest collection of lymphoid
tissue in the body. It is normally palpable in pre-
term infants, in 30% of normal, full-term neo-
nates, and 5–10% of infants and toddlers. A
spleen easily palpable below the costal margin in
any child over the age of 3–4 years should be
considered abnormal until proven otherwise.
Splenomegaly can result from antigenic stimu-
lation (e.g., infection), obstruction of blood flow
(e.g., portal vein obstruction), underlying func-
tional abnormality (e.g., hemolytic anemia), or
infiltration (e.g., leukemia or storage disease, such
as Gaucher’s disease). The most common cause of
acute splenomegaly in children is viral infection,
which is transient and usually moderate. Basic
workup for acute splenomegaly includes a com-
plete blood count with differential, platelet count,
and reticulocyte and atypical lymphocyte counts
to exclude hemolytic anemia and leukemia.
Assessment of IgM antibodies to viral capsid anti-
gen (a rising titer) is indicated to confirm Epstein-
Barr virus or cytomegalovirus. Other infections
should be excluded if these tests are negative.
Conditions for which splenectomy is indi-
cated are listed in Table 85.1.
85  Splenic Disorders
Table 85.1  Diseases or conditions for which splenec-
tomy is indicated
Traumatic injury to spleen
Hematologic diseases including hereditary
spherocytosis; thalassemia major; certain forms of
immune thrombocytopenic purpura (ITP) unresponsive
to medical management; myeloproliferative disorders;
thrombotic thrombocytopenic purpura (TTP)
Hydatid cysts
Hodgkin’s disease (clinical stage I-A or II-A)
Intraoperative splenic injury
Splenic abscesses
Nonparasitic cysts that are symptomatic and greater
than 5 cm
Sinistral portal hypertension secondary to isolated
splenic vein thrombosis or obstruction
Splenic mass presumed to be a primary or undiagnosed
neoplasm, malignancy in an adjacent organ
Metabolic storage diseases such as Gaucher’s disease
Infarcted wandering spleen
85.4 Splenic Trauma
The spleen is the most commonly injured solid
organ in pediatric blunt trauma patients (25–
30%). At present, they are classified accord-
ing to the anatomy of the injury. The optimal
treatment strategy, however, should take into
consideration the hemodynamic status, the ana-
tomic derangement, and the associated injuries.
The ­management of splenic trauma has changed
considerably in the last few decades, especially
in favor of nonoperative management which
ranges from observation and monitoring alone
to angiography/angioembolization, with the aim
to preserve the spleen and its function, espe-
cially in children. Most injuries to the spleen in
children heal spontaneously. Severe trauma can
cause the spleen or its blood vessels to rupture
or fragment (Coccolini et al. 2017).
85.4.1 Causes and Symptoms
The most common cause of injury to the spleen is
blunt abdominal trauma. Blunt trauma is often
caused by a direct blow to the belly, car and
motorcycle accidents, falls, sports mishaps, and
fights. The spleen is the most commonly injured
organ in blunt abdominal trauma; splenic injury
1175
occurs in nearly 25% of injuries of this type.
Penetrating injuries, such as those from stabbing,
gunshot wounds, and accidental impaling, also
account for cases of splenic trauma, although far
less frequently than blunt trauma.
A spleen that has become enlarged and fragile
from disease is sometimes ruptured by a doctor
or medical student in the course of palpating
(feeling) the patient’s abdomen or damaged by a
surgeon in the course of an operation on other
abdominal organs.
Damage to the spleen may cause localized or
general abdominal pain, tenderness, and swell-
ing. Fractured ribs may be present. Splenic
trauma may cause mild or severe internal bleed-
ing, leading to shock and for which symptoms
include rapid heartbeat, shortness of breath,
thirst, pale or clammy skin, weak pulse, low
blood pressure, dizziness, fainting, sweating,
vomiting blood, blood in the stools or urine,
deterioration of vital signs, and loss of con-
sciousness (Coccolini et al. 2017).
85.4.2 Diagnosis
The goal of diagnosis of all abdominal traumas
is to detect and treat life-threatening injuries as
quickly as possible. Ultrasonography—particu-
larly focused abdominal sonographic technique
(FAST)—has now become a standard bedside
technique in many hospitals to check for bleed-
ing in the abdomen. Imaging tests allow doctors
to determine the necessity and type of surgery
required. The contrast CT scan is the gold stan-
dard for the evaluation of blunt abdominal trauma.
However, patients should be hemodynamically
stable, as well as cooperative or sedated. MRI
tests are accurate but costly and less available
in some hospitals, while radionuclide scanning
requires more time (Coccolini et al. 2017).
Peritoneal lavage is another diagnostic tech-
nique in which the abdominal cavity is entered
and flushed to check for bleeding. When patients
exhibit shock, infection, or prolonged inter-
nal bleeding, exploratory laparoscopy may be
warranted.
1176
85.4.3 Treatment
Improved techniques of diagnosis and monitoring
(particularly the introduction of CT scans) as well
as understanding that removal of the spleen cre-
ates future risk of a lowered capacity to fight
infection have modified treatment approaches.
Research over the past two decades has shown
that the spleen has high healing potential and con-
firmed that children are more susceptible to infec-
tion after splenectomy (postsplenectomy sepsis,
PSS). PSS has a mortality rate of over 50%, and
standard procedure now avoids splenectomy as
much as possible (Coccolini et al. 2017).
85.4.4 Nonoperative Treatment
Today, nonoperative treatemt is successful in
95–100% of blunt pediatric trauma patients and
has therefore become the gold standard of treat-
ment in children who have sustained an isolated
blunt splenic injury and are hemodynamically
stable at the time of presentation. Angiography/
angioembolization may only be considered in
carefully selected patients, such as those with
high-grade injuries, transient response to resusci-
tation, and/or persistent blood requirements
(Coccolini et al. 2017; Casale & Perrotta 2011).
85.4.5 Prognosis
The ample blood supply to the spleen can pro-
mote rapid healing. Studies have shown that intra-
abdominal bleeding associated with splenic
trauma stops without surgical intervention in up
to two out of three cases in children. When trauma
patients stabilize during nonoperative therapy,
chances are high that surgery will be avoided and
that spleen injuries will heal themselves. Splenic
trauma patients undergoing diagnostic tests such
as CT and MRI scans have improved chances of
avoiding splenectomy and retaining whole or par-
tial spleen (Coccolini et al. 2017).
T. Doi and T. E. Lobe
85.5 Hematologic Diseases
Indications for splenectomy should be deter-
mined with the close cooperation of a hematolo-
gist/oncologist. Common indications include
hereditary spherocytosis, thalassemia major, and
certain forms of immune thrombocytopenic pur-
pura (ITP) unresponsive to medical management.
Myeloproliferative disorders may lead to massive
splenomegaly and can cause symptoms that are
best relieved by splenectomy, primarily for
symptomatic relief (Alqadi & Saxena 2018;
Schifferli & Kühne 2013).
85.6 Hodgkin’s Disease
Selected patients with clinical Stage I-A or II-A
Hodgkin’s disease may be candidates for a stag-
ing laparotomy or laparoscopy. In the absence of
obvious liver or intra-abdominal nodal disease,
splenectomy is an integral part of the staging pro-
cedure to exclude splenic involvement, which
would alter the method of treatment.
85.7 Iatrogenic (Intraoperative)
Splenic Injury
The spleen may be injured inadvertently during
the performance of intraperitoneal procedures,
especially those involving the distal esophagus,
stomach, distal pancreas, or splenic flexure of the
colon. These injuries may occur directly from
operative retractors or, more often, secondary to
inadvertently avulsed capsular adhesions that can
lead to persistent bleeding. Hemostasis should be
attempted using suture plication, topical hemo-
static agents (including absorbable mesh), elec-
trocautery, or argon beam coagulation so that
splenectomy is not required. However, if rapid
hemostasis is not possible, hemorrhage severe
enough to require blood transfusion is better
managed by formal splenectomy than by repeated
attempts at splenic salvage.
85  Splenic Disorders
85.8 Other Indications
for Splenectomy
Less common indications for splenectomy
include splenic abscesses, cysts, sinistral portal
hypertension secondary to isolated splenic vein
thrombosis or obstruction, or splenic mass pre-
sumed to be a primary or undiagnosed neoplasm.
Splenectomy is occasionally included in en bloc
resection for malignancy in an adjacent organ,
such as the stomach, colon, adrenal gland, or
pancreas. Distal pancreatectomy may include
splenectomy if preservation of the splenic artery
and vein is either contraindicated (malignancy)
or technically impossible.
85.9 Prophylaxis Against
Postsplenectomy Sepsis
The absence of a spleen predisposes to some sep-
ticemia infections. Vaccination and antibiotic
measures are discussed under asplenia. Some
people congenitally completely lack a spleen,
although this is rare.
The absence of the spleen is best confirmed
with a technetium-99m radionuclide scan. This
agent is taken up by the reticuloendothelial cells
and enables better assessment of splenic
function.
Sickle-cell disease can cause a functional
asplenia (or autosplenectomy) by causing infarc-
tions of the spleen during repeated sickle-cell
crises.
Medical care involves four key components:
antibiotic prophylaxis, appropriate immuniza-
tion, aggressive management of suspected infec-
tion, and parent education (Shelby et al. 2020).
85.9.1 Antibiotic Prophylaxis
Antibiotic prophylaxis with penicillin should be
initiated immediately upon the diagnosis of
asplenia because these patients are at significant
risk of pneumococcal infections. For children
younger than 2 years, oral penicillin V may be
given twice a day. Amoxicillin has also been rec-
1177
ommended as an appropriate prophylactic antibi-
otic. Erythromycin is an alternate choice in
patients who are allergic to penicillin (Shelby et
al. 2020).
In general, antimicrobial prophylaxis should
be considered for all children with asplenia or
splenic dysfunction until 5 years of age and for at
least 1 year after surgical splenectomy. Some
experts recommend continuing prophylaxis into
adulthood, particularly for high-risk patients. The
need for long-term prophylaxis for children who
have splenectomy after trauma has not been
agreed upon (Dionne et al. 2017).
85.9.2 Immunization
Immunization remains a key measure to prevent
overwhelming infections in the asplenia/postsple-
nectomy patients. All patients should receive all
standard childhood and adolescent immunizations
including pneumococcal conjugate and/or poly-
saccharide, Haemophilus influenzae type b conju-
gate, and meningococcal conjugate and/or
polysaccharide vaccines, at the recommended age
(Shelby et al. 2020).
Approximately 80% of the pediatric pneumo-
coccal bacteremias in the United States are
caused by the seven serotypes covered in the vac-
cine: 4, 6B, 9V, 14, 18C, 19F, and 23F. The con-
jugated vaccine has been effective in dramatically
reducing the occurrence of invasive pneumococ-
cal disease. In children younger than 2 years, the
incidence of all invasive pneumococcal infec-
tions has decreased by 80% after conjugated vac-
cine was recommended in the routine childhood
immunization schedule. Infections caused by
vaccine and vaccine-related serotypes have
decreased by 90% in older children and adults
(Dionne et al. 2017).
The immunization schedule for pneumococcal
conjugate vaccine (PCV7) consists of a primary
series of four doses (0.5 ml each) at age 2, 4, 6,
and 12–15 months. Catch-up immunization
schedules are published regarding appropriate
dosing schedules for children aged 5 years or
younger. Administration of a single dose of
PCV7 to children of any age is not contraindi-
1178
cated, especially for patients with asplenia or
splenic dysfunction who are at high risk for inva-
sive pneumococcal disease.
The pneumococcal polysaccharide vaccine
against 23 serotypes (PPV23) should be given
after age 24 months for supplemental protection.
PCV7 should be administered first, with adminis-
tration of PPV23 at least 8 weeks after the last
dose of PCV7. A booster dose 3–5 years after the
first dose of PPV23 is appropriate.
Patients should also receive quadrivalent
meningococcal vaccine. Two licensed meningo-
coccal vaccines are available in the United States
against serotypes A, C, Y, and W-135.
Meningococcal conjugate vaccine (MCV4) was
licensed in 2005 for people aged 11–55 years.
Meningococcal polysaccharide vaccine (MPSV4)
is licensed for children aged 2 years and older.
Immunization with MPSV4 is recommended for
children aged 2–10 years and older who have
functional or anatomic asplenia. MCV4 should be
used if the patient is aged 11 years or older.
Because of its ability to induce a T-cell response,
MCV4 is expected to confer immunity for approx-
imately 10 years, as opposed to about 4 years of
immunity from MPSV4. Immunization with
MCV4 should be considered in adolescents aged
3–5 years after receiving MPSV4. Revaccination
schedules with MCV4 are ongoing.
The recommended vaccination schedule for
H influenzae type b is a primary series of three
doses given at age 2, 4, and 6 months or two
doses given at age 2 and 4 months, depending
on the particular conjugate vaccine product
administered. A booster dose at 12 months old is
recommended for all vaccine products. Children
who are undergoing scheduled splenectomy
after completion of their primary series and
booster dose may benefit from an additional
dose of conjugate vaccine at least 7–10 days
before surgery. Catch-up immunization sched-
ules regarding H influenzae type b vaccine are
published.
Yearly influenza vaccines is also recom-
mended to minimize the likelihood of secondary
bacterial infections.
T. Doi and T. E. Lobe
85.9.3 Management of Suspected
Infection
The risk of serious bacterial infection is ever
present in these patients. Many patients have triv-
ial symptoms yet rapidly develop fulminant sep-
sis and death within hours.
All patients with impaired splenic function
with suspected infection must be urgently and
promptly evaluated. Obtain blood; urine; and, if
indicated, cerebrospinal fluid (CSF) cultures.
Initiate broad-spectrum intravenous antibiotics
effective against S pneumoniae, H influenzae
type b, and N meningitidis. Second- or third-­
generation cephalosporins may be the initial
choices. If multiple-drug resistance is a concern,
vancomycin should be added to the regimen. In
addition, many patients require supportive care
with intravenous fluids, volume expanders, and
pressor support.
Because of the potential rapid progression of a
serious bacterial infection, some experts recom-
mend that asplenic patients have access to “stand-
­by” antibiotics, which can be initiated at the first
sign of infection (fever, chills, or malaise). That
the initiation of “stand-by” antibiotics is not a
substitute for seeking immediate medical atten-
tion at the onset of an illness cannot be
overemphasized.
Patients with asplenia are at an increased risk
of sepsis, shock, and meningitis secondary to
Capnocytophaga canimorsus resulting from dog,
cat, or rodent bites. The diagnosis may be made
by means of Gram staining of the buffy coat,
blood, and CSF cultures. Early treatment with
penicillin is the therapy of choice, but cephalo-
sporins, clindamycin, and erythromycin may also
be used (Dionne et al. 2017).
85.10 Wandering Spleen
Wandering spleen is a rare condition. It derives
from the absence or laxity of the lineal ligaments
that hold organ in its normal position in the upper
left abdomen. There are no genetic factors
85  Splenic Disorders
reported for a congenital wandering spleen. The
spleen is suspended by a stalk-like tissue sup-
plied with hilar vessels (vascular pedicle). If the
pedicle is twisted in the course of the movement
of the spleen, the blood supply may be inter-
rupted or blocked (ischemia) to the point of
severe damage to the blood vessels (infarction).
Because there is little or nothing to hold it in
place, the spleen “wanders” in the lower abdo-
men or pelvis where it may be mistaken for an
unidentified abdominal mass.
Symptoms of wandering spleen are typically
those associated with an abnormally large size of
the spleen (splenomegaly) or the unusual posi-
tion of the spleen in the abdomen. Enlargement is
most often the result of twisting (torsion) of the
splenic arteries and veins or, in some cases, the
formation of a blood clot (infarct) in the spleen.
Torsion of the spleen may also draw the torsion
of the distal pancreas or the stomach.
Thrombocytosis is the hallmark of an infarcted or
recently removed spleen.
85.11 Surgical Approaches
The technical aspects of removing the spleen
depend on the specific indication and the experi-
ence of the surgeon. There appears to be no data
to support the routine use of prophylactic antibi-
otics in the perioperative period.
The traditional approach is an open one. In
cases of traumatic injury to the spleen and in
cases of massive splenomegaly, a midline inci-
sion may be preferred. Uncontrolled hemorrhage
from traumatic splenic injury will immediately
focus the surgeon’s attention on the left upper
quadrant of the open abdomen. Coagulated blood
is quickly removed to facilitate visualization and
the surgeon (standing on the patient’s right side
with the patient supine) ups the left hand on the
spleen as the splenoreno and splenophreno liga-
ments are quickly divided either with scissors or
bluntly with the surgeon’s right hand so that the
spleen can be delivered from the surgical wound
for inspection.
At that point, the surgeon will assess whether
the spleen can be repaired. Preoperative images
1179
may suggest the degree of injury so that the sur-
geon can anticipate the findings; however, direct
inspection will be required to make a final
determination.
If the spleen is amenable to repair, this can be
done using any of a variety of techniques. These
include using topical agents to promote coagula-
tion, direct suturing, enveloping the fractured
spleen in surgical mesh, and other techniques.
Ischemic segments can be discarded. The impor-
tant point is to control ongoing hemorrhage. If
hemorrhage cannot be controlled, then removal is
justified.
All forms of splenectomy incorporate similar
surgical steps: the spleen must be mobilized from
its ligamentous (peritoneal) attachments; the
short-gastric vessels must be divided; the colon
must be dissected free of the tail of the pancreas
and lower pole of the spleen; the splenic arteries
and veins must be divided; the pancreas should
be preserved and left intact; any accessory spleen
should be searched for and removed, and in the
case of hematological diseases, the spleen must
be removed and hemostasis assessed.
In the case of trauma, all this happens rather
quickly so as to prevent further hemorrhage.
When an open midline approach is used for
massive splenomegaly, the incision should be
extended sufficiently to allow the surgeon easy
access to all these maneuvers and to prevent acci-
dental rupture of the spleen. The surgeon should
not have to struggle to perform the operation.
A lateral open approach is used by many for
elective splenectomy, primarily for hematologi-
cal or storage diseases. The difference in the
case is principally the type of incision. The
patient, in this instance, is positioned in a lateral
decubitus position with the left side up. A roll is
usually placed under the patient’s right flank to
accentuate elevation of the left side and to flex
the patient to help get the lower ribs out of the
way. A transverse incision then is made through
the layers of the abdominal wall just below or
off the tip of the 12th rib. This incision is not
ideal for cases of hematological diseases that
require concomitant cholecystectomy for chole-
lithiasis as the gallbladder is rather inaccessible
from this approach.
1180
To avoid the potential disadvantages related to
open surgery, surgeons are now drawing their
attention to the laparoscopic techniques.
Laparoscopic splenectomy was first reported in
1993 and quickly gained wide acceptance among
surgeons. Theoretically, laparoscopic splenec-
tomy has the advantage of a shorter hospital stay,
improved cosmesis, and a faster return to daily
activities. However, several concerns about lapa-
roscopic splenectomy still remain. Some authors
doubt the ability of this technique in massive
splenomegaly, while others concern about the
potential risk of overlooking an accessory spleen
in laparoscopic splenectomy (Liu & Fan 2019).
Trocar positions and number will vary depend-
ing on the surgical goal and the surgeon’s prefer-
ence. Laparoscopic splenectomy can be
performed using as few as three strategically
placed trocars, one for the telescope (usually
placed in the umbilicus) and two for instrumenta-
tion. One or more additional trocars may facili-
tate retraction and dissection and can be placed at
the surgeon’s discretion (Fig. 85.2).
The patient’s position is also a matter of pref-
erence and depends somewhat on the planned
procedure.
For simple splenectomy, these authors prefer
to operate between the legs (the so-called “French
Style”) and aim the telescope and instruments
toward the upper abdomen. The patient then is
supine and in a reversed Trendelenburg position,
perhaps with the table rotated to elevate the left
side. This position allows for maximal access and
permits the intestines to gravitate toward the pel-
Fig. 85.2  Trocar sites with patient in lateral position.
Spleen in sac with neck exteriorized from umbilicus
T. Doi and T. E. Lobe
vis and away from the left side so as to best visu-
alize the spleen. This approach also allows easy
access to the gallbladder for its removal when
indicated. And for surgeons who prefer to use a
surgical robot, little has to change to accommo-
date the robot.
Other surgeons prefer to position the patient in
a lateral position or to start supine and move the
patient after the initial trocars have been placed.
These authors have tried every position and found
that approaching the upper abdomen as described
above is the simplest and most reliable in his
hands (Liu & Fan 2019).
We divide the short gastric vessels first as
cephalad as is feasible without risking hemor-
rhage. We then divide the lateral attachments and
carry this dissection cephalad and medial. We
often leave a small attachment to the diaphragm
intact so the spleen when freed will stay in place
when trying to place it in a bag. We then divide
the splenic vessels at the hilum. This can be done
individually or en masse and can be done with an
energy device, with sutures or with an endoscopic
stapler, depending on the size of the patient and
on the expertise of the surgeon (Fig. 85.3).
We then use an endoscopic pouch or bag
inserted through one of the trocar sites to snare
the freed spleen. If we have left a small peritoneal
attachment to the diaphragm, we divide it after
the spleen is in the pouch (Fig.  85.4). We then
withdraw the open end of the pouch through the
Fig. 85.3  Laparoscopic view of splenic hilar dissection
85  Splenic Disorders
Fig. 85.4  Spleen in sac for laparoscopic extraction
Fig. 85.5  Morcellated spleen
trocar site and fracture or morcellate the spleen
so that it can be removed in segments without
spilling its contents in the abdomen (Fig. 85.5).
Depending on the surgeon’s preference, this can
be accomplished with a finger, a sponge forceps
or other similar device. The important point is not
to rupture the bag and spill its contents (Liu &
Fan 2019).
After the spleen is removed, the left upper
abdomen is inspected for hemostasis. If the gall-
bladder is to be removed, we usually do this after
the spleen has been removed successfully.
Careful trocar placement can obviate the need to
insert additional trocars for the cholecystectomy.
Some pediatric hematologists advocate partial
splenectomy for certain conditions such as hered-
itary spherocytosis. Partial splenectomy is a good
method to prevent postsplenectomy infections by
preservation of the immunologic role of the
spleen. Laparoscopic partial splenectomy can be
1181
performed safely when performed by e­ xperienced
laparoscopic surgeons. We divide segmental ves-
sels as they are encountered and after the seg-
ment to be removed is devascularized, we use an
energy device to hemostatically divide the paren-
chyma at the line of ischemia (Stylianos 2019;
Guizzetti 2016; Schimmer et al. 2016).
Splenic autotransplantation with a free graft
for maintenance of specific splenic immunity is
still experimental and of unproven efficacy.
In the case of the wandering spleen, the spleen
can be fixed to the abdominal wall, diaphragm,
directly, or using omentum and surgical mesh.
Conventional laparotomy is challenged recently
by the laparoscopic approach (Feng et al. 2016).
The laparoscopic approach has also proved
beneficial in the management of congenital and
acquired splenic cysts. Splenic cysts may be a
form of cystic hygroma within the splenic paren-
chyma or may be the result of a traumatic injury.
These can be identified laparoscopically, and
their wall can be excised. Simple fenestration is
possible but often results in recurrance (Liu &
Fan 2019).
Echinococcal or hydatid cysts are rare and
also can be treated laparoscopically. Their diag-
nosis should be made preoperatively by labora-
tory and imaging characteristics that are
diagnostic. Under laparoscopic view, these cysts
can be aspirated and refilled with hypertonic
saline.
All patients with hydatid disease should be
considered for percutaneous or surgical treatment
because of the risk of life-threatening complica-
tions of untreated disease. More complicated
1182
Fig. 85.6  Post-op appearance of abdomen
cysts are better managed by open surgery.
Treatment of hydatid cysts is associated with two
technical problems: risk of anaphylaxis from
spillage of cyst fluid containing eggs and larvae
into the peritoneal cavity and recurrence caused
by residual eggs in incompletely removed germi-
nal membranes. To prevent these problems, most
surgeons use a technique in which the cyst con-
tents are aspirated and replaced with a hypertonic
saline solution to kill residual daughter cysts in
the germinal membrane before unroofing and
pericystectomy. The goal of the latter procedure
is to excise the germinal membrane, leaving the
inflammatory and fibrous components of the cyst
wall in situ.
We no longer leave a nasogastric tube in chil-
dren undergoing splenectomy unless we are wor-
ried about the integrity of the stomach where the
short-gastric vessels were divided. Similarly, we
do not leave a urinary bladder catheter in place
after surgery.
T. Doi and T. E. Lobe
Instead, we encourage patients to ambulate
immediately, and we advance their diet as toler-
ated from the immediate postoperative period
(Fig. 85.6).
85.12 Complications and Adverse
Effects
The risk of postsplenectomy infections depends
on the etiology of the disease. The majority of
infections occur within 2 years after splenectomy
and are fatal in almost 50% of cases. The inci-
dence is low in patients with spherocytosis, ITP,
and splenic trauma and higher in Hodgkin’s dis-
ease (8%) (Shelby et al. 2020).
Splenoportal thrombosis following splenec-
tomy is a rare complication. The true rate of
thrombotic complications after splenectomy is not
defined, but early diagnosis and prompt initiation
of anticoagulant therapy is mandatory. Port site
splenosis has been reported after laparoscopic
splenectomy and has to be considered in the dif-
ferential diagnosis of port site pain and a palpable
nodule. Leaks along the pancreatic tail can also
rarely occur after splenectomy (Liu & Fan 2019).
Other postoperative complications of splenec-
tomy depend somewhat on the procedure.
Wound complications may be seen after an
open splenectomy, though the frequencies of
these appear to be independent of the procedure
and are more closely related with surgical tech-
nique. Trocar site infection, hemorrhage, and
herniation can occur, but these are rare and can be
related more to trocar technique than to splenec-
tomy (Liu & Fan 2019).
Patients undergoing open procedures may be
more prone to pulmonary complications, though
we have seen one case of pneumonia after a lapa-
roscopic splenectomy in a teenage girl with
hematological disease who was poorly motivated
to ambulate after surgery. In general, atelectasis
and pneumonia are more closely correlated with
open incisions (Liu & Fan 2019).
Intraabdominal infections and left upper quad-
rant abscess are other potential complications,
but we have not seen this in laparoscopic surgery.
85  Splenic Disorders
When this occurs, it is probably related to inade-
quately controlled hemorrhage in the surgical
bed.
Finally, high-grade injuries to the spleen are at
risk for delayed presentation of splenic artery
aneurysms. We recommend examination for
abdominal bruit and follow-up abdominal ultra-
sounds in all high-grade (III and IV) injuries.
In general, the laparoscopic approach tends to
be associated with fewer adhesions, and thus,
fewer postoperative bowl obstructions when
compared to open procedures (Shelby et al.
2020). This seems to hold true for laparoscopic
splenectomy.
The results of laparoscopic splenectomy in
children are excellent. Recently, the da Vinci
robotic surgery system has been used to remove
the spleen from children (Rice et al. 2012).
85.13 Conclusion
The spleen is the most commonly injured solid
organ in pediatric blunt trauma patients.
Nonoperative management of splenic injuries has
become the treatment of choice in hemodynami-
cally stable patients. Laparoscopic splenectomy
has gained worldwide acceptance for spleen
removal, especially in hematological disorders in
children. The laparoscopic splenectomy is associ-
ated with fewer adhesions and thus fewer postop-
erative bowl obstructions when compared to open
splenectomy. New techniques and smaller instru-
ments have facilitated recent advances in the tech-
nical aspects of laparoscopic splenectomy.
1183
References
Alqadi GO, Saxena AK (2018) Is laparoscopic approach
for wandering spleen in children an option? J Minim
Access Surg 15(2):93–97
Casale M, Perrotta S (2011) Splenectomy for hereditary
spherocytosis: complete, partial or not at all? Expert
Rev Hematol 4(6):627–635
Coccolini F, Montori G, Catena F et  al (2017) Splenic
trauma: WSES classification and guidelines for adult
and pediatric patients. World J Emerg Surg 18(12):40
Dionne B, Dehority W, Brett M, Howdieshell TR (2017)
The Asplenic patient: post-insult immunocompetence,
infection, and vaccination. Surg Infect 18(5):536–544
Feng S, Qiu Y, Li X et al (2016) Laparoscopic versus open
splenectomy in children: a systematic review and
meta-analysis. Pediatr Surg Int 32(3):253–259
Guizzetti L (2016) Total versus partial splenectomy
in pediatric hereditary spherocytosis: a systematic
review and meta-analysis. Pediatr Blood Cancer
63(10):1713–1722
Liu G, Fan Y (2019) Feasibility and safety of laparoscopic
partial splenectomy: a systematic review. World J Surg
43(6):1505–1518
Rice HE, Crary SE, Langer JC, Kemper AR (2012)
Splenectomy in Congenital Hemolytic Anemia
Consortium. Comparative effectiveness of different
types of splenectomy for children with congenital
hemolytic anemias. J Pediatr 160(4):684–689
Schifferli A, Kühne T (2013) Chronic immune thrombo-
cytopenia in children: who needs splenectomy? Semin
Hematol 50(Suppl 1):S58–S62
Schimmer JA, van der Steeg AF, Zuidema WP (2016)
Splenic function after angioembolization for splenic
trauma in children and adults: a systematic review.
Injury 47(3):525–530
Shelby R, Kulaylat AN, Villella A, Michalsky MP,
Diefenbach KA, Aldrink JHJ (2020) A comparison
of robotic-assisted splenectomy and laparoscopic
splenectomyfor children with hematologic disorders.
Pediatr Surg 56(5):1047–1050
Stylianos S (2019) To save a child’s spleen: 50 years from
Toronto to ATOMAC. J Pediatr Surg 54(1):9–15
 Hepatic Tumors in Childhood
Thambipillai Sri Paran and Michael P. La Quaglia
Differential Diagnosis: Traumatic liver hema-
toma, Infective liver abscesses
86.1 Introduction
Primary neoplasms of the liver are rare in child-
hood and constitute 0.3–2% of all pediatric
tumors. Malignant neoplasms account for 1% of
all pediatric malignancies and are the third most
common intra-abdominal neoplasm after neuro-
blastoma and nephroblastoma. The two primary
malignant neoplasms of the liver are hepatoblas-
toma and hepatocellular carcinoma. For both
these malignant tumors, complete resection of
the tumor is necessary to achieve a cure. Detailed
description of hepatic segmental anatomy has
enabled surgeons to resect larger tumors in the
recent years. This appreciation of the anatomy
and the regenerative capability of the liver, which
allows up to 85% of the liver to be safely removed
in small infants, have greatly increased the scope
for cure.
T. S. Paran (*)
Children’s Health Ireland at Crumlin, Dublin, Ireland
and Trinity College Dublin and University College
Dublin, Dublin, Ireland
M. P. La Quaglia
Pediatric Surgery, Memorial Sloan-Kettering Cancer
Center, New York, NY, USA
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_86
86
The incidence of benign liver tumors in chil-
dren is less than their malignant counterpart, and
in a large series of pediatric liver tumors, benign
tumors accounted for less than 35%. The benign
tumors include hemangiomas or vascular malfor-
mations, hepatocellular adenomas, focal nodular
hyperplasia, mesenchymal hamartomas, and var-
ious types of cysts and cystic disease. With wide-
spread use of ultrasonography, fortuitous
discoveries of benign tumors are being made in a
large number of children, and this raises the ques-
tion of optimal surgical management for the
surgeon.
86.2 History
A first glimpse of the hepatic anatomy was pre-
sented by Herophilus and Erasistratus between
310 BC and 280 BC. The first attempt at hepatic
resection did not take place till the late 1880s.
Though Martin et al. reported that hepatoblasto-
mas could be treated by hepatic lobectomy in
1969, real progress in hepatic surgery was mostly
seen following the detailed description of hepatic
segmental anatomy by Couinaud. This knowl-
edge allowed surgeons to get good vascular con-
trol before attempting to divide the liver
parenchyma, thereby avoiding catastrophic
bleeding.
Holton et al. showed that hepatoblastoma was
sensitive to systemic chemotherapy in 1975
1185
1186
(Holton et  al. 1975). Following this, various
authors have shown that neoadjuvant chemother-
apy is useful in reducing the tumor size in hepa-
toblastoma, thereby enabling complete surgical
resection of previously unresectable tumors.
Presently, the standard of practice is to adminis-
ter neoadjuvant systemic chemotherapy to
patients with hepatoblastoma unless the tumor is
clearly resectable at diagnosis.
The first application of hepatic transplantation
to a childhood liver tumor was reported by
Heimann et  al. in 1987 (Heimann et  al. 1987).
Tagge et al. reported the first series of pediatric
liver tumor patients treated by hepatic transplan-
tation in 1992 (Tagge et al. 1992). Total hepatec-
tomy with liver transplantation now has become a
recognized treatment for unresectable hepato-
blastoma confined to the liver, and this treatment
modality is part of the SIOPEL 3 protocol.
86.3 Surgical Anatomy
The liver has two main lobes, a large right and a
smaller left, and conventional description places
their line of fusion on the upper surface of the
Fig. 86.1  Couinaud’s description of the liver into eight
functional divisions. Each segment receives a branch of
the portal vein and hepatic artery and is drained by a
T. S. Paran and M. P. La Quaglia
liver along the attachment of the falciform liga-
ment. However, knowledge of the detailed inter-
nal functional anatomy of the liver is essential for
planning surgical resections. The anatomical
description of the liver by Couinaud is the most
complete and exact and also the most useful for
the operating surgeon (Fig. 86.1). Essentially the
three main hepatic veins within the scissurae
divide the liver into four sectors, each of which
receives a portal pedicle with alternation between
the hepatic veins and portal pedicles. The main
portal scissura contains the middle hepatic vein
and progresses from the middle of the gallbladder
bed anteriorly to the left of the vena cava posteri-
orly. The right and left liver, demarcated by the
main portal scissura, are independent in terms of
portal and arterial vascularization and of biliary
drainage. The distribution of the portal pedicles
and hepatic veins delimits the hepatic segments,
each of which has a unique portal vein, a branch
of hepatic artery, and bile duct. Knowledge of
this anatomy allows control of the vascular struc-
tures before division of the hepatic parenchyma,
thereby making major hepatic resections
feasible.
branch of either the right or left hepatic duct. The three
main hepatic veins demarcate the liver into its four portal
sectors
86  Hepatic Tumors in Childhood
86.4 Evaluation of a Child
with Hepatic Mass
Most hepatic tumors present as an asymptomatic
abdominal mass. Patients presenting with a sus-
pected hepatic mass are first evaluated by history
and physical examination. Blood should be
drawn for the following tests: complete blood
count, liver function tests, coagulation studies,
and measurement of tumor markers such as
serum α-fetoprotein and β-human chorionic
gonadotropin.
Usually the initial radiological evaluation is
with ultrasonography. Ultrasonography will
determine whether the hepatic mass is solid or
cystic and the extent of the mass. A Doppler
ultrasonography is useful in determining the
patency of hepatic vasculature. This is followed
with either a CT scan or magnetic resonance
imaging (MRI). In our experience, MRI provides
the greatest amount of information concerning
both the lesion and surrounding veins and bile
ducts (Pugmire and Towbin 2016). If a malignant
liver tumor is suspected following the initial
scans, then a thoracic CT scan is necessary for
the purpose of tumor staging.
A tissue diagnosis is necessary to confirm
malignancy. Percutaneous needle core biopsy is
useful for the diagnosis of hepatoblastomas but
may not be definitive in the case of hepatocellular
carcinoma. When larger samples are needed, an
open or laparoscopic liver biopsy is necessary.
Once a diagnosis of malignancy is made, a multi-
disciplinary team approach including the pediat-
ric surgeon, oncologist, and an experienced
hepatobiliary surgeon is necessary in the plan-
ning of definitive surgery.
86.5 Malignant Liver Tumors
The most common malignant hepatic tumors
are hepatoblastoma, hepatocellular carcinoma
(hepatoma), and sarcomas. It is estimated that
about 57.8% of these are hepatoblastoma,
33.4% are hepatocellular carcinoma, and 8.8%
are sarcomas.
1187
86.5.1 Hepatoblastoma
86.5.1.1 Incidence and Etiology
Hepatoblastomas are the most common primary
hepatic tumors in children, accounting for up to
64% of hepatic malignant tumors. Hepatoblas­
toma affects one child in one million per year
under 15 years of age. This translates into approx-
imately 50–70 new cases per year in the United
States with a male-to-female ratio of 1.7:1.0.
Over 75% of these tumors occur in children less
than 2 years of age. The median age at diagnosis
of hepatoblastoma is 18  months (La Quaglia
2000). Though congenital hepatoblastomas and
adult-onset hepatoblastomas have been described
in literature, these are rare. Blair et  al. noted a
borderline, but significant, increase in the inci-
dence of hepatoblastoma in 1994 (Blair and
Birch 1994). However, Stiller et al. found no such
increase in the incidence of hepatoblastoma
between 1978 and 1997 (Stiller et al. 2006).
Hepatoblastoma may occur in siblings, and
there is an increased incidence in first-degree
relatives of the patients with familial polyposis
coli. Other conditions associated with hepato-
blastoma include Gardner’s syndrome and the
Beckwith-Wiedemann syndrome. The Beckwith-­
Wiedemann syndrome is also associated with
several abdominal tumors, of which the majority
are Wilms’ tumors.
There is a significant association between
hepatoblastoma and low birth weight. It is
unknown as to whether developmental abnormal-
ities associated with prematurity or interventions,
such as early total parenteral nutrition, are the
cause of the increased incidence of hepatoblas-
toma in these children. The increased survival of
these children may somewhat explain the
increased overall incidence of hepatoblastoma
noted by some authors.
86.5.1.2 Histopathological Subtypes
There are five histological subtypes of hepato-
blastoma reported based on the light microscopic
findings. These subtypes are fetal (cells grow in
trabeculae and resemble fetal hepatic cells),
embryonal (cells grow in noncohesive sheets),
1188
4 atic. A mild anemia associated with a markedly
3 of patients at diagnosis. The platelet count can
Relative Risk
2
1 Measurement of the serum α-fetoprotein is
0 diagnosis of hepatoblastoma and a means of
al lar
l
a
ym cu
Fe
ch be
en ra
es ror
d
m ac
ixe
M ap
M An
Fig. 86.2  Relative risk of death associated with histo-
pathological subtypes of hepatoblastoma (La Quaglia
2000). The small cell undifferentiated or anaplastic vari-
ant has a very poor risk in comparison to the other three
histopathological subtypes
mixed mesenchymal (contain mesenchymal tis-
sue along with the epithelial component), macro-
trabecular, and anaplastic or small cell
(Ranganathan et al. 2020). The histological sub-
typing of hepatoblastoma is important as there is
an association between prognostic risk and the
various subtypes as illustrated in Fig. 86.2. Some
authors have indicated that the fetal histological
subtype has a better prognosis compared to oth-
ers. However, a review of 105 cases at the Armed
Forces Institute of Pathology failed to confirm
this finding. The histological subtype anaplastic
or small cell variant is a rare subtype, but is par-
ticularly virulent with a strong metastatic
potential.
86.5.1.3 Clinical Findings
As mentioned above, the most common present-
ing sign of a hepatoblastoma is an asymptomatic
abdominal mass. The child is often in good
health, and the tumor mass is discovered inciden-
tally when an attentive parent, grandparent, or
clinician discovers the mass on a routine exami-
nation or while bathing the child. A small minor-
ity may have other symptoms such as pain,
irritability, minor gastrointestinal disturbances,
fever, pallor, failure to thrive, and even tumor
rupture. Patients with the anaplastic variant of
hepatoblastoma who often have distant metasta-
ses at diagnoses are more frequently symptom-
T. S. Paran and M. P. La Quaglia
elevated platelet count is observed in the majority
range into the millions, and the etiology is prob-
ably secondary to an abnormal cytokine release.
well established as an initial tumor marker in the
al ll
ce monitoring the therapeutic response. When inter-
n
yo all
br
Em cs
m preting the α-fetoprotein level, it is important to
ti
las realize that normal α-fetoprotein levels are very
high at birth and decrease over the first 6 months
of life. Usually by 6  months of age, the levels
should be below 100 ng/mL, though in some chil-
dren this may take up to 1 year of age. It is esti-
mated that the α-fetoprotein will be markedly
elevated in 84–91% of patients with hepatoblas-
toma. Low initial α-fetoprotein levels have been
associated with poor survival outcome.
86.5.1.4 Imaging
As mentioned earlier, the first imaging study is
usually an abdominal ultrasound, and, if duplex
technique is employed, tumor vascularity can be
gauged and the hepatic veins assessed. This is
usually followed by computerized axial tomogra-
phy or magnetic resonance imaging. A computer-
ized axial tomography with oral contrast is useful
in identifying any pulmonary metastases and in
determining the hepatic involvement and resect-
ability (Fig. 86.3).
86.5.1.5 Staging
Most studies to date have used the clinical group-
ing defined by the Children’s Cancer Group and
the Pediatric Oncology Group as listed in
Table 86.1.
This classification is based on the postopera-
tive extent of disease. Although this classification
is useful in predicting postoperative prognosis, it
does not provide preoperative information. To
assess tumor response and resectability before
and after neoadjuvant chemotherapy, the
International Society of Paediatric Oncology
(SIOP) developed the PRETEXT (pretreatment
extent of disease) staging system (Table  86.2).
The PRETEXT system is based on the radiologi-
cal findings and describes both the number and
86  Hepatic Tumors in Childhood 1189

Table 86.2  International Society of Paediatric Oncology

PRETEXT staging
The liver is divided into four sectors, based on its
surgical anatomy.
• PRETEXT I—three adjoining sectors are free of
tumor.
• PRETEXT II—two adjoining sectors are free of
tumor.
• PRETEXT Ill—one non-sector or two non-adjoining
sectors are free of tumor.
•  PRETEXT IV—no tumor-free sectors.
In addition, extrahepatic tumor is expressed as follows:
V—tumor involving hepatic vein
P—tumor involving portal vein
E—extrahepatic direct spread, limited to enlargement
of the hilar lymph nodes
M—presence of distant metastases

Table 86.3  TNM staging of hepatic tumors

Stage I T1, N0, M0
Stage II T2, N0, M0
Stage III T1, N1, M0
T2 N1, M0
T3 N0, M0
Stage IVA T4, any N, M0
Stage IVB Any T, any N, M1
T0 = no tumor
Fig. 86.3  Typical response of hepatoblastoma to neoad- T1 = solitary tumor </= 2 cm
juvant chemotherapy. CT findings at diagnosis in a child T2  =  solitary tumor </= 2  cm with vascular invasion or
with hepatoblastoma. The tumor involves both right and multiple tumors limited to one lobe without vascular
left hepatic lobes and is inoperable. CT scan in the same invasion
child following four cycles of chemotherapy. The tumor is T3  =  solitary tumor >/= 2  cm with vascular invasion or
confined to left lobe and is eminently resectable multiple tumors limited to one lobe with vascular
invasion
T4 = multiple tumors in more than one lobe or involve-
Table 86.1  Post-surgical clinical group staging ment of a major branch of the portal or hepatic vein
Stage I No metastases, tumor completely resected N0 = no nodal disease
Stage II No metastases, tumor grossly resected with N1 = nodes involved
microscopic residual disease (positive M0 = no distant metastases
margins or tumor rupture/spill at the time of M1 = distant metastases
surgery)
Stage III No distant metastases, tumor unresectable or
resected with gross residual tumor or 86.5.1.6 Treatment and Prognosis
positive lymph nodes Following initial assessment, the first decision
Stage IV Distant metastases regardless of the extent regarding treatment is whether to initiate neoad-
of liver involvement juvant chemotherapy or proceed with resection.
About 46% of hepatic malignancies are resect-
the location of involved liver sectors and takes able at diagnosis. Resection at diagnosis will
into account the invasion of the hepatic and portal avoid or lessen the need for chemotherapy and its
veins as well as extrahepatic and metastatic dis- associated morbidity. However, when resection at
ease. More recently, the TNM classification has presentation is not feasible, neoadjuvant chemo-
been used (Table 86.3). therapy can shrink the tumor size extensively and
1190
enable safer and complete resection at a later
stage. This requires good clinical judgment and
good communication between the pediatric sur-
geon, oncologist, radiologist, and, when neces-
sary, an experienced hepatobiliary surgeon.
Using the clinically established PRETEXT
groups I, II, III, and IV as stems, Meyers et  al.
have created risk stratification trees based on
5-year event-free survival and clinical applicabil-
ity of various trials to predict the prognosis
(Meyers et al. 2017).
For unresectable tumors at diagnosis, the ini-
tial surgical procedure should include a diagnos-
tic biopsy and placement of a vascular access
device for neoadjuvant chemotherapy. At present,
the recommendation for initial treatment of hepa-
toblastomas is with cisplatin, 5-fluorouracil, and
vincristine. Single-agent doxorubicin is some-
times used in very young infants who undergo
complete resection. This regimen consists of just
three doses and may be associated with less long-
and short-term toxicity than multi-agent regi-
mens. Definitive resection of tumor is undertaken
after four cycles of chemotherapy, if complete
resection is feasible. With complete resection of
the primary tumor, overall survival of 60–70% is
achievable with non-stage IV hepatoblastoma
except in the very small group of children with
aggressive anaplastic variant.
It is estimated that approximately 20% of chil-
dren will have stage IV disease at the time of
diagnosis. The overall survival rates are lower in
this group of children, but in our experience, a
rate of 50% is achievable as long as complete
resection at the primary site is accomplished.
Some patients with microscopic residual at the
primary site are curable with continued chemo-
therapy and may benefit from external beam
radiotherapy to this primary hepatic site. Most
pulmonary metastases will resolve fully with
chemotherapy, while resection is reserved for
larger or persistent metastatic lesions. Review by
Lake CM et  al. has looked at the indications,
roles, and outcomes of these surgical approaches
as well as those for the resection of pulmonary
metastases (Lake et al. 2019).
T. S. Paran and M. P. La Quaglia
Hepatic transplantation for unresectable pri-
mary lesions can be effective for tumors confined
to the liver (Khan et al. 2017). Cases with exten-
sive extrahepatic extension or vascular invasion
have had poor outcomes with total hepatectomy
and hepatic transplantation. Chemoembolization
shows promise and involves arteriographic injec-
tion of occluding thrombogenic materials
(Angiostat collagen) or stainless steel coils com-
bined with chemotherapeutic agents, like cispla-
tin or doxorubicin, into the arterial circulation to
the tumor. Using this technique, the concentra-
tion of chemotherapeutic agents can be increased
50- to 100-fold in the embolized tumor.
Interventional radiologist can also use selective
portal venous branch embolization aimed at caus-
ing hypertrophy of the remnant liver segments,
enabling a more radical surgery (Matthew
Hawkins et al. 2018). Others have treated pulmo-
nary metastases with external beam radiotherapy
in an approach similar to that used for Wilms’
tumors but with 18–20  Gy ­ administered.
Pulmonary radiation may be associated with sig-
nificant pulmonary toxicity.
86.5.2 Hepatocellular Carcinoma (or
Hepatoma)
86.5.2.1 Incidence and Epidemiology
Hepatocellular carcinoma accounts for 23% of
pediatric liver tumors and affects about 0.5 chil-
dren in one million per year under the age of 15.
The incidence is bimodal with an early peak
before the age of 5  years and a second peak
between 13 and 15  years of age. It is rare in
infancy, though historical series without patho-
logical review may report a higher rate of infan-
tile hepatocellular carcinoma due to misdiagnosis
of some early hepatoblastomas. The Liver Cancer
Study Group of Japan reported no cases below
the age of 4  years in a series of 2286 patients
(Japan, 1987). There is a male predominance that
ranges from 1.3 to 3.2:1 for hepatocellular carci-
noma. In areas endemic for hepatitis B, the male-­
to-­female ratio may be reversed at 0.2:1.
86  Hepatic Tumors in Childhood
The relative risk for the development of hepato-
cellular carcinoma is 250 for patients with chronic
active hepatitis compared to those without hepati-
tis surface antigen positivity. This knowledge has
enabled the health workers in Taiwan to reduce
their incidence of hepatocellular carcinoma in
children from 0.70 per 100,000 children to 0.36
per 100,000 in 5 years following the introduction
of a universal vaccination program against hepati-
tis B. Other conditions associated with the devel-
opment of hepatocellular carcinoma include
cirrhosis, alpha-1-antitrypsin deficiency, tyrosin-
emia, aflatoxin ingestion, hemochromatosis,
hepatic venous obstruction, androgen and estrogen
exposure, the Alagille syndrome (arteriohepatic
dysplasia), and Thorotrast administration.
86.5.2.2 Clinical Findings
Unlike children with hepatoblastoma, children
with hepatocellular carcinoma are usually symp-
tomatic at diagnosis. Pain is common (38%) and
may even occur in the absence of an obvious mass.
But most of these symptoms are nonspecific and
include anorexia, fatigue, nausea, and vomiting
and weight loss. The α-fetoprotein is elevated in
approximately 85% of patients with most levels
more than 1000  ng/mL.  Though elevated, these
levels are usually lower than those measured in
hepatoblastoma patients. Up to 10% may present
with tumor rupture with signs and symptoms of a
hemoperitoneum. The tumors that rupture are not
necessarily large, and long-term survival with
complete resection has been reported.
86.5.2.3 Staging
The staging schemes listed for hepatoblastoma
are also used for hepatocellular carcinoma in
childhood.
86.5.2.4 Treatment and Outcome
Again, complete resection of the primary tumor
is essential for long-term survival. However, this
is usually difficult at presentation due to the mul-
tifocal nature of the tumor with its extrahepatic
involvement of regional lymph nodes and distant
metastases.
1191
Infiltration with thrombosis of portal and
hepatic venous branches is common, and even
the vena cava may be involved. Chen et  al.
reported an 18.2% complete resection rate in
their series of 55 children (Chen et  al. 1988).
Combination chemotherapy with doxorubicin,
cyclophosphamide, vincristine, and 5-­fluorouracil
has not been found effective in reducing tumor
size in hepatocellular carcinoma. All this trans-
lates into an overall survival rate of zero for chil-
dren with hepatocellular carcinoma that is not
fully resectable at presentation.
Liver transplantation has shown to be useful in
selected patients with unresectable tumor.
However, positive hepatitis B viral serology, non-­
fibrolamellar histological type, and high fre-
quency of local or metastatic spread make most
patients unsuitable for liver transplantation.
86.5.3 Rhabdomyosarcoma
of Extrahepatic Bile Ducts
This form of rhabdomyosarcoma is a very rare
tumor with 40% of patients presenting with dis-
tant metastases (Chavhan et al. 2019). But mor-
tality is most often due to the effects of local
invasion. Rhabdomyosarcoma of the liver, not
involving the bile ducts, has also been reported,
but is extremely rare. The patients’ ages range
from 1 to 9  years at presentation. The typical
symptoms include intermittent jaundice and
sometimes loss of appetite and episodes of chol-
angitis (Charcot’s triad). Hepatomegaly or a pal-
pable abdominal mass is usual, and the diagnosis
may be confused with hepatitis or a choledochal
cyst. As with hepatoblastoma and hepatocellular
carcinoma, tissue diagnosis is necessary.
Neoadjuvant chemotherapy will reduce the tumor
size and allow a cleaner resection at second-look
surgery. During the initial biopsy, hilar and left
gastric lymph node sampling is performed to
determine whether these nodal echelons require
added radiotherapy. Resection appears to improve
survival. Whether these patients may be treated
by chemotherapy alone after establishment of the
1192 T. S. Paran and M. P. La Quaglia

diagnosis and simply observed if a complete

radiological response is documented remains to
be confirmed with future studies.

86.5.4 Primary Hepatic Non-­

Hodgkin’s Lymphoma

Primary non-Hodgkin’s lymphoma of the liver

occurs in childhood and may account for up to
5% of primary hepatic malignancies. These
tumors respond well to chemotherapy, and sur-
gery is not necessary.

86.5.5 Metastatic Hepatic Tumors

Several tumors, including non-Hodgkin’s lym-

phoma, neuroblastoma, rhabdomyosarcoma,
rhabdoid tumors, Wilms’ tumor, the desmoplastic
small round cell tumor, adrenal cortical carci-
noma, and osteogenic sarcomas, are known to
metastasize to the liver in children. There is little
data to determine the correct surgical approach to
these lesions. Criteria for surgical removal of
hepatic metastases include control of the primary
site, a solitary or limited number of metastases,
good performance status, and a reasonable expec-
tation of prolonged survival or cure (Figs.  86.4
and 86.5).
Hepatic metastases from neuroblastoma are
seen in newborns and infants with stage 4S disease
and are a distinct characteristic of this disease.
These lesions usually resolve with time. Wilms’
tumor metastasizes to the liver in about 12% of
cases, and this is usually associated with unfavor-
able histology. In selected patients, resection of
these lesions may have a survival benefit. Most of
the other tumors metastasize to the liver in the late
stages of disease, and surgical resection does not
appear to provide any survival advantage.
86.5.6 Benign Hepatic Tumors
Benign tumors of the liver that occur in child-
hood include hemangiomas or vascular tumors,
hepatocytic adenomas, focal nodular hyperplasia,
Fig. 86.4  An 18-month-old girl with Beckwith-­Wiedemann
syndrome presented with a left Wilms’ tumor and
hepatic metastasis and underwent left nephrectomy and
extended right hepatectomy following neoadjuvant therapy
successfully
mesenchymal hamartomas, and various types of
cysts or cystic disease. Vascular tumors are the
most common and compose greater than 50% of
these benign hepatic tumors.
86.5.6.1 Vascular Tumors
Hemangioma: Hemangiomas are lesions charac-
terized by endothelial-lined vascular spaces that
can vary in size and extent. They are sometimes
classified as hamartomas. The overall incidence
of endothelial-lined vascular tumors of the liver
in childhood is probably unknown since many
are asymptomatic. Vascular lesions taken together
represent 13–18% of symptomatic hepatic tumors
in childhood. Hepatic hemangiomas are twice as
common in females as in males.
86  Hepatic Tumors in Childhood
Fig. 86.5  Right adrenocortical carcinoma with direct
liver invasion; Right hepatectomy and adrenocortical car-
cinoma resection en masse. Child remains well 2  years
post-resection
Diagnosis can be fortuitous on a routine ultra-
sonography or when presented with an abdomi-
nal mass. Multiple hemispherical cutaneous
hemangiomas may be present and warn the phy-
sician of possible visceral lesions. MRI study is
all that is needed to confirm diagnosis, and
asymptomatic lesions are best left alone, as they
tend to regress after the first year of life (Iacobas
et al. 2018). Large lesions may cause congestive
heart failure; if medical treatment is not success-
ful, either hepatic artery embolization (Lungren
et al. 2018) or direct surgical ligation may be nec-
essary (Zavras et al. 2020).
Hemangioendothelioma: These are highly
proliferative cellular lesions of variable malig-
nant potential. In 1 report of 16 infants and chil-
dren, 15 presented with hepatomegaly and 7 with
congestive heart failure, and 4 had associated
cutaneous lesions. The Kasabach-Merritt syn-
drome, a platelet-trapping coagulopathy, has also
been observed. These lesions may appear very
cellular but do not metastasize. If a primary
lesion produces symptoms, resection is indicated
for relief.
1193
Hemangioblastoma: Hemangioblastoma of
the liver is usually associated with Lindau-von
Hippel disease. In infancy and childhood, these
lesions appear very cellular, but distant metasta-
ses are uncommon. Complete resection should be
performed and is usually curative.
86.5.6.2 Mesenchymal Hamartoma
Mesenchymal hamartomas account for six per-
cent of primary liver tumors in childhood, and
there is a male predominance. Two-thirds of
these tumors are diagnosed at less than 1 year of
age. These are usually multicystic, and the cysts
are lined with flattened biliary epithelium or
endothelium. It is postulated that mesenchymal
hamartomas arise in areas of focal intrahepatic
biliary atresia. This results in distal bile duct
obstruction and hepatocellular necrosis.
The majority of mesenchymal hamartomas
present as an enlarging abdominal mass or hepa-
tomegaly and are usually not symptomatic. They
can grow to great sizes causing respiratory
­distress or evidence of caval obstruction. Often
an open biopsy is necessary to make the diagno-
sis. Anatomical resection is the recommended
treatment for large lesions. Because of the mes-
enchymal component, these lesions have a defi-
nite capsule that facilitates enucleation of large
central mesenchymal hamartomas that are not
amenable to lobectomy. Prognosis is good, and in
1 study of 18 patients, 13 who were available for
follow-up were alive and well 1 month to 24 years
after treatment (mean = 5 years).
86.5.6.3 Focal Nodular Hyperplasia
These are benign hepatocellular proliferations
that are rare in children. Less than 2% of hepatic
tumors in childhood are focal nodular hyperpla-
sia or hepatocellular adenomas. The presence of
fibrous septae containing bile ducts and inflam-
matory infiltrate distinguishes focal nodular
hyperplasia from hepatocellular adenoma. These
fibrous septae are seen as a distinct central scar in
the ultrasound and computerized axial tomogram
images. Most patients are less than 5 years of age
at presentation, and there is a female predomi-
1194
nance. There is an association with contraceptive
use in adults, but no such association is reported
in childhood and adolescence.
It is best to remove adenomas because of the
difficulty in histologically differentiating them
from low-grade hepatocellular carcinomas and
because of the uncertainty surrounding future
malignant degeneration. Resection may also
result in symptomatic relief in some children. In
general, anatomic resection is preferred for focal
nodular hyperplasia, and most do well after the
hepatic resection. However, asymptomatic
lesions can be observed using serial abdominal
ultrasonography.
86.5.6.4 Cysts and Cystic Disease
There are multiple case reports of solitary, con-
genital, non-parasitic liver cysts in childhood.
They are extremely rare but have been increas-
ingly noted as incidental findings on ultrasounds
or computerized axial tomograms performed for
other reasons. These usually are simple cysts and
are asymptomatic. To the best of our knowledge,
there has been no report indicating any malignant
degeneration of these cysts.
If there are associated symptoms, such as pain
or jaundice, then aspiration followed by ethanol
injection (sclerotherapy) may be of benefit. If this
is not successful, surgical intervention including
resection, marsupialization, or cyst wall excision
may be considered.
86.6 Conclusions
Malignant hepatic tumors in children are rare.
Hepatoblastomas and hepatocellular carcinomas
together account for almost 80% of primary
malignant liver tumors in children and adoles-
cents. Imaging is crucial in the assessment of chil-
dren with a primary hepatic malignancy. Specific
guidelines for the management of different liver
tumors have been implemented in recent years.
These guidelines have made significant contribu-
tions to improving the treatment and outcomes for
patients with malignant hepatic tumors. A multi-
disciplinary approach is ideal for the management
of these tumors, and care should be provided by
centers with experience in their management.
T. S. Paran and M. P. La Quaglia
References
Blair V, Birch JM (1994) Patterns and temporal trends
in the incidence of malignant disease in chil-
dren: II.  Solid tumours of childhood. Eur J Cancer
30A(10):1498–1511
Chavhan GB, Siddiqui I, Ingley KM, Gupta AA (2019)
Rare malignant liver tumors in children. Pediatr
Radiol 49(11):1404–1421. https://doi.org/10.1007/
s00247-­019-­04402-­8
Chen WJ, Lee JC, Hung WT (1988) Primary malignant
tumor of liver in infants and children in Taiwan. J
Pediatr Surg 23(5):457–461
Heimann A, White PF, Riely CA et  al (1987)
Hepatoblastoma presenting as isosexual precocity.
The clinical importance of histologic and serologic
parameters. J Clin Gastroenterol 9:105–110
Holton CP, Burrington JD, Hatch EI (1975) A mul-
tiple chemotherapeutic approach to the management
of hepatoblastoma. A preliminary report. Cancer
35(4):1083–1087
Iacobas I, Phung TL, Adams DM, Trenor CC 3rd, Blei
F, Fishman DS, Hammill A, Masand PM, Fishman SJ
(2018) Guidance document for hepatic Hemangioma
(infantile and congenital) evaluation and monitoring.
J Pediatr 203:294–300.e2. https://doi.org/10.1016/j.
jpeds.2018.08.012
Khan AS, Brecklin B, Vachharajani N, Subramanian V,
Nadler M, Stoll J, Turmelle Y, Lowell JA, Chapman
WC, Doyle MM (2017) Liver transplantation for
malignant primary pediatric hepatic tumors. J Am
Coll Surg 225(1):103–113. https://doi.org/10.1016/j.
jamcollsurg.2017.02.006
La Quaglia MP (2000) Hepatic tumors in childhood. In:
Blumgart L, Fong Y (eds) Surgery of the liver and bili-
ary tract, vol 2. W.B. Saunders, London, pp 1451–1473
Lake CM, Tiao GM, Bondoc AJ (2019) Surgical manage-
ment of locally-advanced and metastatic hepatoblas-
toma. Semin Pediatr Surg 28(6):150856. https://doi.
org/10.1016/j.sempedsurg.2019.150856
Lungren MP, Towbin AJ, Roebuck DJ, Monroe EJ, Gill
AE, Thakor A, Towbin RB, Cahill AM, Matthew HC
(2018) Role of interventional radiology in managing
pediatric liver tumors: part 1: endovascular inter-
ventions. Pediatr Radiol 48(4):555–564. https://doi.
org/10.1007/s00247-­018-­4068-­1
Matthew Hawkins C, Towbin AJ, Roebuck DJ, Monroe
EJ, Gill AE, Thakor AS, Towbin RB, Cahill AM,
Lungren MP (2018) Role of interventional radiology
in managing pediatric liver tumors: part 2: percuta-
neous interventions. Pediatr Radiol 48(4):565–580.
https://doi.org/10.1007/s00247-­018-­4072-­5
Meyers RL, Maibach R, Hiyama E, Häberle B, Krailo
M, Rangaswami A, Aronson DC, Malogolowkin MH,
Perilongo G, von Schweinitz D, Ansari M, Lopez-­
Terrada D, Tanaka Y, Alaggio R, Leuschner I, Hishiki
T, Schmid I, Watanabe K, Yoshimura K, Feng Y,
Rinaldi E, Saraceno D, Derosa M, Czauderna P (2017)
Risk-stratified staging in paediatric hepatoblastoma: a
unified analysis from the children’s hepatic tumors
86  Hepatic Tumors in Childhood
international collaboration. Lancet Oncol 18(1):122–
131. https://doi.org/10.1016/S1470-­2045(16)30598-­8
Primary liver cancer in Japan; Sixth report by The
Liver Cancer Study Group of Japan. (1987) Cancer
60:1400–1411
Pugmire BS, Towbin AJ (2016) Magnetic resonance
imaging of primary pediatric liver tumors. Pediatr
Radiol 46(6):764–777. https://doi.org/10.1007/
s00247-­016-­3612-­0
Ranganathan S, Lopez-Terrada D, Alaggio R (2020)
Hepatoblastoma and pediatric hepatocellular carci-
noma: an update. Pediatr Dev Pathol 23(2):79–95.
https://doi.org/10.1177/1093526619875228
Stiller CA, Pritchard J, Steliarova-Foucher E (2006) Liver
cancer in European children: incidence and survival,
1195
1978-1997. Report from the automated childhood
cancer information system project. Eur J Cancer
42(13):2115–2123
Tagge EP, Tagge DU, Reyes J et  al (1992) Resection,
including transplantation, for hepatoblastoma and
hepatocellular carcinoma: impact on survival. J
Pediatr Surg 27:292–296. discussion 297
Zavras N, Dimopoulou A, Machairas N, Paspala A,
Vaos G (2020) Infantile hepatic hemangioma: cur-
rent state of the art, controversies, and perspectives.
Eur J Pediatr 179(1):1–8. https://doi.org/10.1007/
s00431-­019-­03504-­7
 Pediatric Liver Transplantation
Khalid Sharif and Alastair J. W. Millar
87.1 Introduction
Liver disease has generally been underestimated
as a cause of death in children. Biliary atresia,
metabolic diseases, acute liver failure, cholestatic
diseases, tumours and retransplantation of the
liver make up the majority of indications for liver
transplantation in children. Advances in the man-
agement of biliary atresia, the initial prompt for
the development of infant liver transplantation,
have achieved some success with timely diagno-
sis, early surgical intervention and close post-­
operative monitoring. Acute liver failure is
becoming less frequent with successful hepatitis
vaccination programs. However, in many cases,
even with expert management, there is inevitable
progression of liver disease. Liver transplantation
offers the only chance of a cure for these unfortu-
nate children.
K. Sharif
Birmingham Women’s and Children’s Hospital,
Birmingham, UK
e-mail: khalid.sharif1@nhs.net
A. J. W. Millar (*)
University of Cape Town and Red Cross War
Memorial Children’s Hospital,
Cape Town, South Africa
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_87
87
87.2 Historical Overview
The first attempted human liver transplant was
performed by Starzl in 1963 (Brenner and
Thomas 2017), but it was not until 5 years later
that long-term success was achieved (Otte 2002;
Ferreira-Meirelles et al. 2015). Since then, there
has been an ever-increasing list of indications for
liver transplantation in infants and children.
Advances in donor identification, donor organ
management, surgical technique, anaesthetic
management, pre- and post-operative care and
refinements in immunosuppression over the last
five decades have resulted in much improved out-
comes. The full story of liver transplantation for
children is yet to be told, but the current expected
10-year survival is now greater than 90% in major
pediatric transplant centres (Ramos-Gonzales
et al. 2019). Excellent quality of life is the rule
rather than the exception. The longest survivor is
more than 50 years after transplantation. Current
anxieties remain over organ donor scarcity, the
long-term side effects of the immunosuppressive
therapy and some ethical issues with regard to
donor organ allocation especially. The focus of
attention has now shifted from an initial target of
early post-transplant survival to quality of life in
the long term. The transformation of a miserable,
jaundiced patient into an active, healthy child
remains a powerful stimulant for pediatricians
and transplant surgeons alike.
1197
1198 K. Sharif and A. J. W. Millar

87.3 Indications Table 87.1  Indications for which liver transplantation

has been performed in children

Liver transplantation should be considered as a       I.  Metabolic (inborn errors of metabolism).

    Alpha-1 antitrypsin
therapeutic option in all cases of acute and     Tyrosinaemia
chronic liver disease before end-stage liver dis-     Glycogen storage disease type III and IV
ease is reached. The most frequent indications     Wilson’s disease
for pediatric liver transplantation are biliary atre-     Neonatal haemochromatosis
    Hypercholesterolaemia
sia (43%), metabolic diseases (13%) and acute     Cystic fibrosis
liver failure (11%). Liver tumours are a group     Hyperoxaluria (+ renal transplant)
where controversy exists, particularly as to how     Haemophilia A + B
far extensive resections can eliminate the need     Protein C deficiency
    Crigler-Najjar syndrome
for transplant versus poor results for recurrent     II.  Acute and chronic hepatitis
disease after failed surgery. A list of conditions     Fulminant hepatic failure (viral, toxin or drug
for which liver transplant has been performed is induced)
summarized in Table 87.1. In more than 50% of     Chronic hepatitis (B, C, etc., toxin, autoimmune,
idiopathic)
children with acute liver failure, the cause is III.  Intrahepatic cholestasis:
unknown. Among the worldwide accepted indi-     Neonatal hepatitis
cations for liver transplantation, inherited meta-     Alagille syndrome
bolic disorders constitute a major portion     Biliary hypoplasia
    Familial cholestasis
(Mazariegos et al. 2014). In some pediatric liver   IV.  Obstructive biliary tract disease
transplant centres, this indication runs second     Biliary atresia
after extrahepatic biliary atresia. National allo-     Choledochal cyst with cirrhosis
cation policies show similarities and differences     V.  Neoplasia
    Hepatoblastoma
between countries. The number of pediatric liver     Hepatocellular carcinoma
transplantations was four to nine per million     Sarcoma
inhabitants younger than 18 years (Fischler et al.     Haemangioendothelioma
2019). VI.  Miscellaneous
    Cryptogenic cirrhosis
The aim of liver transplantation in inherited     Congenital hepatic fibrosis
metabolic disorders is twofold: the first is to save     Caroli’s disease
a patient’s life by addressing the immediate con-     Budd-Chiari syndrome
sequences of the pathologic defect and the sec-     Cirrhosis from prolonged parenteral nutrition
ond is to accomplish phenotypic and functional
cure of disease. In some metabolic diseases, the certain disorders have moved from the list of con-
manifestations are widespread and affect other traindications to acceptable indication (Shneider
organs as well. Examples are haemochromatosis, 2002). One such example is hepatic respiratory
tyrosinemia, Wilson’s disease, glycogen storage chain disorders, although it is essential to exclude
diseases and hyperoxaluria. With experience, extra-hepatic disease before transplantation.
87  Pediatric Liver Transplantation
In recent years, the outcome of the operation
has improved so much for those with primary
liver disease that indications for early transplant
would be evidence of impaired synthetic func-
tion, including prolonged prothrombin time,
low serum albumin and rising bilirubin. Clinical
indicators include presence of ascites, bleeding
from esophageal varices not controlled by endo-
scopic banding or sclerotherapy and poor
response to nutritional resuscitation. Those with
acute liver failure who develop encephalopathy,
hypoglycaemia and a prothrombin time of
greater than 50 seconds should be considered
for early transplant, because almost all of these
children die without transplantation. Timing of
liver transplantation not only affects survival
rate but also may influence neurodevelopmental
outcome.
87.4 Contraindications
There are few conditions for refusal for trans-
plantation. Until recently, hypoxemia (hepato-­
pulmonary syndrome) was considered a relative (B) Family assessment: Transplant candidacy
contraindication for liver transplantation.
However, it has been shown that liver transplan-
tation can be successfully achieved in severely
hypoxemic children and that post-operative cor-
rection of the right to left intrapulmonary shunt
is then obtained. Major cardiorespiratory, neuro-
logical or renal diseases, which would be incom-
patible with quality of life and long-term
survival, malignancy outside the liver and uncon-
trolled systemic infection are considered as
contraindications.
1199
87.5 Assessment
Liver transplant assessment includes thorough eval-
uation of the child and family/caregivers. This pro-
cess allows detailed disease assessment and time for
discussion of treatment options, to prepare for the
transplant procedure and afterwards. This also pro-
vides the opportunity to assess the family’s commit-
ment to sustain long-­ term compliance after
transplantation and to put in place supportive strate-
gies to ensure adherence to treatment regimens,
which need to be lifelong. However, detailed assess-
ment is not possible in children presenting with
acute liver failure or those with acute deterioration of
chronic liver disease. The pre-transplant assessment
protocol from our centre is summarized in Table 87.2.
(A) Child assessment: All children require initial
confirmation of the diagnosis, intensive med-
ical investigation and nutritional resuscitation
to treat the complications of liver disease,
portal hypertension and nutritional depriva-
tion. Disease assessment also includes the
identification of contraindications.
inevitably results in enormous emotional
stress for parents, while waiting for the
appropriate donor. Family life may become
disrupted, especially for those who live afar
from the transplant centre. Compliance is
more difficult to predict in children with
acute liver failure, as time from presentation
to the decision to transplant is much shorter.
Living related transplant includes extensive
evaluation of the potential donors including
both physical and psychological assessment.
1200 K. Sharif and A. J. W. Millar

Table 87.2  Pre-transplant assessment

Anthropometry:
 Height and weight
 Head circumference—under 2 years of age, occipitofrontal circumference (OFC)
 Mid-xiphisternal umbilical circumference (girth)
 Mid-arm circumference (MAC) and triceps skin fold (TSF)
Routine bloods:
 Blood group
 Full blood count (FBC)
 Prothrombin time (PT), partial thromboplastin time (PTT), fibrinogen
 Liver enzymes AST, ALT, GGT, Alk. Phos, Cholesterol
 Total protein, albumin
 Na, K, Urea, Creatinine, Calcium, Potassium, Magnesium Phosphate
 Vitamin A, E, D, K
 N.B. additional bloods may be required if metabolic liver disease is suspected
Serology:
 EBV (IgG)
 CMV (IgG)
 Measles
 Varicella
 Hepatitis A (IgG)
 Hepatitis B (Hepatitis B sAg + Hep B core antibody—hepatitis B s antibody if previously vaccinated)
 Hepatitis C antibody
 HIV 1 and 2
Microbiology:
 MRSA
 Vancomycin resistant enterococcus (VRE)—pseudomonas—B cepacia screen if chronic patient (cystic fibrosis)
 Blood culture if indwelling catheter
 Urine (microscopy, culture and sensitivity)
 Protein/creatinine ratio
 Tubular reabsorption of phosphate
Assessment of severity of liver disease:
 Upper gastrointestinal endoscopy (if required)
 Liver biopsy (if indicated, otherwise obtain previous biopsy result)
 Pediatric hepatology scores (PHS) and pediatric end-stage liver disease (PELD)
Radiology:
 Chest X-ray
 Bone age for rickets or metabolic bone disease
 Abdominal ultrasound and Doppler study to evaluate the diameter of the portal vein and direction of flow, size of spleen, ascites,
vascular anatomy, resistivity index if biliary atresia or reverse flow in the portal vein
 CT/MRI angiography if vascular anatomy uncertain
 Rarely formal angiogram especially to assess the portal venous system for transplant
Cardiology:
 ECG
 ECHO
 Cardiology opinion if needed
 Blood pressure
Neurology:
 EEG (selected cases)
 Developmental assessment
Renal function:
 Chromium EDTA/cystatin C
 Calculated GFR (Schwartz index)
 Urine tubular phosphate reabsorption (TRP)
 Urine protein/creatinine ratio
Respiratory function:
 Oxygen saturations (resting and on exercise)
For children with cystic fibrosis:
 Pulmonary function tests
 Cough swab sputum for MC + S
 Lung perfusion study only if cyanosed
Immunisations:
 Arrange pre-transplant immunisation as needed (suspend from list for 2–4 weeks if live vaccine given).
 Check handheld record for routine DPT, polio, HIB, MenC and MMR
 Ensure live vaccines are given if time permits i.e. Varicella zoster, MMR (if over 6 months), plus advice on completing other
vaccinations, such as Prevenar®, Pneumovax II®, Hepatitis A and B, influenza.
87  Pediatric Liver Transplantation 1201

87.6 Surgical Technique of the left coronary ligament. An appropriate

sized cannula is placed in the inferior mesenteric
Donor operation: Age limits for suitable donors or superior mesenteric vein so that the tip lies at
are being extended due to a shortage of organs. the junction with the splenic vein. A large bore
Pediatric donors can be accepted from 3 months cannula is placed in the aorta, with the tip approx-
of age, while livers procured from older children imately opposite the renal arteries, via the com-
and even young adults can be transplanted into mon iliac vessels ligated and the donor is given
small children after ex  vivo/in situ reduction/ heparin (3 gm/kg). A large bore cannula may be
splitting of the liver graft. However, stable placed in the inferior vena cava, which is con-
cadaver donors from patients <50 years of age nected to suction for drainage of blood.
with a short intensive care unit stay (less than 3 Procurement commences with infusion of cold
days), little requirement for inotropic support and preservation solution (5 C) through both the por-
o

normal or near-normal liver function are pre- tal vein and aorta after cross clamping the aorta at
ferred, with an expected < 5% incidence of the level of the diaphragm and incising the supra-­
impaired function after transplant. Liver biopsy hepatic vena cava within the pericardium. The
is useful if steatosis is suspected. Viral screening porta hepatis is divided distal to the gastroduode-
of the donors is essential. This would include nal artery, which is ligated and the portal vein is
Hepatitis A, B & C, CMV, EBV, sexually trans- divided at the junction with the splenic vein. The
mitted diseases like Syphilis/Toxoplasmosis and proximal part of the superior mesenteric artery is
HIV screening. Core HBV antibody and HCV defined and is dissected down to the aorta.
positive donors would only be considered in The liver is now removed with a patch of the
selected viral infected recipients. aorta including the base of the superior mesen-
Surgical techniques used for donor retrieval teric and the celiac axis. The resected liver
and recipient liver removal and engraftment have includes a cuff of diaphragm around the bare area
evolved over the last 50 years. The majority of along with the retro-hepatic cava and part of the
donor livers are removed as part of a multi-organ right adrenal gland, which is cut through. The
procurement procedure, which would include infra-hepatic inferior vena cava is divided above
various combinations of kidneys, liver, heart or the renal veins. Once the organs have been
heart and lungs, small bowel and pancreas. removed, they are placed in a plastic bag and the
University of Wisconsin solution is widely used liver is further perfused with 500 ml of preserva-
as the preservation solution of choice. Several tion solution via the portal vein, hepatic artery
other preservative solutions like histidine-­ and through the bile ducts. The liver is then
tryptophan-­ketoglutarate (HTK) and Celsior, etc. placed in a further two plastic bags and packed in
are now available. ice for transportation. The retrieval technique for
The two procurement techniques used are a donors with cardiac death (DCD) is slightly dif-
careful dissection and excision technique or the ferent, with rapid cannulation of the aorta via
so-called ‘rapid’ technique, described by Starzl iliac vessels. Also, if other organs are harvested
(Miller et al. 1988; de Ville de Goyet et al. 1995). like the pancreas or small intestine the division of
A midline incision is made from the suprasternal the portal vein and superior mesenteric artery
notch to the pubis and the sternum is opened. The need adjustment.
abdominal part of the operation includes a quick A major constraint has been the shortage of
assessment of all the abdominal organs. Control donor organs of appropriate size. The use of
of the aorta is achieved above the coeliac axis by reduced size adult organs partially alleviated this
incision of the right crus of the diaphragm. The problem but resulted in the waste of a right lobe
inferior vena cava and aorta are identified, dis- of the liver. Splitting the donor liver into two
sected and encircled with tapes below the renal functioning units for two recipients is now rou-
vessels. After a careful search for any vascular tine in good donors. These innovative techniques
abnormalities, the liver is mobilized by division have allowed safe transplantation with a donor to
1202
recipient weight ratio of greater than 15:1. In
neonates, further reduction of a left lateral seg-
ment graft or creation of a monosegment, usually
segment 2, has made transplant in these neonates
possible.
The recipient operation: The recipient opera-
tion commences with an upper abdominal trans-
verse or curved subcostal laparotomy incision,
which may be extended in the midline to the
xiphoid process for extra exposure. The porta
hepatis is dissected first and in children with bili-
ary atresia, this requires the portoenterostomy to
be taken down. The portal vein is dissected and
isolated. The rest of the liver is carefully mobi-
lized. This includes the gastro-hepatic ligament,
the falciform and triangular ligaments together
with the right retroperitoneal reflection.
Mobilization of the liver off the inferior vena
cava is frequently preserved to facilitate reduced
size transplantation or piggy-back engraftment.
This is simply done by carefully dissecting the
diseased liver clear of the vena cava and individu-
ally suturing/ligating all small caudate lobe
hepatic veins. The supra-hepatic and infra-­
hepatic vena cava are dissected and encircled
with tape. Hemostasis of the retroperitoneum
must be ensured with a combination of suture
ligation and cautery. The inferior vena cava (IVC)
is prepared for the donor liver by dividing the
bridges between the separate hepatic veins. This
creates a wide orifice for the hepatic vein to cava
anastomosis. The inferior vena cava should be
incised distally for approximately 2–3  cm to
make a triangular orifice for the ‘piggy-back’
graft. Engraftment should begin with the upper
caval anastomosis, which is usually performed
with continuous monofilament sutures of poly-
propylene or polydioxanone. Prior to completion
of the anastomosis, the liver is flushed clear of
potassium-rich preservation solution via the por-
tal vein with either normal saline or a colloid
solution. The lower cava anastomosis is then per-
formed (only if the cava is replaced), taking care
not to cause any stricture or kinking, a growth
factor of about a third of the diameter of the ves-
sel is usually sufficient to prevent this occurring.
K. Sharif and A. J. W. Millar
The recipient portal vein is usually used for the
anastomosis and in reduced size/split liver donor
graft transplants, where the donor liver is of large
diameter, the bifurcation of the recipient portal
vein is opened to create a trumpeted end for anas-
tomosis. If the portal vein is hypoplastic, portal
venous reconstruction is done using donor iliac
vein. The portal vein anastomosis is carefully
performed using continuous posterior and inter-
rupted anterior sutures or the use of a generous
growth factor. In reduced-size grafts, plenty of
portal vein length should be left to avoid having
any tension on the vein, which may result in
stretching thus compromising flow. The donor
hepatic artery is flushed with heparin saline to
remove air and blood clots and an anastomosis is
done to the recipient common hepatic artery.
End-to-end microvascular techniques are pre-
ferred, while others have used donor iliac artery
vascular grafts to the infra-renal aorta or from the
supra-coeliac aorta with success. The donor liver
is usually revascularised with the removal of the
supra-hepatic clamp followed by the infra-­
hepatic clamp, portal vein and artery. After care-
ful hemostasis of bleeding areas, either from the
cut surface of liver or from any of the other major
bleeding points, the operation is completed by
performing the biliary reconstruction. In biliary
atresia patients and those with a reduced/split or
living donor liver graft, a Roux-en-Y choledo-
chojejunostomy is performed with fine absorb-
able sutures. Occasionally, in pediatric cases, a
duct-to-duct anastomosis may be performed with
a whole liver graft in a recipient with a normal
extra-hepatic biliary system. Stents or T tubes are
optional with some evidence of increased biliary
complications associated with their use. The only
real advantage is access to the biliary system dur-
ing the post-operative period. Finally, hemostasis
is obtained, and the wound is closed with drain-
age to the supra-hepatic and infra-hepatic spaces.
If there is any tension at sheath closure due to
bowel oedema or graft size, it is wise to insert a
temporary patch of Gortex or other non-adherent
material as a ‘tight’ abdominal closure is associ-
ated with an increased incidence of vascular
87  Pediatric Liver Transplantation
Fig. 87.1  Operative photograph shows a reduced size
liver (segments 2, 3 and part of 4) after transplantation
into an infant before abdominal closure. In this case, a
temporary silastic patch would be required to accommo-
date the large liver without any compression of the graft or
increased intra-abdominal compartment pressure
thrombosis and graft dysfunction. It is usually
possible to obtain skin closure over the patch
without too much tension. The patch can be
removed 5–10 days later (Fig. 87.1).
87.7 Living Related Donors
Living related donation of the left lateral seg-
ment, first successfully performed by Strong, has
become widely accepted as a method of acquir-
ing a liver graft in the face of severe donor short-
ages, particularly in countries with cultural or
religious reticence to accept brain death in a ven-
tilated heart-beating donor. There are clear
advantages in the planned nature of the proce-
dure, preferably before end-stage liver disease in
the recipient, namely, the excellent quality of the
graft and short ischemic time. Refinement of the
surgical technique has allowed for the use of liv-
ing donors and decreased donor liver-split grafts,
thus expanding the pool of available organs and
reducing the waiting list (Cuenca et  al. 2017;
Montenovo et al. 2019; Yersitz et al. 2003). The
only advantage to the donor is a psychological
one and there is a current morbidity of around
10% (wound sepsis, hernia, bile leak and adhe-
1203
sive bowel obstruction). There is also a reported
mortality of around 0.2% although there are
many live donor programs running successfully
without mortality. There are ethical concerns,
which appear justified, as with more widespread
transplant activity mortality and morbidity of
donors has been recorded. The donor should first
undergo a thorough screening, both clinical and
psychological without coercion and be given an
option to withdraw from the procedure at any
time before the transplant. It is important to rec-
ognize the limitations of living related liver trans-
plantation as the major source of organs for
children. Parents usually approach living related
liver transplantation with enthusiasm. They
should be advised of the high chance of unsuit-
ability, including the finding of significant pathol-
ogy and the possible complications, including
death.
87.8 Split Liver Transplantation
The donor pool for children has been extended by
the use of cut-down, split, living-related and more
recently, non-heart-beating donors and isolated
hepatocyte transplantation. In the split technique,
the liver of the donor is divided into two func-
tional units thus making maximum use of this
very scarce resource. The procedure may be per-
formed in situ, which has been associated with
less post-reperfusion bleeding from the cut sur-
faces and improved graft function. However, this
technique is time consuming and may not be fea-
sible in many cases. Early results with ex-situ dis-
section on the back table were poor, and this was
in part due to a technical learning curve, pro-
longed preservation times and the use of grafts for
only those recipients in the worst condition.
Recent results are much improved. Clearly, the
infrastructure must be in place to either perform
two transplants simultaneously or alternatively
one hemi-liver graft is exported to another centre.
The in situ technique is similar to a living related
transplant. Cholangiography is essential because
there is considerable variety in the intra-hepatic
1204 K. Sharif and A. J. W. Millar

biliary anatomy. Angiography is desirable but not
essential. Some centres leave the hepatic artery
with the right lobe making it more acceptable as
an ‘export’. The left lateral segment graft remain-
ing is then used locally. The utilization of segment
4 varies among centres, in children needing larger
graft i.e. left lobe, segment 4 is used with the left
graft; in other centres segment 4 is resected prior
to transplant, and in others, segment 4 is left with
right lobe. Likewise, arterial anatomy is carefully
examined and apportioned to each hemi-liver.
There is usually a sufficient peri-ductular vascular
network to ensure adequate blood supply to the
relevant duct systems, so long as the main hepatic
artery branch to each hemi-liver is preserved.
Accessory arteries can be ligated if there is ade-
quate ‘back bleeding’ after reperfusion.
87.9 Medical Management
87.9.1 Post-operative Care
Patients are monitored intensively post-­
operatively and usually require ventilation for a
minimum period of 24–48 h. The post-transplant
protocol from our centre is summarized in
Table 87.3.

Table 87.3  Post-transplant management

Immediate management on arrival to ICU:
 Baseline measurements of FBC, urea & electrolytes, blood gases, Ca, PT, PTT, chest X-ray, lactate.
 Haemoglobin level must be kept between 8 and 10 g/dl (or haematocrit< 0.35).
Measurements of fluid, urine and drain output, frequency of monitoring and investigations:
Hourly:
 Blood pressure, heart rate on ECG monitor, central venous pressure, core/peripheral temperature difference,
cutaneous oxygen saturation, fluid balance including colloid, wound drainage, urine output and drug volumes.
4 hourly for 24 h, if required:
 Arterial blood gases, serum Na, K, Ca, blood glucose and lactate.
 Gastric pH is recorded 6 hourly by nasogastric aspiration (aim to keep pH >5).
Daily:
 Chest X-ray, if indicated
 FBC and clotting (PT, PTT)
 Cyclosporin/tacrolimus level
 Urea, creatinine, calcium, phosphate, magnesium, total proteins, albumin, CRP, PCT.
 Full LFTs (bilirubin total & unconjugated, ALP, ALT, AST, GGT)
 Culture: wound swabs
 Drain fluid
Endotracheal tube aspirates if indicated
(i) Anti-microbials
 Intravenous Tazobactam 90mg/kg/tds
 Metronidazole 8 mg/kg/dose (maximum 500 mg) tds over 1 h (can give over 30 min) for 48 h
 Co-trimoxazole: Up to 5 years 240 mgs daily orally, over 5 years, 480 mgs od orally, over 5 years 480 mgs od
orally
(ii) Anti-virals
 If the donor status for EBV and CMV is known, all transplanted children are to be started on antiviral
prophylaxis.
If donor is EBV NEGATIVE
Recipient status Donor status Treatment
CMV –ve CMV −ve Stop aciclovir
CMV +ve CMV −ve Stop aciclovir
CMV +ve CMV +ve Continue aciclovir
CMV −ve CMV + ve Continue aciclovir
87  Pediatric Liver Transplantation 1205

Table 87.3 (continued)
If donor is EBV POSITIVE
All to continue on aciclovir irrespective of recipient EBV/CMV status.
(iii) Anti-fungals
 mycostatin 100,000 units (= 1 ml) orally qds, if > 10 kg 50,000 units (= 0.5 ml) orally qds, if < 10 kg
amphotericin 3 mg/kg for 7–10 days.
(iv) Gastric acidity prophylaxis
 Ranitidine 3 mg/kg/dose (I/V) tds, if pH still < 5 use sucralfate PO/NGT 250–500 mg QDS and consider using
omeprazole 0.5 mg/kg/bd iv or orally.
(v) Immunosuppression
 Standard immunosuppression (i.e. first graft, no renal failure) begins post-operatively.
 Basilaximab (2 doses): Intra-operatively and day 4 post-transplant
 Tacrolimus (Prograf): First dose 0.2 mg/kg/dose orally given within 6 hrs of transplant.
 Then 0.1 mg/kg/dose bd orally to maintain levels of 10–15 nanogram/ml in the first 2 weeks.
Subsequently adjust the dosage to achieve serum levels of:
 7–11 ng/ml in 3rd/4th weeks
 4–7 ng/ml in 2nd/3rd month
 2–4 ng/ml thereafter
Steroids
 From Day 0: Hydrocortisone IV daily (First dose to be given as child reaches ITU):
Weight kg Dose
<20 50 mg bd
>20 100 mg bd
When oral diet is tolerated commence oral steroids (maximum daily dose 40 mg):
(vi) Antiplatelet treatment
 Aspirin (if platelet > 75,000) 3 mg/kg/day OD orally/NGT (maximum dose 75 mg)
 Dipyridamole (if platelets > 50,000)
 If patient weighs < 10 kg 25 mg tds orally, if weighs >
 10 kg 50 mg tds orally
(vii) Heparin infusion
 For vascular anastomosis at risk (complex vascular reconstruction or small diameter arteries or portal vein)
(viii) Analgesia and sedation
 Analgesia is achieved with morphine in the routine transplant patient with reasonable graft function and is titrated
against pain level be commenced within 72 h of surgery and may be supplemented by nasogastric feeding or
parenteral nutrition in the early phase if there is a delay in restoration of bowel function. Phosphate and
magnesium deficiency is common and requires replacement therapy in nearly all patients.

Liver ultrasound with colour flow Doppler is 87.10 Immunosuppression

performed for the first 5 days and later, as clini-
cally indicated, to confirm vascular patency and
the absence of biliary dilatation.
Hypertension is almost universal in pediatric
transplantation and can initially be managed with
nifedipine sublingually in conjunction with
diuretic agents. Subsequently, calcium channel
blockers may be given in appropriate dosage.
Aspirin 3 mg/kg given on alternate days is
used as prophylaxis against arterial thrombosis
and a proton pump inhibitor is given for gastric
mucosal protection.
There is considerable variation in the selection of
immunosuppressive agents. Most protocols cur-
rently employ triple therapy with tacrolimus,
methylprednisolone and a monoclonal interleu-
kin-­2 inhibitor (CD 25) antibody. Some centres
use steroid-free immunosuppression. In addition,
there are a number of other strategies in place to
reduce the amount of nephrotoxicity, which is a
toxic side effect to both calcineurin inhibitors.
Thus, mycophenolate mofetil, an inosine mono-
phosphate dehydrogenase inhibitor, may be used
1206 K. Sharif and A. J. W. Millar

instead from early on in what is called ‘nephron
sparing immunosuppression’. Rapamycin, a drug
structurally similar to tacrolimus, which prevents
proliferation of T cells but acts at a different stage
of T cell activation from either cyclosporin or
tacrolimus, has the advantage that it is not neph-
rotoxic and does not interfere with transcription
and production of interleukin 2, rather it antago-
nizes the action of interleukin 2 on its receptor. It
has no adverse effects on liver function and may
be synergistic with cyclosporin. The recent con-
sensus conference of the International Liver
Transplantation Society developed guidelines on
immunosuppression in liver transplants recipi-
ents (Charlton et al. 2018). Recently, significant
research is focused on immune regulation with
regulatory T-cells (Safinia et al. 2018).
The methylprednisolone dosage is reduced
over the first week to about 1 mg/kg/day for the
first month and then reduced to a level of 0.3 mg/
kg/day to 0.2 mg/kg as maintenance. This can be
later reduced in some patients to alternate day
therapy or even withdrawn completely. Both
mycophenolate mofetil and rapamycin/sirolimus
can be used as renal sparing should nephrotoxic-
ity become evident. Use of humanised anti-CD25
monoclonal antibodies given before and during
the first week of the transplant has reduced the
incidence of acute rejection in the first 3 months
by around 30%, but long-term graft survival is
essentially the same as when these agents have
not been used. The other polyclonal anti-­
lymphocyte immunoglobulins are rarely used.
day in two divided doses 3 days a week for the
prevention of pneumocystis carinii infection for
at least the 1st year. Intravenous ganciclovir 5
mg/kg/dose 12 hourly is used as prophylaxis
against cytomegalovirus (CMV) and Epstein
Barr virus (EBV) infection, initially for 2
weeks, and this may be extended for up to 3
months in high-risk patients, who have not pre-
viously been exposed to CMV or EBV but have
received a donor graft with previous exposure.
This considerably reduces the incidence of both
cytomegalovirus disease and post-transplantation
lymphoproliferative disorder. Either hyperim-
mune cytomegalovirus globulin or immunoglob-
ulin is also given to assist viral prophylaxis.
Leucocyte filtered blood products are used
throughout to reduce CMV load. Prophylactic
antibiotics are given with induction of anaesthe-
sia and continued for 3–5 days. These are changed
according to cultures taken of blood, secretions,
sputum and urine. Anti-tuberculosis prophylaxis
is given only if evidence of tuberculosis is found
before surgery and if a close family contact has
tuberculosis. Ofloxacin, rifampicin and ethambu-
tol or ethionamide may be used in addition to iso-
niazid but very careful monitoring of liver
function tests is required because all of these
drugs may be hepatotoxic and particularly rifam-
picin may result in a decrease in cyclosporin or
tacrolimus levels due to enzyme P450 induction
with increased drug metabolism.
87.12 Surgical Complications

87.11 Anti-Infection Agents Surgical complications may be reduced to an

absolute minimum with meticulous technique.
Immunosuppression naturally leads to suscepti- These may present early or late, as summarized
bility to bacterial, fungal and viral infections. in Table 87.4.
Fungal infection is a major and potentially fatal Most common surgical complications are as
complication in liver transplantation. Fungal pro- follows:
phylaxis is given as mycostatin orally before the Biliary complications continue to be a signifi-
transplant, to reduce Candida colonization of the cant problem, with an overall incidence of
gut and amphotericin after the transplant and between 10–20%, particularly in living related
continued for a period of several months. From left lateral segment grafts and split liver trans-
the first week after the transplant, trimethoprim-­ plants. These complications include bile leak,
sulphamethoxazole is given at a dose of 6 mg/kg/ anastomotic strictures, and non-anastomostic
87  Pediatric Liver Transplantation 1207

cations are best treated by immediate surgery and

Table 87.4  Summary of common post-operative problems
1. Biliary tract re-anastomosis, if a major leak is suspected.
 Stenosis or stricture Minor leaks from the cut surface of the graft are
 Anastomotic leak—often associated with hepatic generally managed with drainage. Late stricture
artery thrombosis
 Infection formation may be satisfactorily dealt with by
2. Rejection endoscopic or percutaneous balloon dilatation or
 Acute stenting.
 Chronic (vanishing bile duct syndrome) Graft ischemia either from hepatic artery
3. Infection—bacterial, viral, (CMV, EBV, Herpes
thrombosis or portal vein thrombosis can be a
Zoster, hepatitis B). fungal (Candida, Aspergillus),
parasitic (pneumocystis) devastating complication. Hepatic artery throm-
 Abdominal (peri- or intra-hepatic abscess) bosis represents a significant cause of graft loss
 Biliary tree and mortality after pediatric liver transplantation.
 Pulmonary
 Re-activated virus The reported incidence of this complication is
 Gastrointestinal tract 5–7%. The incidence is much less frequent with
 Catheter associated (intravenous, urinary tract) the use of reduced-size liver transplants with
4. Graft vascular injury (thrombosis, stenosis) larger size vessels and microsurgical vessel anas-
 Hepatic artery
 Portal vein tomosis techniques used for living donor trans-
 Inferior vena cava (supra and infra-hepatic) plants. Most centres recommend routine Doppler
ultrasound in the early post-operative period
 Hepatic vein (left lateral segment grafts), Budd-­
Chiari recurrence, hepatic venous outflow ranging from 3–7 days to confirm the patency of
obstruction (HVOO)
5. Renal dysfunction these vessels. Consequences of vascular throm-
bosis are graft necrosis, intra-hepatic abscess,
 Tacrolimus/cyclosporin or other drug-induced injury
 Tubular necrosis due to hypo-perfusion biliary necrosis and bile leakage. A massive rise
 Pre-existing disease (hepato-renal syndrome) in enzyme activity, particularly in the first few
 Hypertension
6. Miscellaneous days after transplant, may be the first signs.
 Encephalopathy (cyclosporin, tacrolimus, Immediate intervention with thrombectomy and
hypertensive, metabolic) re-anastomosis may be successful if the diagno-
 Bowel perforation (steroid, diathermy) sis and treatment are carried out as soon as the
 Diaphragm paresis/paralysis
 Gastrointestinal haemorrhage (peptic ulceration, complication is diagnosed in the first 3–5 days. If
varices) thrombectomy fails, urgent retransplant is
 Obesity (steroids) required. Late thrombosis may be asymptomatic
 Other drug side effects in the first 2 weeks following
and if so can be ignored. Although technical fac-
transplantation
tors usually account for most cases, it is advis-
able to maintain the hematocrit at around 30 to
strictures of the donor bile duct with sludge for- improve microvascular flow, and most centres
mation. Most biliary complications (72%) occur use aspirin and dipyrimidole as long-term pro-
(Kochhar et al. 2013). phylaxis, and in addition, some centres use intra-
Ultrasound and cholangiography are the prin- venous or subcutaneous heparin in the first week
cipal imaging modalities used for detection of post-transplant
these complications. It is imperative with all sus- Portal vein thrombosis can occur early or late.
pected biliary complications to ensure that the Early thrombosis usually presents with a degree
hepatic artery is patent using Doppler ultrasound of liver dysfunction with prolonged clotting
or angiography as hepatic artery thrombosis will while late thrombosis presents with portal hyper-
cause ischemia and necrosis of the biliary tree. tension, which may be heralded by an esophageal
Simple bile leaks are diagnosed in the early post-­ variceal bleed. Immediate thrombectomy may be
operative period by the presence of bile in drain- successful. Where graft portal vein thrombosis is
age fluid or in percutaneous aspirate of fluid established, a meso-portal (Rex) shunt, with a
collections around the liver. Early biliary compli- vein graft taken from the internal jugular vein of
1208
the patient or donor veins from a vascular bank
(if available) and interposed between the superior
mesenteric vein and the left branch of the portal
vein, may be curative. Significant risk factors for
portal vein thrombosis are young age and low
weight at the time of liver transplantation, small
portal vein and in urgent transplants, when the
patient is in a generally poor condition. Overall
risk of portal vein thrombosis (PVT) is 2–5%.
Bowel perforation is a well-recognized com-
plication following liver transplantation (+/-7%).
Contributory factors include previous operation,
steroid therapy, hypoxemia in porto-pulmonary
syndrome and viral infection. The incidence is
higher in children who underwent transplantation
for biliary atresia after a previous Kasai portoen-
terostomy. Diagnosis may be difficult and a high
index of suspicion is needed.
Post-operative fluid collections arising from
the cut surface of the liver has the reported inci-
dence of 30% and 40% in which nearly half
required intervention. These collections can be
due to biliary anastomosis leaks, bile leaks from
the cut surface of a partial liver graft or bowel
perforation. Late presentations may be less acute
and typically present with gram-negative sepsis,
liver abscess or biliary complications.
Inferior vena cava thrombosis may develop
either in the immediate post-operative period pre-
senting with ascites and lower body edema or
later on due to regeneration of the graft and twist-
ing of the caval anastomosis. Thrombolytic ther-
apy may be successful in late thrombosis but
should be avoided in early thrombosis because
uncontrollable bleeding may occur from raw sur-
faces, particularly if a reduced/split liver was
transplanted.
Hepatic venous outflow obstruction (HVOO)
is not an infrequent complication. This can be
due to redundancy of hepatic vein (when the graft
hepatic vein is kept long) or torsion in position-
ing of a partial graft. The correction of the redun-
dancy or torsion is made by pulling the graft
caudally and to the left or right side of the abdom-
inal cavity as determined by Doppler ultrasonog-
raphy and stabilizing the graft by suture of the
falciform ligament to the diaphragm. HVOO can
also be suspected if there is persistence of ascites
K. Sharif and A. J. W. Millar
in the early post-transplant period. This is usually
confirmed by either an angiography or by liver
biopsy findings of congestion and red cell extrav-
asation around central veins.
Diaphragmatic paresis and hernia are rare
complications of liver transplantation. The pos-
sible role of several contributing factors includes
cross-clamping of the IVC at the level of the dia-
phragmatic hiatus, trauma at operation (dissec-
tion and diathermy) and diaphragm thinness
related to low weight and malnutrition.
87.13 Common Medical
Complications
Most patients can be discharged from the inten-
sive care unit within the first week after trans-
plantation. Common medical complications of
transplantation include bacterial, viral, fungal
and opportunistic infections, renal function
impairment, hypertension, rejection and of par-
ticular concern is the post-transplant lymphopro-
liferative syndrome.
Infections: The reported incidence of infection
in the liver transplant population is 1.36 infec-
tion/ patient. The most common sites of infection
are bloodstream (36%) and abdomen (30%).
Gram-positive bacteria (78%) predominated over
gram-negative bacteria (22%). Detailed analysis
of risk factors shows that age < 1 year, body
weight < 10 kg, extra-hepatic biliary atresia,
intraoperative transfusion > 160 ml/ kg, mechani-
cal ventilation > 8 days and PICU stay > 19 days
are associated with a higher risk of infection.
Acute rejection: Despite the availability of
potent immunosuppressive drugs, rejection after
organ transplantation in children remains a seri-
ous concern and may lead to significant morbid-
ity, graft loss, and death of the patient. Diagnosis
of rejection can be made on the basis of clinical,
biochemical and histologic changes and usually
presents in the first few weeks after transplant
with fever, malaise, a tender graft and loose
stools. Diagnosis is confirmed by liver biopsies
performed using the Menghini technique Hypafix
needle (Braun) or spring loaded Tru-Cut needles,
unless biliary dilatation is observed on ultraso-
87  Pediatric Liver Transplantation
nography. Biopsy tissue is routinely assayed for
viral and bacterial activity. The grade of rejection
is assessed, according to established Banff histo-
logical criteria, on a scale of 0–4. Some centres
are trying to evaluate non-invasive tools to diag-
nose acute rejection, such as radiologic findings
on post-transplant Doppler ultrasound. Others
are using Interleukin 5 (IL-5), it is produced in
the liver and is a T cell-derived cytokine that acts
as a potent and specific eosinophil differentiation
factor in humans. During liver allograft rejection,
intragraft IL-5 mRNA and eosinophilia have
been observed. It may be useful as a specific
marker of allograft rejection. Once diagnosed,
acute rejection is treated with three doses of
methyl prednisolone 10 mg/kg given intrave-
nously on successive days with adjusted baseline
immunosuppression. Some patients experience
corticosteroid-resistant acute rejection, the man-
agement of which is not standardized. Various
agents used include the addition of myco-­
phenolate mofetil or sirolimus. Other options are
the use of anti-thymocyte globulins (ATG) or
monoclonal anti-CD3 antibodies, muromonab
CD3 (OKT3). In patients with refractory rejec-
tion despite therapeutic escalation, the risks of
over-immunosuppression, including opportunis-
tic infections and malignancies (especially the
Epstein-Barr virus related post-transplant lym-
phoproliferative disorder) have to be balanced
with the consequences of graft loss due to
rejection.
Late acute cellular rejection: Although acute
rejection is mostly encountered during the first 3
months after liver transplant, it may occur later
on. Late cellular rejection in children is usually
due to low or decreased immunosuppression and
is associated with long-term complications.
Prompt intervention to correct inadequate immu-
nosuppression and careful follow-up to identify
other treatable conditions is essential.
Antibody mediated rejection (AMR) is a well-­
known entity in kidney, heart, and pancreas trans-
plants. AMR in the liver transplant population is
a rare and possibly underdiagnosed condition. In
2016, the Banff working group on Liver Allograft
Pathology published consensus guidelines to
diagnosed AMR in liver transplants. AMR should
1209
be suspected in children with steroid-refractory
rejection, especially in the presence of class II
donor-specific antibodies (DSA) and presence of
C4d staining, along with other features.
Chronic rejection is an irreversible phenome-
non, which is chiefly intrahepatic and ductular
rather than a vascular phenomenon in contrast to
other organ transplants. This is usually mani-
fested by the disruption of bile duct radicals with
development of the vanishing bile duct syndrome.
The incidence seems less frequent with tacroli-
mus based immunosuppressive regimens as
opposed to cyclosporin where an incidence of up
to 10% has been recorded. Late chronic rejection
may also be associated with a vasculopathy
affecting larger arteries.
Chronic graft hepatitis occurs in 20–30% of
children after liver transplantation, but the preva-
lence and causes are not known. Serum liver
associated autoantibodies are often positive. It is
most frequently seen in children transplanted for
cryptogenic cirrhosis (71%). However, neither
hepatitis C nor hepatitis G infection was associ-
ated. Management is with re-introduction or
increase in steroid dose.
Cytomegalovirus (CMV) infection: Cyto-
megalovirus (CMV) infection (seroconversion
or virus isolation) and CMV disease (infection
plus clinical signs and symptoms) have a
reported incidence of 37% and 12% respectively
with significant morbidity and mortality. The
high prevalence of CMV infections supports the
view that clinical signs alone are inadequate to
direct investigations for CMV. Cytomegalovirus
(CMV) infection is best monitored with PP65
antigen and polymerase chain reaction (PCR)
measurement of the virus. Ganciclovir/valgan-
cyclovir remains an important therapeutic
option for the prevention and treatment of CMV
disease in transplant recipients. Prophylactic
treatment with ganciclovir appears the best
strategy to implement in high-risk patients. A
rare association with cytomegalovirus (CMV)
reactivation is haemophagocytic syndrome
(HPS). It is a rare event, which is often fatal.
These patients are treated with a combination of
antiviral agents, immunomodulatory and sup-
portive therapy.
1210
Epstein-Barr virus (EBV) and post-transplant
lymphoproliferative disorder (PTLD). EBV
infection is the main cause of PTLD. Since many
infants are EBV seronegative at the time of trans-
plantation, PTLD is a major concern for these
patients. Post-transplantation lymphoprolifera-
tive disorder (PTLD) presents from the first few
weeks after transplant to several years later with
a mean time of onset around 9 months. First man-
ifestations of PTLD are adenoidal and/or tonsil-
lar involvement. A typical presentation is usually
with acute membranous tonsillitis and associated
cervical lymphadenopathy, which is resistant to
antibiotic therapy. It is important to remember
that tonsillar enlargement in pediatric liver trans-
plant patients does not necessarily imply a diag-
nosis of PTLD.  Furthermore, the presence of
increased numbers of EBV infected cells in ton-
sils from liver transplant recipients by itself does
not indicate an increased risk of developing
PTLD.
However, the disease may be widespread and
gastrointestinal and central nervous system
involvement is common. Currently, there are no
tests to accurately identify pediatric liver trans-
plant patients at risk for post-transplant lympho-
proliferative disorder (PTLD). Attempts have
been made to use cytokine polymorphisms and
real-time quantitative polymerase chain reaction
(qPCR) Epstein-Barr virus (EBV) viral load to
identify patients at risk for PTLD development.
Use of cytokine genotyping, in conjunction with
qPCR for EBV viral load, can significantly
improve the predictive value of diagnostic tests
for identification of patients at high risk for
PTLD. Management strategies include reduction
of immunosuppression, which may require com-
plete withdrawal along with standard anti-­
lymphoma chemotherapy, particularly with the
monoclonal type. Mortality varies from 20% to
70% or more. Prophylactic intravenous ganciclo-
vir given for a prolonged period (two weeks min-
imum) may be effective in preventing EBV
activation which is the promoter of PTLD in most
cases. Rituximab, an anti-CD 20 monoclonal
antibody has been used with good effect. As B
K. Sharif and A. J. W. Millar
cells are largely ablated, replacement immuno-
globulin therapy is required until B cell recovery
has occurred.
Renal impairment: A degree of renal impair-
ment is almost inevitable in those patients suffer-
ing from chronic liver disease and with the
additional burden of the use of nephrotoxic
immunosuppressive drugs, such as cyclosporin
and tacrolimus with other nephrotoxic antibiotics
and antifungal agents, may result in significant
renal impairment of function in the long term.
The importance of renal sparing strategies in
immunosuppression is becoming increasingly
evident as long-term survivors present with drug-­
induced renal failure.
Retransplantation: Ten to 15% of patients
may suffer graft failure at some time and need
retransplantation. Early indications may be pri-
mary non-function, early hepatic arterial throm-
bosis, severe drug-resistant acute rejection and
established chronic rejection. Early retransplan-
tation is technically a much less traumatic proce-
dure than the original transplant, although the
patient may be in a poorer clinical condition.
Outcome largely depends on the indication for
retransplantation and is quite good for technical
causes but less satisfactory for rejection and
infection. An increasingly poorer outcome can be
expected after third and fourth retransplants and
the efficacy and ethics of these interventions are
in question.
87.14 Long-Term Survival
and Quality of Life
One-year survival of > 95% is being achieved in
the best centres, with predicted 10-year survivals
of around 85–90% (Ramos-Gonzales et al. 2019;
Kohli et al. 2018). Patients grafted for acute liver
failure have done less well, with a higher early
death rate usually associated with cerebral com-
plications and multi-organ failure. Excellent
quality of life can be achieved and most children
are fully rehabilitated. It is, however, increas-
ingly evident that prolonged cholestatic jaundice
87  Pediatric Liver Transplantation
and malnutrition in infancy may have late effects
and despite good physical rehabilitation evidence
of significant cognitive deficits, which present
during early schooling as learning difficulties and
attention deficit disorder, are common. Quality of
life may not reach perfection, and also depends
on the way society accepts these imperfections.
As with any immunosuppressed patient, the inci-
dence of neoplasia in a lifetime is greatly
increased.
87.15 Conclusion
Careful planning, extensive preparation of per-
sonnel and a broad base of skills, along with
good teamwork between health professionals,
are required for the development of a successful
pediatric transplant program. These services are
best provided within a pediatric hepatology and
transplant centre. Surgical technique, anaesthetic
skills, and medical care of the highest order are
essential. A patient with a liver transplant is a
patient for life and requires complete commit-
ment from the transplant medical and surgical
team, which cannot be abrogated after discharge
from hospital. Endemic viral and bacterial infec-
tions particularly HBV, CMV, EBV and PTLD
impact negatively on any program. Extended
hospital stay may be required, and this, along
with long-term therapy, may be extremely
expensive. The need for pediatric liver trans-
plants has been assessed at approximately 1–5
children per million per year. Thus, transplant
activity should be concentrated in specific cen-
tres preferably doing more than 12 transplants a
year. The shortage of donor organs will continue
and future efforts must be focused on maximum
use of cadaver donors and increasing living
related donation. Transplant activity is rapidly
increasing throughout the developing world.
This endeavour should be strongly supported as
poor socio-economic status and not a contraindi-
cation to transplantation. Parents with relatively
few material resources have been shown to be
able to diligently care for their children. No child
1211
with end-stage liver disease should be denied the
opportunity of receiving appropriate treatment.
As with any new development, knowledge and
experience improve, costs decline and excellent
outcomes are achieved. These challenges must
be met to offer any infant or child requiring liver
replacement, a chance of a life. The ultimate aim
is to restore the child to normal health such that
he/she can grow up into a productive healthy
adult, who can make his/her contribution to soci-
ety and develop all of his/ her human potential. A
normal pregnancy can be expected in women
who received transplants as children.
References
Brenner DA, Thomas E (2017) Starzl: transplantation pio-
neer. Proc Natl Acad Sci U S A 114:10808–10809
Charlton M et al (2018) International Liver Transplantation
Society consensus statement on immunosuppres-
sion in liver transplant recipients. Transplantation
102:727–743
Cuenca AG et  al (2017) Pediatric liver transplantation.
Semin Pediatr Surg 26:217–223
de Ville de Goyet J et  al (1995) Standardized quick en-­
bloc technique for procurement of cadaveric liver
grafts for pediatric liver transplantation. Transpl Int
8:280–285
Ferreira-Meirelles R et  al (2015) Liver transplantation:
history, outcomes and perspectives. Einstein (Sao
Paulo) 13:149–152
Fischler B et  al (2019) Similarities and differences in
allocation policies for pediatric liver transplanta-
tion across the world. J Pediatr Gastroenterol Nutr
68:700–705
Kochhar G et  al (2013) Biliary complications follow-
ing liver transplantation. World J Gastroenterol
19:2841–2846
Kohli R et  al (2018) Liver transplantation in children:
state of the art and future perspectives. Arch Dis Child
103:192–198
Mazariegos G et al (2014) Liver transplantation for pediat-
ric metabolic disease. Mol Genet Metab 111:418–427
Miller C et al (1988) Rapid flush technique for donor hep-
atectomy: safety and efficacy of an improved method
of liver recovery for transplantation. Transplant Proc
20:948–950
Montenovo MI et al (2019) Living liver donation improves
patient and graft survival in the pediatric population.
Pediatr Transplant 23:e13318. https://doi.org/10.1111/
petr.13318
1212
Otte JB (2002) History of pediatric liver transplantation.
Where are we coming from? Where do we stand?
PediatrTranspl 6:378–387
Ramos-Gonzales G et  al (2019) Predictors of need for
liver transplantation in children undergoing hepato-
portoenterostomy for biliary atresia. J Pediatr Surg
54:1127–1131
Safinia N et  al (2018) Cell therapy in organ transplan-
tation: our experience on the clinical translation of
K. Sharif and A. J. W. Millar
regulatory T-cells. Front Immunol 9:e354. https://doi.
org/10.3389/fimmu.2018.00354
Shneider BL (2002) Pediatric liver transplantation in
metabolic disease: clinical decision making. Pediatr
Transplant 6(1):25–29
Yersitz H et  al (2003) One hundred in situ split-liver
transplantations. A single center experience. Ann Surg
238:496–507
 Part X
Genitourinary Disorders
 Urinary Tract Infection
Thomas de los Reyes and Martin A. Koyle
88.1 Introduction
UTIs may result from a multitude of bacterial
pathogens, usually gram-negative bacteria, and
although they can occur anywhere within the
urinary tract, the kidneys (pyelonephritis) and
bladder (cystitis) represent the most common
sites manifested clinically. Anatomically, pyelo-
nephritis is considered to be an upper UTI, while
cystitis is considered to be a lower
UTI. Pyelonephritis is most commonly associated
with fever. It has been suggested that the younger
the patient, infants and neonates, that are untreated,
especially repeated episodes of pyelonephritis, can
lead to renal scarring and permanent renal damage
(Jacobson et al. 1989). Up to 40% of infants with a
bout of pyelonephritis will show evidence of renal
scarring upon subsequent radioisotope Tc99 DMSA
nuclear scintigraphy of the kidneys (Hewitt et al.
2008). For this reason, the differentiation between
upper and lower tract infections is critical.
However, it is important to appreciate the limita-
tions behind the evidence that shaped our initial
understanding of the relationship between pyelo-
nephritis, scarring, and long-term effects. More
contemporary studies have shown that some scars
T. de los Reyes · M. A. Koyle (*)
The Hospital for Sick Children and University of
Toronto, Toronto, ON, Canada
e-mail: thomas.delosreyes@sickkids.ca;
martin.koyle@sickkids.ca
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_88
88
noted on DMSA nuclear scintigraphy may be con-
genital, rather than as a result of infection (Pohl
and Belman 2009). The degree of scarring may
also be more relevant, rather than the presence of a
scar in itself. In a long-term follow-up of patients
with renal scarring, bilateral scarring was associ-
ated with a decrease in renal function, while uni-
lateral scars were no different from those with no
scarring (Wennerström et  al. 2000). It also
appeared that the presence of renal scarring did not
affect mean 24-h ambulatory blood pressure mea-
surements (Wennerström et al. 2000).
In addition to relating UTIs to the anatomic
location, UTIs can also be classified in several
other ways. A UTI can be categorized as compli-
cated or uncomplicated or whether it is the first
infection or a recurrent episode (Table 88.1).
88.1.1 Complicated vs.
Uncomplicated UTI
A UTI is considered complicated if the patient
has any of the following factors: anatomic or
functional abnormalities of the genitourinary sys-
tem, recent instrumentation, retained foreign
body such as urethral catheters, a documented
fever, or occurring in an infant or neonate.
Conversely, uncomplicated UTIs are infections
that arise in the absence of any of these factors.
Another important consideration that affects
the management and clinical evaluation of a
1215
1216
Table 88.1  Classification of UTI
Anatomic location
 Upper—kidney (pyelonephritis)
 Lower—bladder (cystitis), urethra (urethritis)
Complicating factors
 Complicated infection—infant or neonate, fever,
foreign body or stone, anatomic or functional
abnormalities
 Uncomplicated infection—simple lower tract
infection without fever
Initial or recurrent infection
 First infection—evaluate if complicated or
uncomplicated recurrent infection
 Unresolved—urine cultures are always + with the
same pathogen
 Bacterial persistence—urine cultures become
sterile after initial infection; recurrent infection
occurs with the same bacteria
 Re-infection—the original infection is eradicated,
and then, a different bacterium is isolated with
recurrent infection
child is whether the UTI is the first or a recur-
rent infection. Recurrent infections fall into
three categories, namely, unresolved infections,
infections with bacterial persistence, and true
re-infections.
88.1.2 Unresolved Infection
In patients with unresolved bacteriuria, initial
urinary cultures, as well as all subsequent cul-
tures, will show persistence of the same bacteria.
In these cases, the original bacterial infection has
never been cleared from the urinary tract. This is
often the result of inadequate treatment, due to
either poor compliance or antibiotic resistance. In
recent years, uropathogens have developed
increased resistance to commonly prescribed
antibiotics for UTI.  Notable examples of this
increasing resistance are Escherichia coli’s resis-
tance to ampicillin and vancomycin-resistant
enterococcal infections. The urine culture, at the
time of initial presentation, will reveal not only
the type of bacteria but also the sensitivity and
resistance patterns to various antibiotics. A fol-
low-­up of the results of a urine culture by the cli-
nician can help prevent inadequate and
inappropriate antibiotic treatments of the infec-
tion. Proof of cure urinalysis and culture, how-
T. de los Reyes and M. A. Koyle
ever, are typically not indicated in the vast
majority of patients who have otherwise
responded to initial therapy. Asymptomatic bac-
teriuria, in which a positive urine culture is
obtained in patients who lack symptoms or signs
of an infection, can result in unnecessary treat-
ment which in turn can lead to antibiotic resis-
tance, adverse reactions, and increasing
healthcare costs. Counseling of the patient and
caregiver is also important to ensure that pre-
scribed antimicrobials be taken for the entire pre-
scribed course even if symptoms resolve prior to
course completion.
88.1.3 Bacterial Persistence
In patients with bacterial persistence, the same
bacteria are cultured in the urine, despite the ini-
tiation of and adherence to sensitivity-adjusted
therapy. Bacterial persistence can indicate an
occult nidus or reservoir for the infective process.
The nidus for persisting infection can be retained
foreign body, such as a piece of catheter or ure-
teral stent, or a chronically infected stone. In addi-
tion to an actual physical nidus for recurrent
infection, an anatomic or functional obstruction
can also act as a reservoir for these infections.
Obstruction of urinary flow eliminates the normal
antegrade flow of urine and promotes urinary sta-
sis and chronic infection. Anatomic obstruction of
the urinary system may occur at different levels of
the urinary tract. Supra-vesical obstruction
includes congenital anomalies, such as uretero-
pelvic junction obstruction or obstructed mega-
ureter, while bladder outlet obstruction can occur
from posterior urethral valves, to name a few.
Functional obstruction can include conditions
such as neurogenic bladder, whereby both bladder
storage and emptying can be affected. This would
be in the form of elevated filling pressures result-
ing in hydroureteronephrosis or impaired bladder
contractility resulting in urinary stasis. Such
occurrences, in turn, can serve as a reservoir for
infection in either the upper or lower urinary tract.
Both functional and anatomic obstructions pro-
mote recurrent infections by eliminating the nor-
88  Urinary Tract Infection
mal antegrade flow of urine which is an important
physiologic host defense factor.
88.1.4 Re-infection
In bacterial re-infection, the initial infection is
resolved, and the patient has another documented
UTI. Re-infection can occur with the same patho-
gen but of differing serologic strain or with a dif-
ferent pathogen as a whole. Re-infection may be
a recurrent problem in patients who are geneti-
cally susceptible to UTIs. Treatment strategies
for patients with recurrent infections include
treatment with appropriate antibiotics, surgery to
address anatomic anomalies if present, and pro-
phylaxis with daily antibiotics if the re-infections
are common or accompanied by complicating
factors.
88.2 Historical Overview
UTIs and subsequent sepsis carried a high risk of
death in children prior to the advent of effective
antibiotic therapy, with mortality rates as high as
20% previously described (Zorc et al. 2005). The
development of antimicrobials, improved diag-
nostic capabilities including readily available
point-of-care screening, and an overall under-
standing of sepsis management mean that UTIs
are no longer the fatal illness it once was.
However, UTIs continue to cause significant
morbidity warranting adequate understanding of
clinicians in its diagnosis and management.
88.3 Incidence
During the first few months of life, several studies
have shown that boys have a higher incidence of
UTI at up to 14% compared to girls at 7% (Shaikh
et al. 2008). It has been substantiated that uncir-
cumcised boys, in particular, are the most prone to
UTI within the first year of life, with a tenfold risk
compared to circumcised males (Shaikh et  al.
2008). Despite this, circumcision remains a con-
troversial subject as a protective element for this
1217
Table 88.2  Yearly incidence of UTI in pediatric patients
Age Boys% Girls %
0–1 year 2.7 0.7
1–5 years 0.1–0.2 0.9–1.4
6–16 years 0.04–0.2 0.7–2.3
Source: Foxman B. Epidemiology of urinary tract infec-
tions: incidence, morbidity, and economic costs. Am J
Med. 2002; 113(1A):5S–13S
indication only. In high-risk situations, however,
the surgeon must weigh the risks and benefits.
After the first few months of life and continuing
into adulthood, girls have a higher overall inci-
dence of UTI. In children aged 1–5 years, the inci-
dence of UTI per year in boys is 0.1–0.2%, and in
girls, it is 0.9–1.4% (Foxman 2002). From the age
of 6 to 16, the reported incidence of UTI is stable.
In boys, it is up to 0.4% per year, and in girls, it
approaches 2.3% per year (Table  88.2; Foxman
2002). With the onset of sexual activity, this inci-
dence climbs to 11% in women, but is not very
prevalent in men. Overall, about two-thirds of
women will have a urinary tract infection through-
out their life (Foxman 2002).
88.4 Etiopathogenesis
Some hypothesize that in the first several months
of life, UTIs may be caused by hematogenous
spread of pathogens from transient bacteremia.
After this time period, however, most accept that
UTIs more frequently arise from fecal pathogens.
In these infections, ascending bacteriuria from
contamination and colonization of the perineum
and urethral meatus is thought to be the main fac-
tor in the development of UTI. This may offer an
anatomic explanation regarding the increased
incidence of UTIs in females. The peri-urethral
folds and the moist environment of the vagina
promote bacterial colonization around the ure-
thral meatus. In addition, the shorter length of the
urethra allows ascending infection to spread
more easily to the bladder than in the male.
Bacterial virulence factors enable some to
adhere to the urothelial lining of the genitouri-
nary tract. The process of colonization and subse-
quent proliferation is a complex balance between
1218
Table 88.3  Common uropathogens
Bacteria Incidence %
Escherichia coli 50–93
Enterococcus spp. 0–17
Klebsiella spp. 0–10
Serratia spp. ~1
Staphylococcus aureus ~1
Pseudomonas aeruginosa ~1
Enterobacter cloacae ~1
Streptococcus spp. ~1
Proteus spp. ~1
Source: Edlin RS, Shapiro DJ, Hersh AL, Copp HL. Anti-
biotic resistance patterns of outpatient pediatric urinary
tract infections. J Urol. 2013;190(1):222–227
the host’s immune factors and the virulence fac-
tors expressed by the bacteria. Proximal involve-
ment of the kidneys is believed to result from an
ascending infection originating from the bladder,
either by vesicoureteral reflux (VUR) or in a third
or more patients without VUR, due to the com-
plex relationship between host immunity and
susceptibility and innate pathogenicity and prop-
erties of that bacteria, as discussed below
(Montini et al. 2011).
Numerous organisms including viruses, yeast,
and bacteria can cause UTI. Bacterial infection,
however, is the most common cause of UTI in an
otherwise healthy, immunocompetent child. Of
the uropathogenic bacteria, E. coli is the most
common cause of UTI in both adult and pediatric
populations. E. coli account for up to 83% of bac-
terial UTIs (Edlin et al. 2013; Shaw et al. 1998).
Other common bacteria that cause urinary tract
infections include Enterococcus spp., Klebsiella
spp., Enterobacter cloacae, Serratia spp.,
Staphylococcus aureus, Pseudomonas aerugi-
nosa, Streptococcus spp., and Proteus spp.
(Table 88.3).
88.5 Risk Factors
88.5.1 Gender
Gender plays a large role in susceptibility to
UTI. During the first year of life, UTIs are more
common in boys than in girls. This is more pro-
T. de los Reyes and M. A. Koyle
nounced in febrile infants within the first 6
months after birth, where the incidence is com-
paratively greater for uncircumcised boys as
noted earlier. Thus, the foreskin, especially the
unretractile prepuce that is normal in neonates
and infancy, appears to be a harbinger for early
male UTI.  After the first year of life, the inci-
dence of UTI changes with girls having a higher
incidence than boys well into adulthood.
88.5.2 Circumcision Status
Circumcision remains the most common surgi-
cal procedure performed in the United States
(Morris et al. 2014). There is evidence that show
the benefits of circumcision in decreasing the
risk of sexually transmitted diseases, HIV infec-
tion, development of penile carcinoma, and
UTIs (Morris et  al. 2014). In uncircumcised
boys, the risk of UTI is approximately ten times
more in the first year of life compared to their
circumcised counterparts. The physiologic rea-
son for this increased incidence of infection is
well elucidated. During the first year of life,
there is increased colonization of the peri-ure-
thral tissue of the inner prepuce, allowing the
retrograde ascent of infection more readily into
the bladder. The advantage conferred by cir-
cumcision is diminished in older boys as the
overall incidence of UTI decreases with increas-
ing age. In a meta-­analysis looking into circum-
cision for the prevention of urinary tract
infections published in 2005, 111 circumcisions
would need to be performed to prevent 1 UTI
(Singh-Grewal et  al. 2005). Routine circumci-
sion, however, remains controversial for a vari-
ety of cultural, psychosocial, and ethical
reasons. Combined with a relatively low
reported UTI incidence of 1–3% in boys less
than 1 year of age, many practitioners view the
psychosocial aspects of routine circumcision
may outweigh any potential medical benefits
from the procedure. Although not the purpose of
this chapter, the absence of high-­quality data, as
reflected by the unsuccessful attempt by the
Cochrane collaboration in 2012, means that cir-
88  Urinary Tract Infection
cumcision continues to be a contentious issue
with conflicting guidelines (American Academy
of Pediatrics Task Force on Circumcision 2012;
Jagannath et al. 2012).
88.5.3 Previous Infection
One of the most significant risk factors for UTI is
a previous episode of infection. Of children who
have a UTI in the first year of life, 23% of boys
and 12% of girls are fated to have a further UTI
in the several months following their initial
­infection (Shim et al. 2009). The number of pre-
vious infections also increases the likelihood of
recurrent infection (Keren et al. 2015). This risk
may be reflective of changes that occur physio-
logically in the urinary tract after a UTI which
predispose patients to recurrent infections or may
also be secondary to selection of patients who are
at high risk for UTI due to genetic or anatomic
factors.
88.5.4 Bladder and Bowel
Dysfunction
Bladder and bowel dysfunction (BBD), previ-
ously termed dysfunctional voiding and dysfunc-
tional elimination syndromes, describes any
abnormalities in the storage or emptying of urine,
but recognizes the importance of fecal elimina-
tion, specifically constipation (Austin et  al.
2016). Increased fecal load is thought to affect
bladder storage and emptying, possibly through
direct mechanical compression or through induc-
ing changes in physiological neural stimuli lead-
ing to decreased urge to evacuate (Malykhina
et al. 2017). The resultant urinary stasis may then
increase the risk of UTIs. The Careful Urinary
Tract Infection Evaluation (CUTIE) study
enrolled 195 children without VUR who had 1 or
2 febrile or symptomatic UTI. This study found
1219
the risk of febrile or symptomatic UTI recurrence
was 35% in children with BBD (Keren et  al.
2015; Shaikh et al. 2016). Multiple studies of dif-
fering quality have shown the same conclusions
regarding the importance of BBD and UTI recur-
rence (Meena et al. 2020).
88.6 Pathophysiology
Most pathogenic bacteria that cause UTI arise
from a reservoir in the intestinal tract. As men-
tioned, E. coli is by far the predominant bacte-
rium that causes UTI. One way of differentiating
E. coli strains is based upon differences in the
antigens expressed on the polysaccharide capsule
that surrounds the bacteria (Tullus et  al. 1991).
These antigens are known as K antigens, and it
has been demonstrated that certain K antigenic E.
coli have a much higher propensity for causing
UTI than other strains. Another predictor of a
bacteria’s uropathic potential is its ability to
adhere to the urothelium where they cause infec-
tion. Pili or fimbriae are long filamentous append-
ages composed of protein that project from the
bacterial surface and allow this adhesion to take
place. In E. coli, type 1 pili are highly associated
with bacteria that cause UTI.  Type 1 pili bind
uroplakin, a protein cap that is expressed by the
urothelial cells. Another form of pili, P pili, is
highly associated with strains of E. coli that cause
pyelonephritis. Bacterial adherence, coloniza-
tion, and subsequent infection are a complex pro-
cess that involves a balance between bacterial
virulence factors and the host’s immune response
to invasive bacterial infection and colonization.
88.7 Pathology
The urothelium of the collecting system is com-
prised of a layer of superficial umbrella cells, an
intermediate layer, and basal layers. The latter is
1220
thought to be the location of stem cells that
enable umbrella cell regeneration following reg-
ular turnover similar to epithelial tissue in other
parts of the body. Urinary pathogens, such as E.
coli, are found exclusively in the umbrella cell
layer in part due to adherence factors described
above (Hickling et  al. 2015). A gram stain on
cross sections of pathologic specimens enables
the detection and differentiation of bacteria
based on its staining properties and morphologic
appearance.
88.8 Diagnosis
Diagnosis of UTI in an infant and neonate may
be quite difficult as the signs and symptoms of a
UTI are nonspecific. Signs and symptoms of UTI
in a febrile infant can include abdominal, supra-
pubic, or flank pain, poor appearance, jaundice,
foul smelling urine, and lack of other identifiable
etiology for an unexplained fever (Hoberman
et al. 1993). Among febrile infants in the first 2
years of life without an identifiable etiology for
fever and a temperature greater than 38.0 °C, it
was found that the incidence of UTI was 3.3%
(Craig et al. 2010). UTI is one of the more com-
mon serious causes of infant and neonatal sepsis,
and in the first 8 weeks of life, the incidence of
UTI in a child with an unexplained febrile illness
is as high as 13% (Crain and Gershel 1990). In
any unexplained febrile illness in a child who is
not capable of verbalizing the symptoms, a high
suspicion of UTI must be maintained.
88.8.1 Urinalysis, Microscopy,
and Culture
A urine analysis (UA) will provide information
about the urine which can suggest the present of
UTI. Nitrite, when present, suggests bacteria in
the urine as it is a byproduct of bacterial metab-
olism of nitrates. Although many gram-positive
and gram-negative bacteria are capable of this
conversion, Enterococcus spp., the second or
third leading cause of UTI in pediatric patients,
do not possess this ability. Thus, the test is spe-
T. de los Reyes and M. A. Koyle
cific but not sensitive. A UA in a patient with a
UTI will often also show the presence of leuko-
cyte esterase. Leukocyte esterase is an enzyme
released when white blood cells are lysed in
urine and indicates a significant degree of
leukocyturia.
An additional important part of the UA is the
microscopic examination of the urine. The col-
lected urine sample is typically centrifuged at
2000 revolutions per minute for 10 min (Lin et al.
2000a). The resultant pellet is then examined
under microscope. When there are greater than
five white blood cells (WBC) per high-power
field (hpf), significant leukocyturia exists. As the
number of WBC/hpf increases, the positive pre-
dictive value of the microscopic exam for UTI is
greater (Lin et al. 2000b). Additional sensitivity
can be achieved with the use of a hemocytometer,
which measures the number of WBCs in an
uncentrifuged specimen of urine. The sensitivity
of predicting UTI accurately by finding greater
than 10  WBC/hpf using the hemocytometer is
23% higher as compared to finding greater than
5  WBC/hpf on microscopic exam (Lin et  al.
2000a, b). Overall, the positive predictive value
for a UTI when there is nitrite and leukocyte
esterase on urinalysis coupled with bacteria and
leukocytes in microscopy approaches 100%
(Lohr et al. 1993). Conversely, the negative pre-
dictive value for UTI when these factors are not
present also approaches 100% (Lohr et al. 1993).
When a UTI is suspected, based upon clinical
suspicion, and a UA with microscopy suggests the
diagnosis, bacterial culture of the urine remains
the gold standard for establishing the diagnosis of
UTI. A culture should be sent before the initiation
of antibiotic therapy. However, the initiation of
antimicrobial therapy should not be delayed as
urine culture results typically return 24–48 h after
collection. Traditionally, growth of greater than
105 colony-forming units (CFU) of bacteria within
a urine culture indicates significant infection. This
limit has been challenged in adults and many
symptomatic women who are found to have fewer
bacteria than 105 CFU and often develop signifi-
cant UTI if they are not treated. Similarly, signifi-
cant UTI can be present in febrile children, despite
having fewer than 105 CFU.  In one study, up to
88  Urinary Tract Infection
27% of febrile infants presenting with UTI only
grow 104–105 CFUs upon urinary culture (Lin
et  al. 2000a, b). For this reason, the American
Academy of Pediatrics guidelines on urinary tract
infections, published in 2016, have changed the
diagnosis definition to the presence of pyuria and
at least 50,000 CFUs/mL of a single uropathogen
(Subcommittee on Urinary Tract Infection AAP
2016). The urine culture will also reveal the bac-
teria responsible for the UTI as well as the antibi-
otic sensitivity and resistance pattern of the
bacteria. This then allows for focused antibiotic
treatment strategies.
88.8.2 Urine Collection
The simplest collection method for a sample of
urine for analysis and bacterial culture is a bag
taped to the perineum. This method is helpful if
the urine analysis is negative, making infection
very unlikely. However, this method is not
­helpful if the urine suggests infection since gen-
ital, perineal, and fecal contamination can easily
give a false positive UA.  Two other methods
have been commonly employed to obtain a ster-
ile sample of urine when the proper diagnosis is
critical, such as in an infant or neonate with
unexplained febrile illness. These methods are
suprapubic aspiration of the bladder and sterile
urethral catheterization. To obtain a urine sam-
ple by suprapubic aspiration, a fine needle (23
or 25 gauge) is passed into the bladder by punc-
ture directly above the pubic symphysis. The
needle is directed perpendicularly or with a
slight caudal angle and aspirated as it is
advanced until urine is obtained. Point-of-care
ultrasonography may also help with identifica-
tion of the bladder and needle guidance. Urethral
catheterization is best accomplished with a 5
French feeding tube or specifically designed
urethral catheters which are usually designed
with a more compliant material. The technique
should be performed with as strict adherence to
sterile conditions as possible to prevent contam-
ination of the sample.
Both suprapubic aspiration and urethral cath-
eterization yield acceptable sensitivity for UTI. In
1221
a recent analysis of pain perceived by caregivers
and parents, urethral catheterization appeared to
be better tolerated by infants and neonates than
suprapubic aspiration (Kozer et  al. 2006).
Urethral catheterization is a simple method which
may be better tolerated by patients and is possi-
bly also perceived less invasive by parents.
Urethral catheterization may not be feasible
when it is complicated by anatomic problems
such as a dilated urethra or labial adhesions. In
some instances, urethral catheterization may be
aided by perineal pressure or gentle flushing of
the catheter with sterile saline as it is advanced.
When urethral catheterization is not achievable, a
urine sample can usually be easily obtained by
suprapubic aspiration.
88.8.3 Renal-Bladder Ultrasound
A renal-bladder ultrasound (RBUS) should be
performed in patients with a febrile urinary tract
infection (Subcommittee on Urinary Tract
Infection AAP 2016). The purpose of the ultra-
sound is to assess anatomic features, such as
renal size, architecture, and possible abnormali-
ties such as hydronephrosis or scarring that may
require further investigations. The timing of
when a RBUS should be performed remains con-
troversial. Based on the recommendations by the
American Academy of Pediatrics (AAP) and the
UK National Institute for Health and Clinical
Excellence (NICE), a RBUS should be performed
in patients who are seriously ill or are not exhibit-
ing the expected response to treatment within the
first 48 h of initiation. If a child responds well to
antibiotics, imaging can be postponed to within 6
weeks after the infection.
88.8.4 Further Work-Up
The full spectrum of imaging to identify urologic
pathology, such as a voiding cystourethrogram
(VCUG) and renal scintigraphy, is beyond the
scope of this chapter and will be addressed later
in the text. It is pertinent to note, however, in the
context of febrile urinary tract infections that
1222
recent guidelines including from the AAP recom-
mend performing a VCUG in patients with
abnormalities on ultrasound, such as hydrone-
phrosis, or in those with recurrent febrile UTI
(Subcommittee on Urinary Tract Infection AAP
2016). This is due to the association between
vesicoureteral reflux and pyelonephritis and
potential for subsequent renal damage.
88.9 Differential Diagnosis
Pediatric fever of unknown origin has a broad
differential including infectious, autoimmune, or
oncologic. Infectious causes can be further
divided either into viral, bacterial, fungal, and
other sources or by organ system, such as pneu-
monia, meningitis, or urinary tract infections.
88.10 Management
There are several controversies surrounding the
treatment of pediatric UTI.  These controversies
include the timing of treatment initiation, the
time course of adequate treatment, treatment with
oral versus parenteral antibiotics, and the need
and timing of radiologic imaging. An approach to
treatment of pediatric UTI can be based on
whether it is complicated or uncomplicated
(Koyle & Shifrin 2012).
88.10.1 Uncomplicated UTI
When an uncomplicated UTI has been identi-
fied based upon symptoms, a UA suggesting
infection, or a positive urine culture, oral anti-
biotic therapy should be initiated. The most
common pathogen is E. coli, and empiric anti-
biotic therapy should be guided by local resis-
tance patterns. Options for empiric therapy
include narrow-­ spectrum antibiotics such as
nitrofurantoin, sulfonamide-­ based antibiotics
such as trimethoprim-­ sulfamethoxazole (TMP-
SMX), or cephalosporins such as cefuroxime.
The duration of oral antibiotic therapy has been
an area of controversy in pediatric UTI treatment.
T. de los Reyes and M. A. Koyle
In a Cochrane collaboration review between
short-term (2–4 days) and long-term therapy
(7–10 days), there was no evidence that long-
term therapy was advantageous (Michael et  al.
2003). The recurrence rate and re-infection rate
remained the same up to 15 months after treat-
ment with either short- or long-term therapy
(Michael et  al. 2003). A more recent Cochrane
collaboration review published in 2012 compared
conventional 10-day antibiotic therapy to single-
dose therapy (Fitzgerald et  al. 2012). This had
similar outcomes to the aforementioned study,
with regard to having no difference in recurrence
or re-­infection rate. However, persistent bacteri-
uria at the end of treatment was reported in 24%
of children receiving single-dose therapy com-
pared to 10% of children who received conven-
tional 10-day therapy (Fitzgerald et  al. 2012).
Although there is not enough evidence that defin-
itively shows superiority of one duration versus
another, the totality of available evidence points
toward a 3–4-day course as reflected by the
European Society of Pediatric Urology/European
Association of Urology guidelines which is the
most recently available across all relevant bodies
(Stein et al. 2015). It is important to emphasize
that children with uncomplicated UTI are more
likely under the care of a primary care provider.
The surgical consultant is usually involved with a
child with a febrile UTI and a known or suspected
genitourinary surgical issue.
88.10.2 Complicated UTI
A common scenario presented to the consulting
surgeon will be an unexplained high fever in a
neonate or infant (Fig. 88.1). In this scenario, the
clinical should make a clinical judgment about
the need for admission to the hospital. If the child
shows evidence of dehydration and toxemia, and
if admission to the hospital is warranted, paren-
teral antibiotics are typically administered.
Broad-spectrum antibiotics are initiated after
proper urine collection for urinalysis and culture
and sensitivity has been performed. If the child
improves in 24–48  h, these can be switched to
oral antibiotics to complete a full course of treat-
88  Urinary Tract Infection 1223

2 month – 2 year
old with
unexplained fever

Consider UTI

Does the infant’s Obtain urine

“toxicity” warrant YES specimen by SP
immediate aspiration or urethral
antimicrobial catheterization;
therapy? culture urine

NO

Option: Perform Option: Obtain urine Initiate antimicrobial

urinalysis on specimen for culture by SP therapy parenterally;
collected by most aspiration or urethral consider
convenient method catheterization hospitalization

YES

Urinalysis positive for Urine Culture YES 7-14 day of antimicrobial

leukocytes, nitrite or positive? therapy guided by culture
WBCs? results

NO NO
YES Renal and Bladder
Clinical response
UTI unlikely in the Ultrasound as soon as
within 48 hours? convenient
absence of specific
symptoms. Follow
clinical course. NO UTI NO
Reconsider UTI if
fever persists. Renal and Bladder VCUG or Renal Nuclear
Ultrasound now; consider scan if ultrasound shows
repeating urine and blood abnormalities or recurrent/
culture atypical infection

Fig. 88.1  Algorithm for the treatment of infants with unexplained fever

ment spanning 7–14 days. If the urine culture
proves to be positive, antibiotic therapy can be
tailored to the sensitivities of the bacteria respon-
sible for the episode. Many antibiotic regimens
have been utilized for the treatment of pyelone-
phritis. The Cochrane collaboration published in
2014 reviewed trials that compared regimens and
outcomes of pyelonephritis (Strohmeier et  al.
2014). There was no difference in the various
treatment regimens published for the treatment of
pyelonephritis (Strohmeier et al. 2014). Most cli-
nicians utilize a third- or fourth-generation ceph-
alosporin such as ceftriaxone possibly in
conjunction with an aminoglycoside during the
acute phase of treatment. The patient is then tran-
sitioned to an equivalent oral medication once
fevers have abated, usually after approximately
48  h. By this time, urine culture results should
also be available to tailor antibiotic therapy based
on sensitivities. In the same Cochrane review
above, the evidence also showed that oral
­antibiotics alone are as effective as a short course
(3–4 days) of IV antibiotics followed by oral
therapy for a total duration of 10–14 days in
appropriately selected patients (Strohmeier et al.
2014). Typically, these are children who have a
fever but are otherwise able to tolerate oral hydra-
tion and non-toxic appearing. Oral antibiotic
agents that concentrate in the urine but obtain
poor tissue and blood levels should be avoided in
1224
patients who have had a febrile component to
their UTI. Nitrofurantoin is one of these antibiot-
ics, which remains a good choice for uncompli-
cated UTI and UTI prophylaxis.
88.10.3 Catheter-­Associated UTI
Catheter-associated UTI (CAUTI) is among the
most frequent cause of healthcare-associated
infections in the pediatric setting. CAUTI is
defined by the United States Centers for Disease
Control and Prevention as “a UTI where an
indwelling urinary catheter was in place for more
than 2 consecutive days in an inpatient location
on the date of the event, with the day of device
placement being day 1, and an indwelling cathe-
ter was in place on the date of event or the day
before.” Complications of CAUTI include
­pyelonephritis, sepsis, and death. Risk factors for
CAUTI include duration of catheterization, with
those having a catheter in place for a longer
period at an increased risk of 8% for each addi-
tional day that the catheter remained in place
(Gould et al. 2010). A history of prior catheter-
ization has also been shown to increase the risk of
CAUTI, possibly due to the colonization of the
urethra persisting after catheter removal (Gould
et al. 2010). This then serves as a nidus for infec-
tion upon re-catheterization due to disruption of
the urothelium. It is not clear, however, as to what
time interval constitutes a positive history for
prior catheterization. Treatment strategies to
tackle CAUTI include using urinary catheters
only when indicated, aseptic techniques in cath-
eter insertion, and daily evaluation regarding the
need for a catheter and potential catheter removal
(Gould et al. 2010). Unfortunately, most preven-
tion strategies are evaluated as a CAUTI preven-
tion bundle rather than the different components
T. de los Reyes and M. A. Koyle
of the above. Therefore, it is difficult to discern
which element has the greatest impact in CAUTI
prevention or reduction.
88.10.4 Antibiotic Prophylaxis
The use of continuous antibiotic prophylaxis in
pediatric UTI remains a controversial issue, espe-
cially in the era of increased antimicrobial resis-
tance by common uropathogens such as E. coli.
In several randomized trials involving patients
with known vesicoureteral reflux, the use of pro-
phylactic antibiotic therapy has been shown to
decrease the incidence of recurrent febrile UTI
(Brandström et  al. 2011; RIVUR Trial
Investigators et al. 2014). However, the benefits
of antibiotic prophylaxis in preventing renal
injury are controversial. Thus, most clinicians
would agree that beginning antibiotic prophy-
Table 88.4  Surgical conditions leading to febrile UTI
Anatomic
Vesicoureteral reflux
Ureterocele
Ureteropelvic junction obstruction
Posterior urethral valves
Ureteral or renal calculi
Obstructed megaureter
Functional
Neurogenic bladder—Myelomeningocele
Dysfunctional voiding—Hinman’s syndrome
laxis is warranted in young infants after an epi-
Table 88.5  Prophylactic antibiotic regimens for preven-
tion of UTI
Trimethoprim—1–2 mg/kg daily
Trimethoprim-sulfamethoxazole (TMP-SMX)—1–2 mg/
kg TMP, 10–15 mg/kg SMX
Nitrofurantoin—1–2 mg/kg daily
Cefixime—2 mg/kg daily
88  Urinary Tract Infection 1225

sode of pyelonephritis until imaging studies have trolled trial in children with dilating vesicoureteral
been completed to rule out surgical conditions reflux. J Pediatr Urol 7(6):594–600
Craig JC, Williams GJ, Jones M et  al (2010) The accu-
leading to febrile UTI (Table 88.4). Many differ- racy of clinical symptoms and signs for the diagnosis
ent antibiotics have been utilized to prevent of serious bacterial infection in young febrile children:
recurrent UTI.  The ideal antibiotic would have prospective cohort study of 15 781 febrile illnesses.
little effect upon the flora of the gut and be easy BMJ 340:c1594
Crain EF, Gershel JC (1990) Urinary tract infections in
to administer for the child and the parent since febrile infants younger than 8 weeks of age. Pediatrics
compliance is a major issue with these regimens. 86(3):363–367
Common regimens used include nitrofurantoin, Edlin RS, Shapiro DJ, Hersh AL, Copp HL (2013)
cefixime, or trimethoprim (Table 88.5). Antibiotic resistance patterns of outpatient pediatric
urinary tract infections. J Urol 190(1):222–227
Fitzgerald A, Mori R, Lakhanpaul M, Tullus K (2012)
Antibiotics for treating lower urinary tract infection in
88.11 Conclusion children. Cochrane Database Syst Rev (8):CD006857
Foxman B (2002) Epidemiology of urinary tract infec-
tions: incidence, morbidity, and economic costs. Am
UTI in pediatric patients can arise from a variety J Med 113(1A):5S–13S
of causes, encompassing simple cystitis to life-­ Gould CV, Umscheid CA, Agarwal RK, Kuntz G,
threatening urosepsis secondary to surgical Pegues DA, Healthcare Infection Control Practices
anomalies of the urogenital tract. There are a Advisory Committee (2010) Guideline for prevention
of catheter-­associated urinary tract infections 2009.
number of complex host and bacterial virulence Infect Control Hosp Epidemiol 31(4):319–326
factors that also make children susceptible to Hewitt IK, Zucchetta P, Rigon L et al (2008) Early treat-
UTI. It is important that the clinician be familiar ment of acute pyelonephritis in children fails to reduce
with the spectrum of UTI and children at risk of renal scarring: data from the Italian Renal Infection
Study Trials. Pediatrics 122(3):486–490
UTI, so that infants and children who need to be Hickling DR, Sun TT, Wu XR (2015) Anatomy and physi-
evaluated for anatomic or functional abnormali- ology of the urinary tract: relation to host defence and
ties of the urinary tract are correctly identified. microbial infection. Microbiol Spectr 3(4). https://doi.
The hallmark sign of one of these children is an org/10.1128/microbiolspec.UTI-­ 0016-­2012
Hoberman A, Chao HP, Keller DM, Hickey R, Davis HW,
infant in the first few years of life with evidence Ellis D (1993) Prevalence of urinary tract infection in
of a UTI and high fever. This child must be con- febrile infants. J Pediatr 123(1):17–23
sidered to have pyelonephritis and an anatomic Jacobson SH, Eklöf O, Eriksson CG, Lins LE, Tidgren B,
abnormality, until proven otherwise. Winberg J (1989) Development of hypertension and
uraemia after pyelonephritis in childhood: 27 year fol-
low up. BMJ 299(6701):703–706
Jagannath VA, Fedorowicz Z, Sud V, Verma AK,
Hajebrahimi S (2012) Routine neonatal circumci-
References sion for the prevention of urinary tract infections in
infancy. Cochrane Database Syst Rev 11:CD009129
American Academy of Pediatrics Task Force on Keren R, Shaikh N, Pohl H et al (2015) Risk factors for
Circumcision (2012) Male circumcision. Pediatrics recurrent urinary tract infection and renal scarring.
130(3):e756–e785 Pediatrics 136(1):e13–e21
Austin PF, Bauer SB, Bower W et al (2016) The standard- Koyle MA, Shifrin D (2012) Issues in febrile urinary
ization of terminology of lower urinary tract func- tract infection management. Pediatr Clin North Am
tion in children and adolescents: Update report from 59(4):909–922
the standardization committee of the International Kozer E, Rosenbloom E, Goldman D, Lavy G, Rosenfeld
Children’s Continence Society. Neurourol Urodyn N, Goldman M (2006) Pain in infants who are younger
35(4):471–481 than 2 months during suprapubic aspiration and trans-
Brandström P, Jodal U, Sillén U, Hansson S (2011) The urethral bladder catheterization: a randomized, con-
Swedish reflux trial: review of a randomized, con- trolled study. Pediatrics 118(1):e51–e56
1226
Lin DS, Huang FY, Chiu NC et  al (2000a) Comparison
of hemocytometer leukocyte counts and standard uri-
nalyses for predicting urinary tract infections in febrile
infants. Pediatr Infect Dis J 19(3):223–227
Lin DS, Huang SH, Lin CC et  al (2000b) Urinary tract
infection in febrile infants younger than eight weeks
of Age. Pediatrics 105(2):E20
Lohr JA, Portilla MG, Geuder TG, Dunn ML, Dudley SM
(1993) Making a presumptive diagnosis of urinary
tract infection by using a urinalysis performed in an
on-site laboratory. J Pediatr 122(1):22–25
Malykhina AP, Brodie KE, Wilcox DT (2017)
Genitourinary and gastrointestinal co-morbidities in
children: The role of neural circuits in regulation of
visceral function. J Pediatr Urol 13(2):177–182
Meena J, Mathew G, Hari P, Sinha A, Bagga A (2020)
Prevalence of bladder and bowel dysfunction in toilet-­
trained children with urinary tract infection and/or
primary vesicoureteral reflux: a systematic review and
meta-analysis. Front Pediatr. 8(84) Published 2020
Mar 31
Michael M, Hodson EM, Craig JC, Martin S, Moyer VA
(2003) Short versus standard duration oral antibiotic
therapy for acute urinary tract infection in children.
Cochrane Database Syst Rev (1):CD003966
Montini G, Tullus K, Hewitt I (2011) Febrile uri-
nary tract infections in children. N Engl J Med
365(3):239–250
Morris BJ, Bailis SA, Wiswell TE (2014) Circumcision
rates in the United States: rising or falling? What
effect might the new affirmative pediatric policy state-
ment have? Mayo Clin Proc 89(5):677–686
Pohl HG, Belman AB (2009) The “top-down” approach
to the evaluation of children with febrile urinary
tract infection. Adv Urol 2009:783409. https://doi.
org/10.1155/2009/783409
RIVUR Trial Investigators, Hoberman A, Greenfield
SP et  al (2014) Antimicrobial prophylaxis for chil-
dren with vesicoureteral reflux. N Engl J Med
370(25):2367–2376
T. de los Reyes and M. A. Koyle
Shaikh N, Hoberman A, Keren R et al (2016) Recurrent
urinary tract infections in children with bladder and
bowel dysfunction. Pediatrics 137(1):e20152982
Shaikh N, Morone NE, Bost JE, Farrell MH (2008)
Prevalence of urinary tract infection in childhood: a
meta-analysis. Pediatr Infect Dis J 27(4):302–308
Shaw KN, Gorelick M, McGowan KL, Yakscoe NM,
Schwartz JS (1998) Prevalence of urinary tract infec-
tion in febrile young children in the emergency depart-
ment. Pediatrics 102(2):e16
Shim YH, Lee JW, Lee SJ (2009) The risk factors of
recurrent urinary tract infection in infants with normal
urinary systems. Pediatr Nephrol 24(2):309–312
Singh-Grewal D, Macdessi J, Craig J (2005) Circumcision
for the prevention of urinary tract infection in boys: a
systematic review of randomised trials and observa-
tional studies. Arch Dis Child 90(8):853–858
Stein R, Dogan HS, Hoebeke P et al (2015) Urinary tract
infections in children: EAU/ESPU guidelines. Eur
Urol 67(3):546–558
Strohmeier Y, Hodson EM, Willis NS, Webster AC, Craig
JC (2014) Antibiotics for acute pyelonephritis in chil-
dren. Cochrane Database Syst Rev (7):CD003772.
Published 2014 Jul 28
Subcommittee on Urinary Tract Infection (2016)
Reaffirmation of AAP clinical practice guideline: the
diagnosis and management of the initial urinary tract
infection in febrile infants and young children 2-24
months of age. Pediatrics 138(6):e20163026
Tullus K, Jacobson SH, Katouli M, Brauner A (1991)
Relative importance of eight virulence characteristics of
pyelonephritogenic Escherichia coli strains assessed by
multivariate statistical analysis. J Urol 146(4):1153–1155
Wennerström M, Hansson S, Hedner T, Himmelmann
A, Jodal U (2000) Ambulatory blood pressure 16-26
years after the first urinary tract infection in childhood.
J Hypertens 18(4):485–491
Zorc JJ, Kiddoo DA, Shawn KN (2005) Diagnosis and
management of pediatric urinary tract infections. Clin
Microbiol Rev 18:417–422
 Imaging of the Paediatric
Urogenital Tract
89
Michael Riccabona

Abbreviations UT Urinary tract

(a)CDS (Amplitude-coded) colour Doppler
sonography
(a)PN (Acute) Pyelonephritis
CA Contrast agent
ce Contrast-enhanced
ce-US Contrast-enhanced ultrasound
ce-VUS Contrast-enhanced voiding
urosonography
CM Contrast media
CT Computed tomography
DMSA Dimercaptosuccinic acid
IVU Intravenous urography
KUB Kidney–ureter–bladder abdominal film
MAG3 Dynamic/diuretic renal scintigraphy
MRT Magnetic resonance tomography
MRU Magnetic resonance urography
MU Megaureter
PCN Percutaneous nephrostomy
PUJ(O) Pelvi-ureteral junction (obstruction)
PUV Posterior urethral valve
RNC Radionuclide cystography
Tbc Tuberculosis
Tu Tumour
UCA Ultrasound contrast agents
UG(T) Urogenital (tract)
US Ultrasound
M. Riccabona (*)
Department of Pediatric Radiology, University Clinic
of Radiology, Medical University Graz, Graz, Austria
e-mail: michael.riccabona@medunigraz.at
UTI Urinary tract infection
UVJ(O) Ureterovesical junction (obstruction)
VCUG Voiding cystourethrography
VUR Vesico-ureteral reflux
XPN Xanthogranulomatous pyelonephritis
89.1 Introduction
When discussing imaging of the paediatric uro-
genital (UG) system, the basic rationale of any
imaging should be reconsidered and remem-
bered: imaging should only be used if it has an
impact on diagnosis, treatment, further manage-
ment and/or patient outcome. Particularly in pae-
diatrics, the imaging modality should be selected
considering the specific paediatric needs and dis-
eases: imaging should try to avoid ionising radia-
tion whenever possible, the least invasive
procedure applicable for the individual query
should be selected (e.g. try to avoid catheterism if
possible, reduce examinations that need sedation
etc.) and any imaging that does not impact man-
agement should be avoided. Furthermore, the
recommended imaging method should be easily
accessible and available. Numerous recommen-
dations exist on how to image when and for what;
those from the ESPR abdominal imaging task
force are listed in the Further Reading section—
as a help to select the appropriate approach and

© Springer Nature Switzerland AG 2023 1227

P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_89
1228
then tailor the necessary imaging individually in
terms of a more personalised medicine. And one
more basic comment on imaging in children:
children are not small adults, therefore state-of-
the-art imaging of infants and children requires
dedicated techniques (equipment, transducer, fil-
tering ...), as well as age-adapted investigation
protocols (ultrasound, CT, MR etc.). This means
that everyone who deals with children needs to
have not only the adequate equipment available,
but also has undergone dedicated training in pae-
diatric imaging.
89.2 Imaging Methods
There are a number of imaging methods that can
be used and are applied for imaging the paediat-
ric UG-tract (UGT): ultrasound (US), abdominal
plain film, intravenous urography (IVU) (modi-
fied and adapted), voiding cystourethrography
(VCUG), as well as other forms of urethra- or
uretero-pyelography (antegrade or retrograde),
including the therapeutic percutaneous nephros-
tomy (PCN), scintigraphy, computed tomogra-
phy (CT) or magnetic resonance tomography
(MRT). The details of specific imaging, such as
scintigraphy, CT or MRT in children are not dis-
cussed here, but the chapter briefly outlines the
basic requirements for more commonly per-
formed procedures such as US, IVU or VCUG,
which, in some centres, are in the hands of paedi-
atric surgeons or paediatric nephro-/urologists.
Interventional radiology in the paediatric UGT is
discussed in the respective chapters. Note that the
role and potential of various imaging modalities
are constantly evolving, as well as the patient
treatment concepts and management strategies;
this naturally has implications on imaging algo-
rithms. Additionally, whoever performs paediat-
ric uroradiologic studies should have undergone
dedicated training in paediatric imaging tech-
niques to grant the optimal diagnosis at the low-
est achievable invasiveness by high-quality
imaging and child-adapted procedures.
Ultrasound (US) has emerged from an ini-
tially orienting imaging tool to a dedicated high-­
end examination method and a range of
M. Riccabona
age-adapted sector, vector, curved array and lin-
ear transducers are necessary to reliably perform
the task. Today, high-frequency, high-resolution
transducers and some modern US techniques are
applied routinely, such as (amplitude-coded)
colour Doppler Sonography [(a)CDS], motion
mode, harmonic and high-resolution imaging, as
well as image compounding. Other modern
methods, such as the three-dimensional US (e.g.
for conspicuously displaying the rendered dilated
collecting system, for calculating renal parenchy-
mal volume, for virtual cystoscopy or assessment
of particularly uterine anomalies) or US elastog-
raphy (e.g. for chronic kidney disease or trans-
plant queries) have not found a place in routine
paediatric UG US, whereas contrast-enhanced
(ce) US has become an increasingly appreciated
tool for many conditions, such as intraluminal
queries (VUR = ce-VUS, US-guided nephros-
tomy and drain placement, US-genitography) or
iv. applications (e.g. assessment of complicated
cysts or renal trauma) (Fig. 89.1). These applica-
tions may help to reliably establish a final diag-
nosis in many conditions, reducing the need for
other (irradiating) imaging in many conditions.
However, to properly explore this quickly grow-
ing, vast potential of US, some prerequisites need
to be considered:
• Investigations should be performed in physio-
logically hydrated patients with a sufficiently
filled bladder.
• Every US investigation includes the entire uri-
nary tract (UT) and possibly the genital tract
(GT), supplemented by an abdominal survey.
• For a thorough evaluation, a detailed study of
the renal parenchyma using linear transducers
from a lateral, ventral or dorsal approach is
mandatory, as well as pre- and post-void
assessment of the bladder, the retrovesical
space and the kidneys/the upper collecting
system.
• (a)CDS is very helpful in various conditions,
such as urinary tract infection (UTI).
• Kidney volume should be calculated and
assessed in relation to growth charts.
• Note that particularly initial studies should
always include a post-void assessment.
89  Imaging of the Paediatric Urogenital Tract
a b
Bl
PCD
e f
PCD
U
g h
Bl
Fig. 89.1  Applications of intracavitary contrast-enhanced
Ultrasound in paediatric urogenital radiology: some typi-
cal examples. (a) Echogenic appearance of the urinary
bladder ((Bl), axial section) filled with ultrasound contrast
agent. (b) Parasagittal oblique section depicting a contrast
filled ureter (U) behind the contrast filled urinary bladder
(Bl) on ce-VUS; note the wide-open bladder neck and the
pathological course of the transmural ureter. (c) Perineal
US during voiding during ce-VUS nicely demonstrating
the normal male urethra (arrow). Bl: urinary bladder. (d)
Coronal view of the kidney on ce-VUS; dual image display
(left side: contrast image, right side: fundamental US
image) Contrast depicted in the renal pelvis, without dila-
tation of the calices, in middegree VUR (III°); PCD: pelvi-
calyceal system. (e) High-grade reflux into a dilated
pelvicalyceal system (PCD) with a dilated and wide ureter
1229
c
Bl
U
Bl
Bl
*
at the ureteropelvic junction. (f) Coronal view of the kid-
ney on ce-VUS: Intrarenal reflux displayed in the right
upper corner image (arrows) seen as echogenic bubbles
beyond the caliceal border in the renal medulla. (g) Axial
pelvic section — dual image display (left side: contrast
image, right side: fundamental US image) during ce-VUS
showing a few contrast bubbles in a dilated ureter (arrow)
behind the bladder (Bl) — as such VUR is diagnosed
which was invisible on VCUG (as the dilution effect of the
radiopaque contrast did not create enough contrast to be
visualized on fluoroscopy). (h) ce-VUS with US genitog-
raphy, perineal view: during voiding after filling the uri-
nary bladder (Bl) contrast influx from the urethra (arrow)
into the vagina (arrowhead) through a urethrovaginal sinus
tract (Asterix) is depicted, in a baby girl with adrenogenital
syndrome
1230 M. Riccabona

In dilated or distended pelvi-caliceal systems
or UTs, formerly called “hydronephrosis”, now
replaced by the terms PCD (= pelvi-caliceal dis-
tention) or UTD (=UT dilatation), standardised
(mostly axial) measurements of caliceal dilata-
tion and the width of the ureter and renal pelvis
are mandatory; grading of PCD/UTD should be
performed using standardised classifications,
such as the ESPR or the new American sugges-
tions (Tables 89.1 and 89.2).

Table 89.1  Grading of pelvi-calyceal respectively urogenital tract dilatation. Pelvi-caliceal Distention (PCD) grading
by US (has replaced the old term “Hydronephrosis”) as used in Europe and suggested by the ESPR task force
Typ 0 Typ I Typ II Typ III Typ IV Typ V

PCD 0: no collecting system visible, usually considered normal

PCD I: just the renal pelvis visible with an axial diameter less than 5–7 mm, usually considered normal
PCD II: axial renal pelvis diameter less than 5/7–10 mm, some calyces with normal forniceal shape visible
PCD III: marked dilatation of the renal calyces and pelvis larger than 10 mm with reduced forniceal and papillar dif-
ferentiation without parenchymal narrowing, often seen in dilating VUR and obstructive uropathy
PCD IV: gross dilatation of the collecting system with narrowing of the parenchyma, in high grade and obstructive
uropathy
PCD V: in some places used additionally to communicate an extreme PCD/UTD with only a thin, membrane-­like
residual renal parenchyma. Differentiation from MCDK is commonly achievable, as in MCDK the residual dysplastic
parenchyma is usually positioned centrally, with surrounding and exophytic cysts of varying size

Adapted suggestion, based on Hofmann V. and on Fernbach SK et al. Ultrasound grading system of hydronephrosis:
introduction to the system used by the Society for Fetal Urology. Pediatr Radiol 1993, 478–480) (Riccabona et  al.,
Pediatr Radiol 2008, 38: 138–145, updated 2017; see also Riccabona M (ed) Paediatric urogenital Radiology 3rd edn,
Springer 2019; and Riccabona M, Ultrasound of the Urogenital tract in neonates, infants and children. In Riccabona M
(ed) Pediatric Ultrasound, requisites and applications, 2nd edn, Springer 2020
89  Imaging of the Paediatric Urogenital Tract 1231

Table 89.2  Grading of pelvi-calyceal respectively urogenital tract dilatation. Urinary tract dilatation (UTD) grading
issued by the American Societies (has replaced the old term “Hydronephrosis”) more commonly used in the USA—
introduced by a multi-society consensus statement (adapted suggestion, based on Nguyen HT et al., Multidisciplinary
consensus on the classification of prenatal and postnatal urinary tract dilation—UTD classification system. J Pediatr
Urol 2014) composed of pre- and postnatal presentations leading to a risk-based assessment guiding management and
respective imaging strategy
Prenatal presentation
Normal Abnormal
[ap pelvis 4–7 (<28 w)/7–10 (>28 w) mm] [ap pelvis >7 (<28 w)/>10 (>28 w) mm]
if no other pathology (normal parenchyma, other pathology (cysts or narrow/dysplastic
slim ureter, bladder normal, slim calices) + no parenchyma, abnormal ureter, bladder pathology,
oligohydramnios dilated/clubbed calices) + oligohydramnios

UTD A1 = low risk UTD A2/3 = increased/high risk

Postnatal presentation

No or some dilatation (>48 h) No or some dilatation (>48 h) No or some dilatation (>48 h)

[ap pelvis <15 mm], [ap pelvis >15 mm] [ap pelvis >15 mm]
if no other pathology
(normal parenchyma, slim Peripherily dilated calices Peripherily dilated calices
ureter, bladder normal, Normal parenchyma Parenchymla pathology
sometimes only centrally Abnormal ureter Abnormal ureter
visible calices with normal Bladder normal Bladder abnormal
shape)

UTD P1 = low risk UTD P2 = intermediate risk UTD P3 = high risk

Risk based management

UTD A1 UTD A2/3 UTD P1 UTD P2 UTD P3

Prenatally
1 additional US Additional US - - -
at-4–6 w
Postnatally
US at 2 d–1 mo US at 2 d–1 mo Follow-up Follow-up Follow-up
US at 6 w–6 mo depending on US 1–6 mo US 1–3 mo US at 1 mo
findings
VCUG at VCUG at VCUG
discretion of discretion of recommended
Clinician Clinician

Antibiotics at Antibiotics at Antibiotics

discretion of discretion of Recommended
clinician clinician

Others: Others: Functional scan Functional scan Functional scan

Risk adaption if Specialist not at discretion of at discretion of
indicated consultation recommended clinician clinician

Abbreviations: ap anterior-posterior (axial) section, d days, MCDK multicystic dysplastic kidney, mo months, PCD
pelvo-caliceal dilatation/distention, US ultrasound, UTD urinary tract dilatation, w weeks
1232
Assessment of vesico-ureteral reflux (VUR)
based on changing dilatation of the ureteropelvic
system is unreliable, only the use of US contrast
agents (UCA) instilled into the bladder (called ce-
voiding urosonography = ce-VUS) enables a reli-
able and adequate sonographic assessment of
VUR (see Fig.  89.1a–g). Include surrounding
body compartments (i.e. perform an orienting
overview (“sonoscope”) of the other pelvic struc-
tures) as well as of the liver, the pancreas, the
spleen, the adrenal glands and adjacent intestinal
parts, in order to not miss a potential disease that is
clinically mimicking UT pathology or that is part
of an underlying systemic or syndromal condition
(e.g. US of the neonatal spine in babies with neu-
rogenic bladder). Furthermore, it is recommended
to thoroughly evaluate the inner genitalia in female
neonates with severe UT malformations or single
kidneys, as there is a relatively high incidence of
embryologically determined associated genital
malformations. The diagnosis of these conditions
will be difficult after the 1st months of life due to
the regression of the uterus und ovaries, whereas a
reliable diagnosis can easily be achieved by US,
supplemented by sonographic genitography in the
neonatal period (potentially supplemented by fluo-
roscopy using the same catheters see Fig. 89.1h).
Furthermore, in certain conditions, such as syn-
dromal disease, adequate additional assessment of
other body areas (heart, brain, spinal canal) must
be initiated, with many of these queries being suf-
ficiently answered by US.
Voiding cystourethrography (VCUG): In
spite of US advances, VCUG still remains an
important investigation for VUR, particularly in
countries where UCA is not approved for paedi-
atric use or pre-operatively. Furthermore, VCUG
enables a detailed anatomic assessment of the
urethra and is superior to US in depicting inter-
mittently posing diverticula, in demonstrating
intrarenal reflux and by offering a panoramic
overview of the entire (refluxing) UT. However,
VCUG poses a significant radiation burden, par-
ticularly to girls—thus indications must be very
strict (and constantly reassessed, potentially
investigations can be replaced by ce-VUS) and a
proper fluoroscopy technique is essential:
M. Riccabona
• The radiopaque contrast media (CM) should
always be instilled into a (nearly) empty blad-
der, with CM instillation under physiologic
pressure to avoid overextension and non-­
physiological conditions. The way varies
depending on how the various centres, both
transurethral as well as suprapubic punctures,
are used (the latter particularly for psycholog-
ical and socio-cultural reasons, as well as in
cases where a catheterisation is impossible
such as in PUV or urethral trauma). Note that
a balloon of a urethral catheter must be
deflated to avoid incorrect findings.
• Cyclic studies are recommended in neonates and
infants to increase VUR detection rate—this is
more difficult with a suprapubic puncture.
• Adequate fluoroscopy units adapted for paedi-
atric needs are mandatory—performing
VCUG with blind-spot films or using a C-arm
device is inadequate.
• A VCUG should always include a fluoro-
scopic assessment during the early filling
phase in an anterior-posterior projection, an
oblique view of the distal ureteral region, a
lateral or oblique view of the urethra during
voiding, an assessment of the upper collecting
system during and after voiding, as well as a
post-void assessment for residual urine and of
potentially refluxed CM drainage dynamics.
Detailed guidelines on how to properly per-
form VCUG can be found in the paediatric
­radiology literature (e.g. Riccabona et al. 2008);
VCUG-based VUR grading has internationally
been standardised.
Plain film still has a role in assessment of UT
calculi, as well as in assessment of various drains,
catheters and devices; it, furthermore, is helpful
for the assessment of associated skeletal changes,
such as spina bifida. It should be focused on the
kidney, the ureter and the bladder (“KUB”) by
using shutters (not retrospective collimation in
digital devices!); an age-adapted technique (dose,
grids, resolution of film-foil combination or digi-
tal detectors) is essential.
Intravenous urography (IVU) was one of the
mainstays in uroradiology decades ago that allows
89  Imaging of the Paediatric Urogenital Tract
for a reliable assessment of the collecting system
anatomy and reveals information on renal function
and urinary drainage. However, today it has lost
most of its importance in the paediatric setting, as
US, scintigraphy and MR-Urography (MRU)
enable a superior assessment with less radiation in
most paediatric queries. Still, in areas with
restricted access to these modalities, for a focused
pre- or post-operative assessment, in some condi-
tions where maximal spatial resolution is indis-
pensable for early diagnosis (e.g. for medullary
sponge kidney), and in suspected urolithiasis that
cannot be sufficiently assessed using the other
means respectively or for planning lithotripsy, a
modified and adapted IVU may still be an option.
If IVU is performed, the number of films should
be restricted, and targeting (shutters—not retro-
spective grabbing or collimating on digital devices,
gonad protectors) as well as timing should be indi-
vidually adapted to the diagnostically relevant
moments and areas. X-ray dose, film-foil combi-
nation, digital detectors, grids and CM dose need
to be adapted according to the patient’s age and
necessary resolution. With this approach, a signifi-
cant reduction of the film number (usually 2–4
films are sufficient for diagnosis) and of the deliv-
ered radiation dose can be achieved.
Scintigraphy is usually performed by paediat-
ric nuclear medicine specialists according to the
standardised protocols issued by the paediatric
nuclear medicine groups and societies/associa-
tions. It uses tracers labelled with radioactive
material (usually Technetium 99m) that are spe-
cifically handled by the renal parenchyma and/or
secreted into the urine. The change in radioactiv-
ity over time is measured by a gamma camera,
for functional assessment.
Static renal scintigraphy uses DMSA (dimer-
captosuccinic acid). This tracer is specifically
taken up by the tubuli, and it allows for an exqui-
site assessment of the functioning renal paren-
chyma. It is used for split renal size assessment
and assessment of focal lesions, such as acute
pyelonephritis (aPN) or scars. Spatial resolution
is restricted and can be improved using SPECT;
however, due to an often-increased radiation bur-
1233
den, this latter technique is rather reluctantly
used in children.
Dynamic (diuretic) renography uses the glo-
merularly and tubularly handled MAG3
(Mercapto-acetyltriglycine) allowing for differ-
entiation of an arterial phase, a parenchymal
phase, an excretion phase and a drainage phase.
Standardised hydration is essential as well as
diuresis provocation using Furosemid to produce
reliable results; a bladder catheter should be
inserted in infants and particularly in patients
with (high degree) VUR also to avoid misdiagno-
sis by bladder filling induced “pseudo-­
obstruction”. It is mainly used for the assessment
of obstructive uropathy, such as pelvi-ureteral
junction obstruction (PUJ(O)) or megaureter
(MU).
Radionuclide cystography (RNC) uses either
the late phase of a dynamic renogram with the
bladder filled by the renally excreted MAG3
(only applicable in toilet-trained and cooperative
patients—without catheterisation), or the isotope
(generally Tc99m colloid) is installed directly into
the bladder via suprapubic puncture or a trans-
urethral catheter. Any activity increase in the
upper collecting system proofs the VUR.  Some
grading and functional assessment is achievable
but anatomic resolution is restricted; RNC, there-
fore, is used mostly only for screening purposes
and for follow-up examinations.
Interventional radiologic procedures are less
common in children—however, still PCN (usu-
ally US-guided, potentially complemented by
fluoroscopy or increasingly intra-cavitarily
applied UCA) is still performed (often by paedi-
atric urologists), and intravascular diagnostic and
therapeutic procedures may become necessary,
such as embolisation of an acute (tumour, trau-
matic) haemorrhage or depiction and treatment
of renal artery stenosis—the latter commonly
performed only by interventional (paediatric)
radiologists.
CT and MRI are usually performed by paedi-
atric radiologists and will not be covered in detail.
CT has a significant radiation burden and,
therefore, is very reluctantly used in children. If
1234
UT-CT is considered in neonates, infants or chil-
dren, age-adapted low-dose protocols with
adapted parameters, reconstruction algorithms,
as well as CM amount and timing are mandatory;
normal protocols used in adults are not accept-
able for routine paediatric use. The generally
accepted indications for CT in children’s UGT
are: severe abdominal trauma, tumour assess-
ment, assessment of renal abscesses (particularly
if MRI is not available), complex malformations
that involve the pelvic skeleton (if this informa-
tion is relevant for planning surgery, e.g. bladder
exstrophy), and assessment of complicated uroli-
thiasis and infections, such as renal tuberculosis
(Tbc) or xanthogranulomatous pyelonephritis
(XPN). Note that increasingly, most of these que-
ries are investigated by MRI if available.
MRI is becoming the ideal one-stop-shop
imaging modality, particularly for congenital
UGT anomalies, as MRU constitutes a compre-
hensive investigation without a radiation burden
that allows for both anatomic and (if diuretic
stress and CM are applied) functional assess-
ments. Its setbacks are the restricted availability,
the sedation needs particularly in infants and
young children, some spatial and temporal reso-
lution issues, as well as artefacts derived from
patient motion. Nevertheless, MRI has taken
over, not only as the imaging assessment of com-
plex UG anomalies, but also the imaging assess-
ment and staging of most UG tumours in
childhood; thus, increasingly replacing CT. And
finally, MRI is promoted for assessment of equiv-
ocal or complicated UTI (additionally applying
diffusion-weighted sequences), as well as assess-
ment of genital or complex anorectal malforma-
tions and anomalies.
89.2.1 Typical Imaging Findings
in Common Paediatric
Urological Conditions
One of the most common queries is pre- or neo-
natally detected UTD/PCD. The task of early
imaging is to depict those neonates and infants
who benefit from early treatment and appropriate
management with consequent monitoring to pre-
M. Riccabona
vent future harm to the kidney. Basically, a neo-
natal US study in the well-hydrated patient at the
end of the 1st week of life enables differentiation
of the patients who do not need any further imag-
ing, those patients who need follow-up or addi-
tional studies and patients who need urgent
treatment, such as for severe bilateral obstruction
or posterior urethral valve (PUV). The latter
might often need an earlier study (within the first
days of life), if indicated by prenatal US findings
and clinical presentation. The most common
diagnosis in this patient group is a “physiologi-
cal” dilatation without any need for further stud-
ies or follow-up (Fig.  89.2). The other relevant
group is those with (dilating) VUR and obstruc-
tive uropathy. The most important sonographic
signs (constituting most of the “extended US cri-
teria” that improve US diagnosis and differentia-
tion, Table  89.3) that allow differentiation of
obstructive versus non-obstructive dilatation are:
thickening of the renal pelvic wall and the con-
figuration of the calices, which remain normally
shaped in low-pressure pelvi-ectasia, whereas
they become distorted or rounded with more or
less parenchymal narrowing in patients with
increased intrapelvic pressure. Additionally,
bladder wall changes or dilatation of the ureter—
even if intermittent—may hint towards urine
transport and drainage problems, particularly
VUR.  Note that already neonatally renoparen-
chymal changes due to hypodysplasia, such as
increased echogenicity, reduced cortico-­
medullary differentiation or renal cysts may be
present. These findings usually indicate a rela-
tively restricted prognosis with high probability
of renal insufficiency that sometimes manifests
in infancy. Creative new sonographic approaches,
such as perineal US (Fig.  89.3) or ce-VUS/
US-urethrography can help in the further specifi-
cation of the condition, either by directly show-
ing a urethral valve or enabling a reliable
assessment of VUR. As these UCA-based meth-
ods are increasingly advocated for radiation
safety issues, the paediatric urologist should
make him-/herself familiar with the respective
imaging appearance.
In high-grade neonatal PCD, early US may
underestimate its severity as the physiologically
89  Imaging of the Paediatric Urogenital Tract 1235

a b

Fig. 89.2  US of a normal neonatal kidney. Longitudinal 14 MHz linear transducer: Note some echogenic papillae,
(a, no pelvi-calyceal distension) and axial (b, slight physi- with a physiologically pronounced cortico-medullary dif-
ologic distension of the renal pelvis) section with a ferentiation and echogenic cortex

Table 89.3  Extended ultrasound criteria

List of extended ultrasound criteria
These sonographic features may indicate UT pathology and might prompt further investigations (individual
personalised decision based on all information available):
 Urethra: valve-like shape of open bladder neck, suspected stenosis or valve on perineal US during voiding,
utriculus cysts or suspected duplications etc.
 Bladder: atypical shape and size, bladder wall thickening and trabeculation, diverticula, ostium pathology,
significant residual urine, persistent urachus
 Ureter: dilatation, tortuosity, ureteral wall thickening, ectopic orifice, duplex ureter
 Kidney: position (ectopic? malrotation? fusion such as Horseshoe kidney …), caliceal dilatation and shape (e.g.
PCD III or higher, clubbing, or obvious variation of dilatation during investigation/during and after voiding),
pelvic wall thickening (“urothelial sign”), abnormal/asymmetric kidney size, parenchymal abnormalities
(echogenicity and cortico-medullary differentiation, narrowing, cysts ...), vascular anomalies (regional vessel
rarefaction, accessory renal artery crossing area of pelvi-ureteric junction, abnormal course of renal vein etc.)

reduced renal function during the first days of life
will prevent the collecting system from reaching
its maximum dilatation. Therefore, a repeated
investigation after 2–4 (6) weeks is usually neces-
sary. The amount of dilatation does not necessar-
ily equal the degree of obstruction or renal
function; kidneys with significantly reduced func-
tion may exhibit less dilatation than systems that
function well. Therefore, US is not sufficient for
assessment and grading of obstruction. Earlier,
IVU was generally used for the assessment of
urine drainage dynamics; today, this has been
replaced by diuretic renography and functional
MRU. Note that these functional studies (i.e. scin-
tigraphy and MRU) should only be performed
after kidney maturation, which is at the earliest
after the 6th week, better the 3rd month of life.
Particularly in patients with a dilated ureter,
differentiation of obstructive uropathy from VUR
is mandatory. In these cases—particularly in neo-
nates and boys—ce-VUS or VCUG is recom-
mended for comprehensive assessment of VUR,
including the bladder and the urethra. If VUR is
detected, a base assessment of renal function by
DMSA scintigraphy should be performed, usu-
ally at approximately 3–6 months of age. VCUG
1236
Fig. 89.3  Perineal US during voiding in a neonate show-
ing the urine-filled and nicely distended male urethra
(arrow) and a cystiform urethral diverticulum (arrow-
head). Bl: bladder, A: anal canal, S:symphysis. Note that
urine in itself may act as a sufficient contrast medium on
US so that not always UCA are needed
is mandatory before surgery, whereas VUR fol-
low-­up or VUR screening (if indicated at all) can
be performed by RNC or ce-VUS.
In obstructive uropathy, differentiation of the
level of obstruction is important. Besides the rare
but serious infravesical obstruction secondary to
a tumour or a PUV, the obstruction can be either
at the level of the ureterovesical junction (UVJ)
or pelvi-ureteric junction (PUJ). In UVJ obstruc-
tion (UVJO), a “megaureter” is present that can
be usually and finely visualised by US, as well as
a potential ureterocele and an often associated
duplex system (Fig. 89.4a). US can additionally
assess and document (by motion-mode or using
cine loop clips) ureteral peristalsis, particularly
useful for follow-up (Fig 89.4b). PUJO is sono-
graphically characterised by significant dilatation
of the renal pelvi-caliceal system (PCD III–V),
with more or less parenchymal narrowing with-
out a visible ureter (Fig. 89.5a, b), unless a com-
bined stenosis or a combination with dilating
VUR is present. US examinations should always
measure the maximum axial distension of the
renal pelvis and the calices, as well as the maxi-
mal narrowing of the parenchyma, and assess
renal vascularity and perfusion. CDS may depict
additional renal arteries compromising the PUJ,
and Doppler trace analysis may exhibit asym-
metrically elevated resistance indexes in severe
acute obstruction, particularly under diuretic
stimulation with Furosemid (= diuresis sonogra-
M. Riccabona
phy). Furthermore, echoes and sedimentations
may be present within a dilated system and may
hint towards intraluminal infection (but also other
phenomena may cause those echoes—therefore,
US is not proof of such an infection, but can also
not “rule out” an UTI!). This condition is rela-
tively frequent in MU patients under antibiotic
prophylaxis and may be clinically silent; urine
samples may produce equivocal results, as a
greater amount of urine will come from the nor-
mal functioning non-­ affected contra-lateral
healthy system.
Once the initial diagnosis is established, man-
agement varies: patients with non-obstructive
dilatation or megacalycosis will usually be fol-
lowed by repeated US examinations. Patients
with obstructive uropathy or deterioration of mild
(“non-obstructive”) dilatation need additional
functional imaging, particularly for deciding on
surgery and for preoperative anatomic assess-
ment. US offers a non-invasive method for fre-
quent follow-ups, but remains restricted in terms
of functional assessment—therefore, diuretic
scintigraphy, at present, remains the standard
investigation for grading and follow-up of
obstruction; usually, the decision for surgery is
based on these findings (Fig. 89.5c). Increasingly,
functional diuretic dynamic MRU is used for a
combined approach to assess anatomy and func-
tion (Fig. 89.5d–h).
Another common paediatric condition is uri-
nary tract infection (UTI). Today, imaging
increasingly focuses on differentiation of upper
versus lower UTI, which clinically—particularly
in infants—can be difficult. Potentially associ-
ated UT malformations need to be evaluated, but
often these conditions have already been diag-
nosed by various screening programs. The
importance of bladder function disturbances is
increasingly recognised; therefore—particularly
in older girls—bladder function assessment con-
stitutes an essential part of the work-up after
UTI.
Again, US forms the mainstay of initial imaging
allowing for assessment of UT anomalies, and
using a supplementing aCDS enables a quite reli-
able assessment of renal involvement in UTI. Typical
US criteria for (acute) pyelonephritis (aPN) are:
89  Imaging of the Paediatric Urogenital Tract 1237

a b

Fig. 89.4  US in megaureter. (a) US image of an uretero-
coele (using an 8 MHz curved linear array and harmonic
imaging) protruding into the bladder at the ostium, with
its corresponding distal megaureter (parasagittal section,
slightly oblique). (b) Cross-section of a slightly dilated
distal left ureter behind the bladder in an infant using an
8 MHz curved linear array and harmonic imaging, with a
motion-mode trace documenting lack of ureteral
peristalsis

a b c

d e f g h

Fig. 89.5  Imaging in obstructive uropathy. (a, b) Typical
US image of a pelvi-ureteric junction obstruction with
gross pelvic dilatation, dilated calices (+ + 1) and paren-
chymal narrowing (+ + 2), in longitudinal (a) and axial (b)
section using an 8 MHz curved linear array with harmonic
imaging in an infant. (c) MAG3 dynamic diuretic renogra-
phy showing the asymmetrically deteriorated drainage of
the affected kidney (red line) in a decompensated pelvi-
ureteric junction obstruction. (d–h) MRU in pelvi-ureteric
junction obstruction: the T2-weighted “water image”
shows the grossly dilated pelvi-calyceal system with mal-
rotation on the left side (d). The serial T1 weighted gado-
linium-enhanced images show asymmetric uptake (e),
asymmetric excretion (f), asymmetric diuretic response (f,
3d-reconstructed image) and delayed excretion of contrast
urine into the dilated left collecting system without suffi-
cient urinary drainage (h, 3d-reconstruction)
1238
a b
Fig. 89.6  Imaging in acute UTI. (a) Cross-section with
an 8-MHz curved array transducer in a child with febrile
UTI demonstrating an area of increased parenchymal
echogenicity and reduced cortico-medullary differentia-
tion (+ +), consistent with focal renal involvement. (b)
Amplitude-coded colour Doppler sonography (= power
• Enlarged kidney
• Focal or diffuse parenchymal changes
• Increased echogenicity of the perirenal sinus
• Thickening of the pelvic-, ureteral- and blad-
der wall (Fig. 89.6a)
There may be (but must not be) echoes within
the urine. Furthermore, focal or diffuse, particu-
larly asymmetric perfusion or vascularity distur-
bances caused by inflammatory oedema, vascular
compression or necrosis, as well as scars can be
reliably visualised by aCDS (Fig.  89.6b). In
patients with ambiguous, equivocal or inconsis-
tent findings, as well as in situations where aCDS
is not available or of suboptimal quality, static
DMSA scintigraphy or increasingly MRI using
diffusion-weighted sequences are considered the
gold standard for assessment of suspected aPN
(however, only indicated if there is a therapeutic
implication) (Fig. 89.6c). Increasingly, MRI (and
in children, much rarely CT) are used for evalua-
tion of complications, such as abscess formation,
XPN, renal Tbc, as well as for differential diag-
nosis in sonographically unclear lesions such as
complicated, infected cysts or tumours. The latter
can be also better addressed by i.v. ce-US. VUR
assessment is considered indicated in all infants
and all patients with renal involvement or renal
scarring after the infection—best at approximal 4
weeks after the onset of an upper UTI.
Haematuria is a relatively rare event in child-
hood. There are a number of causes, most of them
M. Riccabona
c
Doppler) in the same child depicts a focal perfusion defect
at the site of the parenchymal abnormality, confirming
renal segmental involvement in acute pyelonephritis. (c)
Static renal DMSA scintigraphy in UTI, posterior acquisi-
tion, showing a photopenic defect in the upper pole (aster-
isk) in a different child with acute pyelonephritis
constituting a nephrologic problem, such as glo-
merulonephritis or UTI including haemorrhagic
cystitis. However, other severe UT conditions
may also manifest with macroscopic haematuria
such as haemolytic uraemic syndrome, renal vein
thrombosis, or bladder and kidney tumours.
Furthermore, conditions such as stress haematu-
ria, retro-aortal left renal vein with nutcracker
syndrome, idiopathic familiar microscopic hae-
maturia, and haematuria associated with reflux-
ing or even obstructing uropathy may also cause
(microscopic) haematuria; renal infarction is an
extremely rare event in children. In all these con-
ditions, a comprehensive abdominal US includ-
ing (a)CDS is indicated as the first imaging study
to supplement urine (erythrocyte) microscopy
and laboratory findings and often helps in the dif-
ferential diagnosis.
If (macroscopic) haematuria is associated
with flank pain or dysuria or even abdominal col-
ics, urolithiasis must be considered. US is the
first imaging study allowing assessment of the
collecting system and the renal parenchyma that
may exhibit signs of nephrocalcinosis and other
underlying nephropathies, or show different enti-
ties that need to be considered for differential
diagnosis, for example, bleeding renal or bladder
tumours, such as large angiomyolipoma or rhab-
domyosarcoma. Particularly, for detection of dis-
tal ureteral calculi a sufficiently filled bladder
and a detailed study of the distal ureters are man-
datory. CDS may be helpful by demonstrating the
89  Imaging of the Paediatric Urogenital Tract
a b
Fig. 89.7  Imaging in haematuria. (a) Colour Doppler
sonography demonstrates the asymmetric urine inflow
into the urinary bladder from the right ostium, and the
twinkling colour signals at the area of the left ostium/left
transmural ureter deriving from an ostial calculus (arrow).
(b) Longitudinal section of the distal ureter (curved array,
ureteric bladder inflow, enabling a differentiation
of asymmetric urine inflow or complete obstruc-
tion (Fig. 89.7a, b). Furthermore, the “twinkling
sign” may improve the detection of even small
calculi in poorly distended distal ureters
(Fig. 89.7c). Note that in the acutely obstructed
kidney, US usually exhibits only little dilatation;
the parenchyma is echogenic and the kidney
itself appears enlarged and swollen. Doppler flow
profiles may demonstrate an asymmetrically ele-
vated resistance index. The calculi themselves
are echogenic, though blood clots and mixed
concrements with low calcium content may also
appear less echogenic and produce less shadow-
ing, whereas intraluminal fungus may appear
similar to a calculus (i.e. very echogenic, with
some dorsal shadowing). In cases with ureteral
obstruction, the PUJ is often wide open, the prox-
imal ureter is dilated and thus can be followed to
the level of the obstruction (Fig.  89.8a–c).
Sometimes, differentiation of papillary or wall
calcifications from intraluminal concrements
may be difficult—in these cases, positioning
manoeuvres that demonstrate floating of concre-
ments may be helpful. Note that other sedimenta-
tions in the papillae or the distal tubules may also
cause similar appearances and twinkling artefacts
such as in nephrocalcinosis or the physiologic
hyperechoic medullae and papillae of neonates.
1239
c
harmonic imaging, 8 MHz) in an infant with haematuria
and hypercalcuria demonstrates a distal ureteral calculus
(+ +), with dorsal shadowing (arrow). (c) Same child, at
maximum of peristaltic wave, with better dilated ureter,
where the calculus exhibits twinkling signals on CDS (=
“twinkling sign”) (arrow)
In many cases, US (possibly supplemented by
a KUB film) is sufficient for diagnosis
(Fig.  89.8d). In some instances, particularly for
planning lithotripsy, an adapted IVU may become
necessary, for example for differentiation of a
duplex system or assessment of multiple ureteral
calculi (Fig.  89.8e, f). Low dose stone CT has
become the major imaging method in adults with
urolithiasis, but, at present, is still used reluc-
tantly in children due to its radiation burden—it,
however, can be used as a problem-solving tool
using a dedicated unenhanced low dose paediat-
ric age- and size-/weight-adapted CT-protocol.
Otherwise, only some rare conditions, such as
prolonged complicated stone disease with infec-
tion, XPN or renal tumours with calcifications, as
well as gross or complicated nephrolithiasis in an
underlying problem as well as acute or traumatic
haemorrhage, contrast-enhanced spiral CT may
become indicated (Fig.  89.9a–c). At present,
MRI plays no role in imaging urolithiasis.
Kidney and bladder tumours are rare but
important entities that only manifest clinically in
part. They are often detected incidentally when
doing an US for various unspecific symptoms.
such as abdominal pain, haematuria, increased
blood pressure or follow-up in various syndromes
that are associated with a higher risk for renal
tumours. The task of imaging is to detect and
1240
a b
d e
Fig. 89.8  Imaging in infants and children with urolithia-
sis. (a–c) US images pelvi-calyceal dilatation (a), an
echogenic pelvic wall with indirect signs for a duplex sys-
tem such as a central “parenchymal bridge” (b) and a cal-
culus in the mid-ureter ( + + ) that exhibits the twinkling
confirm the tumour, to try to give some informa-
tion on the tumour entity, to properly stage the
tumour and search for metastasis, to give the pre-
operatively necessary anatomic-topographic
information and eventually, to monitor the
patients for follow-up during and after treatment.
US usually is the initial investigation, and it
detects or confirms the tumour. MRI, at present,
is propagated for complete anatomic assessment
M. Riccabona
f
sign on CDS. (c–f) KUB film (d) and adapted IVU using
just two films for assessment of anatomy and drainage in
the same girl as a–c with ureteral urolithiasis and sus-
pected duplex system for planning shock wave lithotripsy
(e = 15 min after contrast, f = 60 min after i.v. contrast)
as well as characterisation and potentially staging
of paediatric UG tumours. CT may become nec-
essary for a thorough work-up, particularly con-
cerning lung metastases or suspected
calcifications (differential diagnoses versus, e.g.
neuroblastoma), as well as in situations where
(paediatric) MRI is not available.
In children, every renal cyst or renal polycystic
disease should prompt a thorough family assess-
89  Imaging of the Paediatric Urogenital Tract
a b
Fig. 89.9  Low dose unenhanced spiral multi-slice CT
(adapted paediatric protocol). (a–c) Multi-slice CT (a)
with sagittal reconstruction (b) and volume rendering (c)
in an infant with a large concrement that filled the entire
ment and at least one follow-up examination.
Basically, US is the primary investigation of
choice in renal cysts, which are usually diagnosed
and followed sonographically. Rarely, for exam-
ple, for a complete diagnostic work-up in suspi-
cion of cystic tumours, a range of modalities may
become necessary, including plain film, CT or
MRI (e.g. also for DD against tertiary calices or
caliceal diverticula) and scintigraphy, depending
on the underlying setting, the clinical suspicion
and method availability. Note that microcysts may
remain invisible to imaging, just altering the (US)
appearance of the renal parenchyma. Every com-
plicated cyst or every undefined cystic disease
(particularly if cyst size or number increases over
time) must be investigated by additional sectional
imaging, such as CM-enhanced spiral CT or
MRU; in complicated cysts i.v. ce-US may
improve differential diagnosis by, for example,
detecting enhancement in nodular areas or within
thickened septae, then confirming the suspicion of
a tumorous condition. International consensus
statements have been issued to streamline and rec-
ommend imaging algorithms in cystic kidney dis-
ease based on the new insights into pathophysiology
(e.g. model of ciliopathies, uromodelin pathology)
both in the initial diagnosis (that very much relies
on genetics and family history) and for follow-up,
also respecting new treatment options with the
respective imaging needs (e.g. MRI) for research
purposes or to study the treatment effects.
1241
c
collecting system, performed for therapy planning (i.e.
additional parenchymal calcifications that would obviate
lithotripsy or hinder percutaneous lithoapraxy)
Abdominal and urogenital trauma exists in
children, too. Depending on the severity of
trauma, as well as the trauma mechanism and
impact, different imaging approaches are advo-
cated. In mild trauma, as well as after renal
biopsy, a comprehensive US study may suf-
fice—with a mandatory follow-up after 12–24 h
(Fig. 89.10a). In severe or multiple traumas, the
role of US is restricted to the 1-min “FAST”
examination in the emergency room used to
check for free fluid; all other imaging is then
achieved by contrast-enhanced multi-detector
spiral CT. This approach grants a quick and reli-
able imaging of all necessary aspects (not only
the kidney and ureter, but also other abdominal
parenchymal organs, the spine, the major ves-
sels and thorax etc.) as necessary for proper
patient management and decision making, par-
ticularly as a conservative approach is increas-
ingly advocated. It also allows for detailed
grading of the injury. In cases with suspected
bladder injuries or ureteral as well as urethral
tears, dedicated additional imaging should be
performed, such as fluoroscopic- or
CT-cystography (complete filling of bladder
essential) and retrograde or antegrade urethra-
and ureterography. But note that again, only
age-/size-adapted paediatric CT protocols must
be used and have to be established beforehand—
as there is little time to adapt and optimise CT
protocols in emergency situations.
1242
a b
d e
Fig. 89.10  Post-operative imaging. (a, b) Longitudinal
US image of a perirenal/subcapsular haematoma at the
lower pole of the left kidney in a patient after renal biopsy,
much clearly visible using aCDS ( + + ) than just on plain
grey-­scale US (↔). (c, d) US after macroblast® injection
for cystoscopic VUR treatment: note the echogenic
deposit ( + + ) at the left transmural distal ureter/ostium
(c), with slight ureteral dilatation (d). (e–g) Adapted IVU
Preoperative imaging is outlined by the diag-
nostic imaging work-up of the various conditions
discussed above. In order to enable an efficient
use of imaging, a profound communication
should be established between all physicians and
specialties involved. For example, a PUJO diag-
nosed by US.  Diuretic scintigraphy has shown
decompensated obstruction with asymmetrically
reduced renal function of the affected kidney. If
any additional imaging is needed, the important
questions to answer are:
• Is there a duplex system?
• Is there VUR?
• Is there any vascular anomaly that may be
associated with the UPJO or might pose a
peri-operative risk?
M. Riccabona
c
f g
after ureteral re-­implantation and closure of a cutaneous
ureterostoma in an infant with obstructive megaureter
who sonographically exhibited gross dilatation with
asymmetrically elevated resistance indices post-opera-
tively. A series of three films demonstrates the partial
obstruction after 20 min (e) and after Furosemide applica-
tion (f), but sufficient clearance as response to gravitation
after upright positioning (g) obviating intervention
In this situation, preoperative imaging con-
sists of VCUG, MRU with an included
MR-Angiography sequence for the delineation of
the exact anatomy of the collecting system,
assessment of a potential duplex system and renal
vascular supply; IVU is not suitable to answer all
these questions, and ce-VUS may miss short-­
lasting, low-grade VUR into a non-dilated ureter
(however, probably this condition is of less
importance).
Usually, US is sufficient for post-operative
and post-interventional monitoring (Fig. 89.10a–
d). However, more intense post-operative imag-
ing may become necessary in cases with a
complicated post-operative course, as US may be
restricted in reliably answering all management
relevant function-related questions. Therefore,
89  Imaging of the Paediatric Urogenital Tract
antegrade pyelography (e.g. after pyeloplasty,
using the peri-operatively placed drain) may be
helpful; some centres even regularly perform an
antegrade pyelogram as part of the assessment
before drain removal (this can also be achieved
using UCA and US). For questions concerning
drain position and function, UCA may be applied
intraluminally to reduce the need for fluoroscopy
and can depict drain malposition with extravasa-
tion or assess ureteric patency. If no drain is in
place (such as in patients after anti-reflux proce-
dures), an adapted IVU (potentially only consist-
ing of two delayed films) or scintigraphy may
answer the question (Fig.  89.10e–g). Seldom
may other imaging, such as MRI or VCUG,
become necessary in the post-operative setting,
except for rare cases after tumour or trauma sur-
gery. Note that the value of post-operative imag-
ing is restricted; for example, pre-existing
chronic dilatation will not quickly resolve and
may physiologically even increase temporarily
post-operatively—thus, dilatation cannot be
used to assess re-obstruction. Therefore, don’t
overstress your imaging department with too
many unnecessary US exams just done for
“reassurance” without real clinical necessity, as
these results may be even misleading.
89.3 Imaging Algorithms
In order to standardise the imaging procedures,
to allow for comparison during follow-up, to
enable a reliable work-up even after working
hours and to grant some basic quality at reason-
able costs, imaging algorithms and recommenda-
tions are have become important; these should be
applied for imaging work-up of the various con-
ditions discussed above. They try to enable an
effective use of the various modalities without
increasing the risk of missing important condi-
tions that would need urgent treatment. The
European Society of Uroradiology, as well as
The European Society of Paediatric Radiology,
are engaged in developing and have issued
numerous specific paediatric imaging recom-
mendations for common queries, such as, for
1243
work-up of neonatal UTD/PCD, obstructive
uropathy in childhood, or imaging in
UTI. Furthermore, these societies have been and
are issuing procedural guidelines that define a
basic minimum standard for the various modali-
ties used in paediatric UG radiology. These open
access recommendations (mostly published in
Pediatr Radiol) try to take specific paediatric
considerations and growth variations into account
(as the probability of various diseases as well as
management options may differ depending on
sex and age), and address radiation protection
issues, as children are far more sensitive to ionis-
ing radiation than adults and need utmost radia-
tion protection.
For example, the imaging algorithm in UTI in
childhood shows slight differences between girls
and boys (VCUG versus ce-VUS) and for infants
versus older patients, where bladder function
studies become very important, while the need
for VUR assessment constantly decreases
(Table 89.4). The imaging algorithm for neona-
tal PCD is strategised according to the amount
of dilatation on prenatal US as well as the find-
ings on the first postnatal examination
(Table 89.5).
Note that all imaging recommendations and
algorithms constitute a basic general minimum
consensus; individual adaptation to the specific
patient’s condition (in terms of a more “person-
alised” approach) as well as adjustment of the
algorithms according to local facilities and avail-
ability and expertise remain essential.
Furthermore, these recommendations will need
constant updating because not only changes in
management strategies and advances in imaging
techniques will affect the recommendations and
induce potential changes, but also new insights
into pathophysiology and natural history of cer-
tain conditions may lead to altered imaging
requirement. As a general message—not all that
can be imaged needs to be imaged! If there is no
therapeutic impact, or some studies may even
carry the risk of misinterpretation, one should
abstain from studies that could possibly create
confusion and the need for further (follow-up)
studies.
1244 M. Riccabona

Table 89.4  Flowchart “Imaging algorithm in UTI in infants and children”

UTI (critria & diagnosis = clincal & laboratory: catheter urine sample /culture and blood count) *1

US incl. (power) Doppler Pyonephrosis nephrostomy

Normal Normal or equivocal*2 (Pyelo)nephritis*2

Clinically cystitis - e.g., poor quality, DDx … - aPN, upper UTI, scar …
- poor Doppler …
Clinically upper UTI

STOP *2
DMSA or MRI (*2) acutely
possibly follow-up
- if therapeutically or
diagnostically necessary
Follow-up US
- timing depends on clinical course
and initial findings
Normal - assess scars & renal growth
VUR evaluation
- always in neonates / infants
- possibly in age< 5 years
-in recurrent UTI & scaring /RNP
- ce-VUS particularly in girls &
* 1 Urine: lecucyturia, posoiutve notrite, positive caulture, urine calictonin …
Blood : leucocytosis, C-reactive protein,
follow-up
NOTE: reliable clinical diagnosis essential = most important entry criteria for imaging - VCUG (boys, pre-operatively)
US cannot relaible diagnose or rule out UTI
- RNC follow-up
* 2 for DDx and complications (e.g., Tu or cysts versus abscess) Late DMSA or MRI
for other complicated situations such as XPN, Tbc … or for
infected stone, inflammatory pseudotumor etc:MRI (or CT)
- after 4-6 (-12) months
- for scars
CT Indiactions: complicated stone disease, MRI not available
Bladder function studies
- in older, toilet trained patients

Adapted from Riccabona et al., Pediatr Radiol 2008, 38: 138–145

Suggested imaging algorithm in children with UTI
aPN acute pyelonephritis, DMSA static renal scintigraphy, ee-US echo-enhanced uro-sonography, RNC radionuclide
cystography, Tbc tuberculosis, Tu tumour, US ultrasound, UTI urinary tract infection, VCUG voiding cystourethrogra-
phy, VUR vesico-ureteral reflux, XPN xanthogranulomatous pyelonephritis
89  Imaging of the Paediatric Urogenital Tract 1245

Table 89.5  Flowchart “Imaging in neonates with antenatal diagnosis of PCD”

Prenatal US

Normal no US at all?
(grade 0/I, plevis < 5/8 mm)

mild dilatation moderate dilatation severe dilatation

(grade II, plevis 5-10 mm) grade III, pelvis 8/10 mm (grade IV: bilateral, dysplastic
unilateral grade IV + no other changes) parenchyma, supected PUV)

suffiicent hydration + sufficient bladder filling, use extended US criteria

Delayed US early US Urgent US

(day 7–14) (day 3–10) (day 1-3) + ce-VUS / VCUG
If suspicion of VUR + VCUG / ce-VUS

PCD I-II PCD III-IV Early DMSA / MRU?

No other pathology other pathology US-guided intervention?
complex malformation …

stop VCUG / ce-VUS? MRU? If signs of obstruction

potentially US follow-up
after some months
Scintigraphy?
Depends on individual findings
e.g. US-follow-up intervalls …
+ DMSA in VUR, …
or increasing dilatation etc …
+ MAG3 scintigraphy
+ (functional) ce-MRU
(+ adapted paediatric IVU)*
BUT wait for mature renal
function (> 8-12 weeks)

Adapted from Riccabona M, Fotter R, 2004, Reorientation and future trends in paediatric uroradiology: Minutes of a
symposium. Pediatr Radiol 34, 295–301
Suggested imaging algorithm in neonates with prenatally diagnosed PCD:
DMSA static renal scintigraphy, ce-VUS contrast-enhanced voiding uro-­sonography, IVU intravenous urography, MAG3
dynamic renography, MRU MR-urography, PCD pelvi-calyceal distension, PUV posterior urethral valve, US ultra-
sound, UT urinary tract, VCUG voiding cystourethrography, VUR vesico-­ureteral reflux
*Adapted paediatric IVU: only if no diuretic functional scintigraphy/ce-MRU available
1246
89.4 Conclusion
Many paediatric UG(T) conditions and queries
need dedicated and tailored expert imaging, best
provided by a paediatric radiology unit. Due to
the different conditions one encounters in child-
hood imaging, algorithms differ from adults, and
the imaging needs and protocols have to be
adapted individually according to the patient age,
gender, query and therapeutic consequence.
Usually, imaging starts with a dedicated US,
complemented by either a VCUG or MRI,
depending on the underlying query. Existing
imaging and procedural recommendations help
to optimally tailor and perform imaging in paedi-
atric uroradiology. And finally—also respecting
economic aspects and capacities, as well as local
variances and availabilities—imaging has always
to be questioned in terms of efficacy: studies that
do not serve the patient and impact on manage-
ment or prognosis and only serve for falsely
­pacifying referring physicians or parents should
be avoided.
Further Reading
Arthurs O, Easty M, Riccabona M (2020) Imaging of the
kidneys, urinary tract and pelvis in children. In: Adam
A, Dixon AK, Gillard JH, Schaefer-Prokop CM (eds)
Grainger and Allison’s diagnostic radiology, Volume
II, Section G pediatric imaging. Elsevier, New York,
pp 1846–1885
Fotter R (2001) Pediatric uroradiology. Springer, Berlin
Gordon I, Riccabona M (2003) Investigating the newborn
kidney update on imaging techniques. Semin Neonatol
8:269–278
Kao SC (2005) The urinary tract. In: Carty H, Brunelle F,
Stringer D, Kao SC (eds) Imaging children, 2nd edn.
Elsevier, New York, pp 537–882
Ključevšek D, Ključevšek T (2020) Efficacy of contrast-­
enhanced percutaneous nephrosonography to evaluate
urinary tract patency in children. J Clin Ultrasound
48:410–415. (Online June 16th)
Ključevšek D, Riccabona M, Ording Müller LS et  al
(2020) Intracavitary contrast-enhanced ultrasonogra-
phy in children: review with procedural recommen-
dations and clinical applications from the European
Society of Paediatric Radiology Abdominal Imaging
Task Force. Pediatr Radiol 50:596–606. https://doi.
org/10.1007/s00247-­019-­04611-­1
Nguyen HT, Benson CB, Bromley B et  al (2014)
Multidisciplinary consensus on the classification of
M. Riccabona
prenatal and postnatal urinary tract dilation (UTD
classification system). J Pediatr Urol 10:982–998
Puri P, Höllwarth M (2006) Pediatric surgery. Springer,
Berlin
Riccabona M (ed) (2002) Paediatric uroradiology. Eur J
Radiol 43 (special issue)
Riccabona M (2011) The pediatric kidney. In: Quaia E
(ed) Radiological imaging of the kidney. Springer,
Berlin, pp 675–714. ISBN 978-3-54087596-3
Riccabona M (ed) (2019) Pediatric urogenital radiology,
3rd edn. Springer, New-York
Riccabona M, Fotter R (2006) Radiographic studies in
children with kidney disorders: What to do and when.
In: Hogg R (ed) Kidney disorders in children and ado-
lescents. Taylor & Francis, Birmingham, pp 15–34
Riccabona M, Avni FE, Blickman JG et al (2008) Imaging
recommendations in paediatric uroradiology: Minutes
of the ESPR Workgroup session on urinary tract infec-
tion, fetal hydronephrosis, urinary tract ultrasonogra-
phy and voiding cysto-­urethrography ESPR-meeting,
Barcelona/Spain, June 2007. ESUR Paediatric
Guideline Subcommittee and ESPR Paediatric
Uroradiology Work Group. Pediatr Radiol 38:138–145
Riccabona M, Avni FE, Blickman JG et al (2009) Imaging
recommendations in paediatric uroradiology, part II:
urolithiasis and haematuria in children, paediatric
obstructive uropathy, and postnatal work-up of foe-
tally diagnosed high grade hydronephrosis. Minutes
of a mini-symposium at the ESPR annual meeting,
Edinburg, June. Pediatr Radiol 39:891–898
Riccabona M, Avni FE, Dacher JN et  al (2010) ESPR
Uroradiology Task Force and ESUR Paediatric
Working Group: Imaging and procedural recommen-
dations in paediatric uroradiology, Part III. Minutes of
the ESPR Uroradiology Task Force mini-symposium
on intravenous urography, uro-CT and MR-urography
in childhood. Pediatr Radiol 40:1315–1320
Riccabona M, Avni F, Dacher JN et  al (2011) ESPR
Uroradiology Task Force and ESUR Paediatric
Working Group: imaging recommendations in pae-
diatric uroradiology, part IV Minutes of the ESPR
Uroradiology Task Force mini-symposium on imaging
in childhood renal hypertension and imaging of renal
trauma in children. Pediatr Radiol 41:939–944
Riccabona M, Avni F, Damasio B et  al (2012) ESPR
Uroradiology Task Force and ESUR Paediatric
Working Group—Imaging recommendations in pae-
diatric uroradiology, part V: childhood cystic kidney
disease, childhood renal transplantation, and contrast-­
enhanced ultrasound in children. Pediatr Radiol
42:1275–1283
Riccabona M, Lobo ML, Willi U et  al (2014a) ESPR
Uroradiology Task Force and ESUR Paediatric
Working Group—imaging recommendations in paedi-
atric uroradiology, part VI: childhood renal biopsy and
imaging of neonatal and infant genital tract. Pediatr
Radiol 44:496–502
Riccabona M, Vivier HP, Ntoulj A et  al (2014b) ESPR
Uroradiology Task Force—Imaging recommen-
dations in paediatric uroradiology—part VII:
89  Imaging of the Paediatric Urogenital Tract
standardized terminology, impact of existing rec-
ommendations, and update on contrast-enhanced
ultrasound of the paediatric urogenital tract. Pediatr
Radiol 44:1478–1484
Riccabona M, Darge K, Lobo ML et  al (2015) ESPR
Uroradiology Task Force—imaging recommenda-
tions in paediatric uroradiology—part VIII: retro-
grade urethrography, imaging in disorders of sexual
development, and imaging in childhood testicular
torsion. Report on the mini-symposium at the ESPR
1247
meeting in Amsterdam, June 2014. Pediatr Radiol
45:2023–2028
Riccabona M, Lobo ML, Ording-Muller LS et al (2017)
ESPR Abdominal (GU and GI) Imaging Task Force—
imaging recommendations in paediatric uroradiology,
part IX: imaging in anorectal and cloacal malforma-
tion, imaging in childhood ovarian torsion, and efforts
in standardising pediatric uroradiology terminology.
Report on the mini-symposium at the ESPR meeting
in Graz, June 2015. Pediatr Radiol 47:1369–1380
 Management of Antenatal
Hydronephrosis
Jack S. Elder
90.1 Introduction
An abnormality involving the genitourinary tract
is detected in 1–2% of pregnancies, depending on
the sonographic criteria. The goal of manage-
ment is to recognize and treat congenital anoma-
lies that may adversely affect renal function or
cause urinary tract infection (UTI) or sepsis.
Many structural abnormalities of the urinary tract
are characterized by hydronephrosis, which is
frequently assumed to be obstructive. However,
often antenatal hydronephrosis (ANH) results
from nonobstructive causes, including vesicoure-
teral reflux (VUR), multicystic dysplastic kidney,
and certain abnormalities of the ureteropelvic
and ureterovesical junction.
90.2 Development of the Kidney
and Renal Function
The kidney is derived from the ureteral bud and the
metanephric blastema. During the 5th week of ges-
tation, the ureteral bud arises from the mesonephric
(Wolffian) duct and penetrates the metanephric
blastema, which is an area of undifferentiated mes-
enchyme on the nephrogenic ridge. The ureteral
bud undergoes a series of approximately 15 genera-
J. S. Elder (*)
Division of Pediatric Urology, Massachusetts General
Hospital, Boston, MA, USA
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_90
90
tions of divisions, and by 20 weeks’ gestation, it
forms the entire collecting system, that is, the ureter,
renal pelvis, calyces, papillary ducts, and collecting
tubules. Under the inductive influence of the ure-
teral bud, nephron differentiation begins during the
7th week of gestation. By 20 weeks’ gestation,
when the collecting system is completely devel-
oped, approximately one-third of the nephrons are
present. Nephrogenesis continues at a nearly expo-
nential rate and is complete at 36 weeks’ gestation.
Throughout gestation, the placenta functions as the
fetal hemodialyzer, and the fetal kidneys play a
minor role in the maintenance of fetal salt and water
homeostasis. Formation of urine begins between the
5th and 9th weeks of gestation.
The rate of urine production increases
throughout gestation, and at term, volumes have
been reported to be 51 mL/h. The glomerular fil-
tration rate (GFR) has been measured at 6 mL/
min/1.73  m2 at 28 weeks’ gestation, increasing
to 25  mL/min/1.73  m2 at term, and thereafter
triples by 3 months of age. The main factors
responsible for this increase in GFR after birth
include an increase in the capillary surface area
available for filtration, changes in intrarenal vas-
cular resistance, and redistribution of renal blood
flow to the cortical nephrons, which are much
more numerous than the medullary nephrons. A
congenital obstructive lesion of the urinary tract
may have a deleterious effect on renal function.
Severe early obstructive uropathy disrupts nephro-
genesis and results in renal dysplasia.
1249
1250
90.3 The Fetus with Antenatal
Hydronephrosis
When a fetus is identified with a suspected uri-
nary tract abnormality, the goals of manage-
ment include determining the differential
diagnosis, assessment of associated anomalies,
and determining the fetal and postnatal risk of
the anomaly. Hydronephrosis is recognized by
demonstrating a dilated renal pelvis and calyces.
The ureter and bladder may be dilated also. The
likelihood of having a significant urinary tract
abnormality is directly proportional to the sever-
ity of hydronephrosis. If the renal pelvic diameter
is more than 2 cm, 94% have a significant abnor-
mality of the urinary tract requiring surgery or
long-term urologic follow-up. If the fetal renal
pelvic diameter is between 1.0 and 1.5 cm, 50%
have an abnormality, and if the dilated renal pelvis
is less than 1  cm, only 3% have a significant
abnormality. A renal pelvic diameter of at least
4  mm before 27 weeks’ gestation and at least
7 mm after 27 weeks’ gestation has been consid-
ered significant. The later the sonogram is per-
formed, the more likely an existing abnormality
will be detected, because the obstructed renal pel-
vis gradually enlarges throughout gestation. In
addition, in utero the fetus is usually upside down
in the uterus and urine is draining uphill. For
example, Fugelseth and colleagues reported that
only one-third of a series of women carrying
babies with a urologic anomaly had an abnormal
ultrasound (US) study at 15–21 weeks’ gestation.
The differential diagnosis of ANH is provided
in Table 90.1. Virtually all of these conditions can
cause bilateral hydronephrosis. A distended blad-
der and bilateral hydronephrosis are suggestive
of bladder outlet obstruction, such as posterior
urethral valves or a large ectopic ureterocele
obstructing the bladder neck, but fetuses with
nonobstructive conditions, such as high-grade
VUR or prune belly syndrome, also may have
bilateral hydroureteronephrosis and a distended
bladder. In fetuses with a urologic anomaly, asso-
ciated abnormalities are common. For example,
in an early series of fetuses with bilateral hydro-
nephrosis and oligohydramnios, 16 of 31 (55%)
J. S. Elder
Table 90.1  Causes of antenatal hydronephrosis
Anomalous UPJ/UPJ obstructiona
Multicystic kidneya
Retrocaval ureter
Primary obstructive megauretera
Nonrefluxing nonobstructed megauretera
Vesicoureteral refluxa,b
Midureteral stricturea
Ectopic ureterocelea,b
Ectopic uretera
Posterior urethral valvesa,b
Prune belly syndromea,b
Urethral atresiab
Hydrocolposa,b
Pelvic tumora,b
Cloacal abnormalitya,b
May be unilateral or bilateral
a
Bladder may be distended
b
had an associated structural or chromosomal
abnormality (Reuss et al. 1988). Congenital heart
disease and neurologic deformities can often be
detected, if they are present. In contrast, large
bowel abnormalities, such as imperforate anus,
are more difficult to detect by prenatal sonogra-
phy, whereas recognition of small bowel anoma-
lies, such as atresia, are usually straightforward.
The considerations in determining fetal man-
agement include overall fetal well-being, gesta-
tional age, whether the hydronephrosis is
unilateral or bilateral, and the volume of amniotic
fluid. Until recently, there were no guidelines for
determining how frequently to image the fetus or
whether a specific intervention was necessary. If
hydronephrosis is unilateral, usually no specific
fetal therapy is necessary. For example, if the
hydronephrosis is secondary to a ureteropelvic
junction (UPJ) obstruction, even if function is
poor, the kidney has a significant capacity for
improvement in function following neonatal
pyeloplasty. Even with bilateral UPJ obstruction
(characterized by bilateral hydronephrosis and a
normal bladder), the amniotic fluid volume and
pulmonary development typically are normal.
Consequently, specific interventions, such as per-
cutaneous drainage of the fetal kidney or early
delivery to allow immediate urologic surgery, are
unwarranted. These same principles apply to pri-
mary obstructive megaureter.
90  Management of Antenatal Hydronephrosis
The primary life-threatening congenital uro-
logic anomalies include posterior urethral valves,
urethral atresia, and prune belly syndrome, which
are usually characterized by bilateral hydroure-
teronephrosis and a distended bladder that does
not empty in a male fetus, and many have oligo-
hydramnios. Approximately one-third of infants
with urethral valves eventually develop renal
insufficiency or end-stage renal disease. Although
prune belly syndrome is considered nonobstruc-
tive, neonates with this condition frequently have
renal insufficiency, in large part because of con-
genital renal insufficiency and also from renal
deterioration in children with repeated episodes of
pyelonephritis. Urethral atresia is nearly always
fatal, because the kidneys are usually dysplastic.
A significant adverse prognostic factor is oligohy-
dramnios, which prevents normal pulmonary
development. In fetuses with severe obstructive
uropathy and renal dysplasia, neonatal demise
usually results from pulmonary hypoplasia rather
than chronic renal failure. Intuitively, it would
seem that treatment of the obstructed fetal urinary
tract by diverting the urine into the empty amni-
otic space might allow normal renal development
to occur and restore amniotic fluid dynamics,
stimulating lung development. Experimental pro-
cedures have been performed, including percuta-
neous placement of a vesicoamniotic shunt,
creation of a fetal vesicostomy or pyelostomy, and
even percutaneous urethral valve ablation through
a miniscope. The complication rate is high,
including shunt migration, urinary ascites, stimu-
lation of preterm labor, and chorioamnionitis.
Furthermore, in many cases, irreversible renal
dysplasia has already occurred, and although the
procedure may be successful technically, often
the baby is stillborn, dies of pulmonary hypopla-
sia, or is alive with end-stage renal disease.
Nevertheless, some fetuses may benefit from
aggressive intervention if the kidneys do not have
irreversible dysplasia. Unfavorable prognostic
factors include:
–– Prolonged oligohydramnios
–– Renal cortical cysts
–– Urinary Na >100 mEq/L, Cl > 90 mEq/L, and
osmolarity>210 mOsm/L
1251
–– Beta2-microglobulin > 6 mg/L
–– Reduced lung area and thoracic or abdominal
circumference
Unfavorable urinary electrolytes may reflect stale
urine in the fetal urinary tract. Consequently, perina-
tal centers typically obtain two or three sequential
samples, as subsequent samples yield fetal urine that
is more reflective of true fetal renal function.
90.4 Guidelines on Antenatal
Hydronephrosis
In 2010, the Society for Fetal Urology (SFU)
published a consensus statement regarding the
evaluation and management of ANH. This docu-
ment was a comprehensive attempt to provide a
guide to antenatal and postnatal management of
ANH and identify areas of controversy and pri-
oritize research endeavors. The SFU did not
provide guidance on when a voiding cystoure-
throgram (VCUG) should be performed. With
regard to antibiotic prophylaxis, the authors rec-
ommended that neonates with an increased risk
of UTI (girls, uncircumcised boys, moderate to
severe hydronephrosis, and familial VUR)
should receive prophylaxis until the initial eval-
uation is completed and management is planned
with the family. In 2012, the American
Urological Association (AUA) was asked to
establish formal guidelines for ANH, but they
thought that the SFU was thorough and did not
need updating. The European Society for
Paediatric Urology has published recommenda-
tions on the management of various congenital
urinary tract malformations, but has not
addressed ANH specifically.
In 2013, the Indian Society of Pediatric
Nephrology published updated guidelines on
ANH (Sinha et  al. 2013). These recommenda-
tions were based on systematic reviews of the
literature from 1990 to 2012. Level 1 recommen-
dations included those applicable to most sub-
jects based on consistent information. Level 2
recommendations were those that were based on
equivocal or insufficient information. Highlights
of the recommendations include:
1252
1. It is recommended that ANH be diagnosed
and its severity graded based on anteroposte-
rior diameter (APD) of the fetal renal pelvis
(1B). ANH is present if the APD is ≥4 mm in
second trimester and ≥7  mm in the third
trimester.
2. Termination of pregnancy is not recom-
mended in fetuses with unilateral or bilateral
ANH, except in presence of extrarenal life-­
threatening abnormality (1D).
3. It is recommended that all newborns with a
history of ANH should have postnatal US
examination within the first week of life (1B).
In neonates with suspected posterior urethral
valves, oligohydramnios or severe bilateral
hydronephrosis, ultrasonography should be
performed within 24–48 h of birth (1C). In all
other cases, the US should be performed pref-
erably within 3–7 days, or before hospital dis-
charge (1C).
4. It is recommended that assessment of the sever-
ity of postnatal hydronephrosis be based on the
classification proposed by SFU or anteroposte-
rior diameter of the renal pelvis (1B).
5. It is recommended that neonates with normal
US examination in the first week of life should
undergo a repeat study at 4–6 weeks (1C).
Infants with isolated mild unilateral or bilateral
hydronephrosis (APD <10  mm or SFU grade
1–2) should be followed by sequential US alone,
for resolution or progression of findings (1C).
6. It is recommended that a micturating cysto-
urethrogram (MCU) be performed in patients
with unilateral or bilateral hydronephrosis
with renal pelvic APD > 10 mm, SFU grade
3–4, or ureteric dilatation (1B). MCU should
be performed early, within 24–72 h of life, in
patients with suspected lower urinary tract
obstruction (1D). In other cases, the proce-
dure should be done at 4–6 weeks of age.
MCU is recommended for infants with ante-
natally detected hydronephrosis who develop
a urinary tract infection (1C).
7. It is recommended that infants with moderate
to severe unilateral or bilateral hydronephro-
sis (SFU grade 3–4, APD >10  mm) who do
not show VUR should undergo diuretic renog-
raphy (1C). Infants with hydronephrosis and
dilated ureter(s) and no evidence of VUR
J. S. Elder
should undergo diuretic renography (2C). The
preferred radiopharmaceuticals are 99mTc-­
mercaptoacetyltriglycine (99mTc-MAG3),
99m
Tc-ethylenedicysteine ( Tc-EC) or 99mTc-­
99m
diethylenetriaminepentaacetic acid (DTPA)
(2D). The differential function is estimated
and renogram curve is inspected for a pattern
of drainage. Diuretic renography should be
performed after 6–8 weeks of age (2D). The
procedure may be repeated after 3–6 months
in infants where US shows worsening of pel-
vicalyceal dilatation (2D).
8. It is suggested that surgery be considered in
patients with obstructed hydronephrosis, and
either reduced differential renal function or its
worsening on repeat evaluation (2C). It is sug-
gested that surgery be considered in patients
with bilateral hydronephrosis or hydronephro-
sis in a solitary kidney showing worsening
dilatation and deterioration of function (2D).
9. It is recommended that parents of all infants
with antenatal or postnatal hydronephrosis be
counseled regarding the risk of urinary tract
infections and the need for prompt manage-
ment (1B). It is recommended that infants
with postnatally confirmed moderate or severe
hydronephrosis (SFU 3–4; renal APD >
10  mm) or dilated ureter receive antibiotic
prophylaxis while awaiting evaluation (1C). It
is recommended that all patients detected to
have VUR receive antibiotic prophylaxis
through the first year of life (1B).
In 2014, a multidisciplinary consensus confer-
ence involving eight societies was held and
included pediatric urologists, pediatric nephrolo-
gists, pediatric radiologists, and maternal-fetal
medicine specialists. The purpose was to try to
standardize the fetal evaluation and early postna-
tal management of babies with ANH. For exam-
ple, several different classification schemes had
been published for significant and insignificant
pelvocaliceal dilation. In addition, although the
SFU 4-point grading scale is used by most pediat-
ric urologists, most pediatric radiologists classify
hydronephrosis into mild, moderate, and severe.
The US parameters include anterior-posterior
renal pelvic diameter (APRPD), calyceal dila-
tion, whether it involves the major and/or minor
90  Management of Antenatal Hydronephrosis
calyces, parenchymal thickness and appearance,
normal or abnormal ureter, and normal or abnor-
mal bladder (Nguyen et al. 2010, 2014). Normal
values for urinary tract dilation (UTD) are
APRPD:
Antenatal 16–27 weeks <4 mm
>28 weeks <7 mm
Postnatal (>48 h) <10 mm
Assuming there is no calyceal dilation, the
kidneys have a normal appearance, and the ureter
and bladder are normal, then this is considered a
normal study.
The consensus group then categorized ANH into
antenatal and postnatal risk groups. For antenatal,
there are two risk groups: Low Risk and Increased
Risk. For postnatal, there are three risk groups: Low
Risk, Intermediate Risk, and High Risk. The panel
recommended that all seven urinary tract parame-
ters be described in a written report.
For antenatal presentation, if the APRPD is
4–7  mm at 16–27 weeks or 7–10  mm at > 28
weeks, and there is central or no calyceal dila-
tion, the fetus is UTD A1, Low Risk. In follow-
up, for UTD A1, the panel suggested one
­ optional, at the discretion of the clinician (but is
additional antenatal US at > 32 weeks, and after
birth, a renal US at >48 h to 1 month of age, and
a second renal US 6 months later. Genetic screen-
ing is not indicated unless there are associated
congenital malformations.
If the APRPD is >7  mm at 16–27 weeks or
>10 mm at >28 weeks with any peripheral caly-
ceal dilation or any other upper urinary tract
abnormality, they are classified as UTD A2–3, or
Increased Risk. The assigned risk is based on the
most concerning feature. For UTD A2–3, the
panel recommended a follow-up US in 4–6 weeks,
although with suspected posterior urethral valves
(PUV) or severe bilateral hydronephrosis, more
frequent follow-up was recommended until deliv-
ery. Following delivery, a renal US after 48 h but
before 1 month was suggested, again with more
immediate evaluation if PUV is suspected or there
is significant bilateral hydronephrosis. In addi-
tion, specialist consultation with pediatric urology
or nephrology was recommended.
For postnatal presentation, at >48  h, an
APRPD <10  mm is Normal. If the APRPD is
1253
10–15 mm, there is central calyceal dilation, and
all other parameters are normal, it is classified
UTD P1, Low Risk. SFU hydronephrosis grades
1 and 2 correspond to UTD P1. The panel recom-
mends a follow-up renal US in 1–6 months. A
VCUG and antibiotic prophylaxis are optional, at
the discretion of the clinician. A renal scan is not
recommended.
If the postnatal APRPD is >15  mm, there is
peripheral calyceal dilation and/or abnormal ure-
ters, it is classified UTD P2, Intermediate Risk.
SFU hydronephrosis grade 3 corresponds to UTD
P2. The panel recommends a follow-up renal US
in 1–3 months. A VCUG, antibiotic prophylaxis,
and a functional renal scan are optional, at the
discretion of the clinician.
If the APRPD is >15 mm and there is periph-
eral calyceal dilation, abnormal parenchymal
thickness, abnormal parenchymal appearance,
abnormal ureters, and/or abnormal bladder, it is
classified UTD P3, High Risk. SFU hydrone-
phrosis grade 4 corresponds to UTD P3. The con-
sensus group recommends a follow-up renal US
in 1 month. A VCUG and antibiotic prophylaxis
are recommended. A functional renal scan is
virtually always recommended).
This classification scheme was validated at an
NIH Consensus Conference by the participants.
In addition, it was evaluated in a retrospective
study of 490 patients (Hodhod et al. 2016). The
authors found that the UTD classification appro-
priately identified babies that were likely to
require surgical intervention, while the SFU
hydronephrosis grading system was most likely
to predict the likelihood of resolution of
hydronephrosis.
90.5 Management
of the Newborn
with Antenatal
Hydronephrosis
90.5.1 Management in the Nursery
At birth, the abdomen is inspected to detect the
presence of a mass, which most often is second-
ary to a multicystic dysplastic kidney or UPJ
1254
obstruction. In male newborns with posterior
­urethral valves, often a walnut-shaped mass, rep-
resenting the bladder, is palpable just superior to
the pubic symphysis. Newborns should also be
evaluated for anomalies involving other organ
systems, as urinary tract abnormalities often
occur in babies with congenital heart disease,
lung abnormalities, and anorectal malformations.
Renal function should be monitored with serial
serum creatinine levels, particularly in infants
with bilateral hydronephrosis or a unilateral
poorly functioning kidney. At birth, the serum
creatinine level is identical to the mother’s, but
by 1 week of age, the creatinine should decrease
to 0.4 mg/dL. The exception is premature infants,
in whom the creatinine may not decrease until
these children reach 34–35 weeks’ conceptional
age because of renal immaturity before that age.
90.5.2 Antibiotic Prophylaxis
Neonates with hydronephrosis who are at risk
for UTI should be placed on antibiotic prophy-
laxis with either amoxicillin 20  mg/kg daily or
cephalexin 15 mg/kg daily. At 2 months beyond
term, the prophylaxis is usually changed to
trimethoprim-­sulfamethoxazole suspension,
cephalexin, or nitrofurantoin because these med-
ications provide broader antibacterial coverage.
In addition, circumcision should be considered
in male neonates to minimize the likelihood of
UTI.  However, which babies are at risk and
therefore who needs prophylaxis is controver-
sial. The risk group UTD P3, High Risk deserves
prophylaxis. However, there is significant vari-
ability among pediatric urologists and nephrolo-
gists in prescribing prophylaxis for children with
hydronephrosis. For example, 30% of Canadian
pediatric nephrologists recommended CAP for
bilateral low-grade ANH compared with 11% of
pediatric urologists (Braga et  al. 2014). With
high-grade ANH, 73% of nephrologists and 38%
of urologists recommended CAP.  Their recom-
mendations for VCUG also were quite variable.
Early studies suggested that children with
VUR, ectopic ureter and ureterocele, and posterior
urethral valves benefit from prophylaxis, but those
with hydronephrosis secondary to an abnormality
J. S. Elder
of the UPJ or ureterovesical junction are not at
increased risk. More recently, infants with high-
grade hydronephrosis (OR 2.40), female gender
(OR 3.16), and uncircumcised males (OR 3.63)
were at the highest risk, but multivariate analysis
suggested that prophylaxis was not beneficial
(Zaraba et al. 2014). However, a subsequent pro-
spective study from the same institution indicated
that ‘lack of continuous antibiotic prophylaxis’,
hydroureteronephosis, and VUR were significant
risk factors for febrile UTI (Braga et  al. 2015).
Another retrospective study found that the risk of
febrile UTI was 7.9% in children receiving pro-
phylaxis vs. 18.7% in those not receiving prophy-
laxis. Children with ureteral dilation >11  mm,
ureterovesical obstruction, and high-grade VUR
were at the greatest risk for a febrile UTI (Herz
2014). Most agree that children with low-grade
hydronephrosis do not benefit from prophylaxis.
Antibiotic resistance is a definite concern when
prescribing prophylaxis unnecessarily.
90.5.3 Initial Radiologic Evaluation
Radiologic evaluation should be performed to
delineate the abnormality responsible for changes
in prenatal sonography. Serial renal sonograms, a
voiding cystourethrogram (VCUG), and a
diuretic renogram usually provide the diagnostic
information necessary to guide management,
although all of these studies are unnecessary in
many children.
90.5.3.1 Renal/Bladder Ultrasound
A renal and bladder sonogram should be obtained
first. Because neonates have transient oliguria, a
dilated or obstructed collecting system may
appear normal for the first 24–72  h of life
(Fig.  90.1). Ideally, if unilateral hydronephrosis
was present prenatally, the renal sonogram should
not be obtained until 72 h to maximize its sensi-
tivity. It is most appropriate to wait until 3–4
weeks of age unless the antenatal US showed
severe unilateral or bilateral hydronephrosis.
Renal length, degree of caliectasis, parenchy-
mal thickness, and presence or absence of
ureteral dilation should be documented. The
­
severity of hydronephrosis can be graded from 1
90  Management of Antenatal Hydronephrosis 1255

a b

Fig. 90.1 (a) Ultrasound of left kidney at 12 h. Normal study. Echolucent areas (arrows) in renal cortex are renal pyra-
mids (normal finding). (b) Ultrasound of same patient at 6 weeks shows grade 4 hydronephrosis

Table 90.2  Grading of hydronephrosis with the likelihood that a significant urological
Renal Image condition is present. Lee et al performed a meta-
Renal analysis of reports of ANH and determined that
Grade of Central renal parenchymal the risk of finding postnatal pathology was
hydronephrosis complex thickness
0 Intact Normal
11.9% with mild ANH, 45.1% for moderate, and
1 Slight splitting Normal 88.3% for severe ANH. Their definitions of mild,
2 Evident splitting, Normal moderate, and severe hydronephrosis depended
complex on gestational age at the time of diagnosis.
confined within Similarly, Sidhu et al performed a meta-analysis
renal border
3 Wide splitting Normal
and found that when postnatal hydronephrosis
pelvis dilated was SFU grade 1 or 2, there was stabilization or
outside renal resolution of pelviectasis in 98%, whereas when
border; calyces there was SFU grade 3 or 4, there was stabiliza-
uniformly dilated
tion or resolution in only 51%.
4 Further dilatation Thin
of pelvis and The bladder should be imaged to detect a
calyces (calyces dilated posterior urethra (urethral valves), bladder
may wall thickening, ureteral dilation, inadequate blad-
appearconvex)
der emptying, or a ureterocele. Perineal sonogra-
phy may demonstrate a dilated prostatic urethra,
to 4 using the Society for Fetal Urology (SFU) which is consistent with posterior urethral valves.
grading scale (Table  90.2). Inexperienced radi-
ologists may misinterpret a normal neonatal kid-
90.5.3.2 Voiding Cystourethrogram
ney with hypoechoic pyramids for caliectasis.
In selected cases, a VCUG should be performed.
Most significant urologic anomalies that require
This study may demonstrate VUR, posterior ure-
surgical correction or long-term urological fol-
thral valves, or a bladder diverticulum. A radio-
low-up are associated with SFU grade 3 or 4
graphic cystogram is preferred over a radionuclide
hydronephrosis. More sophisticated analyses,
cystogram because the latter does not provide
such as the renal resistive index, as well as uri-
sufficient delineation of bladder and urethral
nary studies have been assessed, but efforts to
anatomy and because VUR, if present, cannot be
demonstrate obstruction have been inconsistent.
graded. In an analysis by the AUA Pediatric
The degree of pelvocaliectasis correlates closely
Vesicoureteral Reflux Guidelines Committee, an
1256
overall VUR detection rate of 16.2% was found.
The mean incidence of VUR into a non-dilated
kidney was 4.1%. In cases with ANH and a nor-
mal postnatal sonogram, the incidence of VUR
was 17%. The prevalence of VUR was signifi-
cantly higher in girls than boys with ANH. The
likelihood is highest if there is SFU grade 3 or 4
hydronephrosis or if a dilated ureter is identified.
The chance of identifying VUR on a VCUG is
less if there is only SFU grade 1 or 2.
90.5.3.3 What If the Initial Sonogram
Is Normal?
A common dilemma is whether a full evaluation is
necessary if the initial renal sonogram is normal.
Assuming a significant degree of fetal renal pelvic
dilatation (i.e., >4  mm anteroposterior pelvic
diameter before 33 weeks, 7 mm diameter after 33
weeks) was present, the child may have
VUR. Blane and colleagues reported that 12% of
children with grade V, 31% with grade IV, and
80% with grade III VUR had a normal renal sono-
gram. However, the AUA Reflux Guidelines deter-
mined that the mean incidence of VUR into a
non-dilated kidney was 4.1%. Because VUR may
cause intermittent renal pelvic dilation, theoreti-
cally babies with prenatal hydronephrosis and a
normal postnatal sonogram may have VUR, and
early diagnosis and medical treatment of VUR
may reduce the likelihood of developing reflux
nephropathy. On the other hand, others have advo-
cated performing a VCUG only if the postnatal
sonogram is abnormal; however, in these reports,
neonates with a normal postnatal renal sonogram
were not systematically evaluated to determine the
real incidence of VUR in this group. Currently,
most do not recommend a VCUG unless the post-
natal renal sonogram demonstrates grade 3 or 4
hydronephrosis and/or ureteral dilation.
90.5.4 Follow-Up Evaluation
and Treatment
If bilateral hydronephrosis or unilateral hydrone-
phrosis in a solitary kidney is present, then close
monitoring of the serum creatinine and electro-
lytes is necessary. If the hydronephrosis is caused
by posterior urethral valves, then valve ablation
J. S. Elder
should be performed before hospital discharge. If
the hydronephrosis is secondary to VUR, the
infant should be placed on prophylaxis and man-
aged as described later in this chapter. If the
hydronephrosis is SFU grade 3 or 4, the VCUG is
negative and bilateral UPJ or ureterovesical junc-
tion obstruction is suspected, evaluation with
diuretic renography is indicated. If unilateral
hydronephrosis and a normal contralateral kidney
are present, abnormalities in serum creatinine or
electrolytes are uncommon. Nevertheless, these
serum studies should be drawn to document that
renal function is normal. Usually, follow-up func-
tional radiographic studies can be delayed until
6–12 weeks of age, when renal function is more
mature and studies of renal function and obstruc-
tion are more likely to be accurate. If the sono-
gram and VCUG are normal, then only a follow-up
sonogram in 6–8 weeks is necessary. In general, if
hydronephrosis is discovered on the initial post-
natal sonogram, pediatric urologic or nephrologic
consultation is advisable to direct subsequent
radiologic evaluation and plan therapy.
90.5.4.1 Diuretic Renogram
The diuretic renogram is used to determine
whether upper urinary tract obstruction is pres-
ent. It is used to assess the differential renal func-
tion and efficiency of drainage of the kidneys.
Infants with grade 3 and 4, and occasionally
grade 2, hydronephrosis should undergo this
study. Mercaptoacetyl triglycine (MAG-3) is
generally used and is secreted by the renal
tubules. It provides excellent images with mini-
mal background activity. During a diuretic reno-
gram, a small dose of the radiopharmaceutical is
injected intravenously. During the first 2–3 min,
renal parenchymal uptake is analyzed and com-
pared, allowing computation of differential renal
function. Subsequently, excretion is evaluated.
After 20–30  min, furosemide 0.5–1  mg/kg is
injected intravenously, and the rapidity and pat-
tern of drainage from the kidneys to the bladder
are analyzed. If no upper urinary tract obstruction
is present, then normally half of the radionuclide
is cleared from the renal pelvis within 10–20 min,
termed the t½. A t½ > 20 min is consistent with
upper urinary tract obstruction, but it is not diag-
nostic of obstruction, because there are factors
90  Management of Antenatal Hydronephrosis 1257

that can prolong the t½ in addition to an obstruc- to establish a diagnosis and establish a plan of
tive lesion (see below). A t½ between 15 and management. In particularly complicated cases,
20  min is indeterminate. The images generated however, cystoscopy with retrograde pyelogra-
usually provide an accurate assessment of the site phy, computed tomography (CT) scan, antegrade
of obstruction. With unilateral hydronephrosis, pyelography, or a Whitaker antegrade perfusion
the normal kidney should be used as a control for test is necessary.
interpretation, and ideally it has already
­accumulated and excreted most of the radionu-
clide before the furosemide is administered. 90.6 Congenital Anomalies
Numerous variables affect the outcome of the Causing ANH
diuretic renogram. For example, newborn kidneys
are functionally immature, and in some cases, 90.6.1 UPJ Obstruction or
even normal kidneys may not demonstrate normal Anomalous UPJ
drainage following diuretic administration.
Dehydration prolongs parenchymal transit and The most common cause of severe hydronephro-
can blunt the diuretic response. Giving an insuf- sis without a dilated ureter or bladder in newborn
ficient dose of furosemide may result in slow or infants is UPJ obstruction, which results from an
inadequate drainage. In addition, a full bladder intrinsic fibrotic narrowing at the junction
may impede bladder drainage. Furthermore, if between the ureter and renal pelvis (Fig. 90.2). At
VUR is present, continuous catheter drainage is times, an accessory artery to the lower pole of the
mandatory to prevent the radionuclide from kidney also causes extrinsic obstruction, but this
refluxing from the bladder into the dilated upper finding is rare in newborns with hydronephrosis;
tract, causing a prolonged washout phase. it is much more likely to occur in older children
Consequently, a urethral catheter should be and adults. In kidneys with a UPJ obstruction,
inserted and bladder drainage measured. renal function may be significantly impaired
from pressure atrophy.
90.5.4.2 Magnetic Resonance The anomaly is corrected by performing a
Urography pyeloplasty, in which the stenotic segment is
Magnetic resonance urography (MRU) occasion- excised and the normal ureter and renal pelvis are
ally is used to evaluate suspected upper urinary reattached. Success rates are 91–98%. Lesser
tract pathology. The child is hydrated and given degrees of UPJ narrowing may cause mild hydro-
intravenous furosemide. Next, gadolinium-­ nephrosis, which is usually nonobstructive, and
DTPA, which is filtered and excreted, is injected typically these kidneys function normally.
intravenously and routine T1-weighted and fat-­ Another cause of mild hydronephrosis is fetal
suppressed fast spin-echo T2-weighted imaging folds of the upper ureter (Fig. 90.3), which also
is performed through the kidneys, ureters, and are nonobstructive. The spectrum of nonobstruc-
bladder. This study provides superb images of the tive UPJ abnormalities often is termed ‘anoma-
pathology, and methodology has been developed lous UPJ’.
to allow assessment of differential renal function Hydronephrosis in many newborns gradually
and drainage. There is no radiation exposure, but diminishes or resolves over months to years. The
younger children need sedation or general anes- goal of early evaluation is to determine whether a
thesia. It is the procedure of choice for delineat- true anatomic obstruction is present that should be
ing complex genitourinary pathology (e.g., corrected or whether it is safe to follow the infant
cross-fused ectopia with hydronephrosis and/or non-operatively. Defining what constitutes
segmental multicystic kidney, cloacal anomaly). obstructive and nonobstructive hydronephrosis is
a constant source of debate in pediatric urology.
90.5.4.3 Ancillary Studies Cartwright and colleagues in 1992 studied 80
In most cases, a renal sonogram, VCUG, and neonates with suspected UPJ obstruction. Of 39
diuretic renogram provide sufficient information with unilateral hydronephrosis and at least 35%
1258
a c
Fig. 90.2 (a) Ultrasound shows grade 4 left hydrone-
phrosis without dilated ureter. VCUG was normal (b)
MAG-3 diuretic renogram. Differential renal function
46% left, 54% right. Right kidney drains spontaneously
differential renal function who were managed
non-operatively, only six (15%) later underwent
pyeloplasty, primarily because of deteriorating
differential renal function on renal scintigraphy.
Following pyeloplasty, the differential renal func-
tion returned to its initial level in these patients.
One might question whether early pyeloplasty in
these patients would have allowed the renal func-
J. S. Elder
before furosemide is administered at 20 min. There is no
drainage from the left kidney, consistent with left UPJ
obstruction. (c) Follow-up renal US 4 months following
left pyeloplasty showing improved hydronephrosis
tion to improve to 50% (normal). The remaining
patients managed non-operatively maintained dif-
ferential function greater than 40%. Onen et  al
reported 104 consecutive neonates with unilateral
hydronephrosis managed non-operatively, with
follow-up as long as 5 years. In follow-up, only
seven (7%) underwent pyeloplasty because of
reduction in differential renal function of more
90  Management of Antenatal Hydronephrosis
Fig. 90.3  Excretory urogram in infant with bilateral
grade 2 hydronephrosis shows bilateral fetal folds of
upper ureter (arrows). Note sharp calyces (normal) [from
previous chapter]
than 10% or progression of hydronephrosis.
Pyeloplasty returned differential renal function to
prepyeloplasty levels in all cases. Of 16 patients
with significantly reduced renal function on initial
scan and grade 4 hydronephrosis, rapid improve-
ment was noted on follow-­up diuretic renography
in 15 and the washout curve became nonobstruc-
tive in six. In addition, hydronephrosis disap-
peared in six, improved in six, remained stable in
three, and deteriorated in one. The physiology of
the resolution or reduction in hydronephrosis and
the improvement in differential renal function in
these babies is unknown. In another more recent
report from the same institution, of 19 newborns
with bilateral grade 3 or 4 hydronephrosis, a total
of 13 kidneys were subjected to pyeloplasty. Of
those managed non-operatively, 21 kidneys were
grade 0–2 and grade 3 in two kidneys. The mean
follow-up time to achieve maximum improve-
ment in hydronephrosis was 10 months.
Although these studies suggest that it is safe
or appropriate to manage neonates with a sus-
1259
pected UPJ obstruction non-operatively, an
infant’s kidney has greater capacity for improve-
ment in differential renal function than an older
child’s. In addition, all of these studies base dif-
ferential renal function on the uptake during the
first 2–3 min of the study, and there is variability
in the way this percentage is calculated. Finally,
these studies have not reported the pattern of
washout on diuretic renography. In a review of
renal biopsies obtained at pyeloplasty at the
author’s previous institution, 63% showed mini-
mal or no obstructive histologic changes; how-
ever, of those with differential function greater
than 40%, 21% showed significant histopatho-
logic changes, including reduced glomerular
number, glomerular hyalinization, interstitial
inflammation, and dysplastic glomeruli (Elder
et  al. 1995). Of the kidneys with a differential
function less than 40%, 33% showed minimal or
no histologic changes. Overall, in 25% of the
patients the findings on renal biopsy did not cor-
relate with the computed differential renal func-
tion and reflect the need for more sensitive
markers of obstruction. The approach of many
pediatric urologists to neonates with a suspected
UPJ obstruction is as follows. The hydronephro-
sis is graded from 1 to 4 using the SFU grading
scale and a VCUG is obtained for grade 3 and 4
hydronephrosis. Nearly all infants requiring
pyeloplasty have grade 3 or 4 hydronephrosis,
and those with grade 1 or 2 hydronephrosis do
not seem to be at significant long-term risk. If a
neonate has an abdominal mass from a hydrone-
phrotic kidney, bilateral severe hydronephrosis,
or a solitary kidney, a prompt MAG-3 diuretic
renogram is obtained. If signs of obstruction are
apparent, prompt pyeloplasty is performed.
Otherwise, the diuretic renogram is obtained at
6–12 weeks of age. If the study is obtained when
the kidneys are immature, the diuretic response
to furosemide might show delayed drainage even
with an unobstructed UPJ. If diuretic renography
shows at least >40% differential renal function
and there is some drainage on the diuretic reno-
gram, the child can be managed non-operatively,
regardless of the drainage pattern. A follow-up
renal sonogram is performed 3 months later. If
the hydronephrosis is unchanged from baseline, a
repeat MAG-3 diuretic renogram is obtained. If
1260
there is deterioration in differential renal function
or worsening of the diuretic washout curve,
pyeloplasty is recommended. However, if these
parameters remain stable or improved and the
child does not develop a UTI, follow up 3–6
months later with another renal sonogram or
MAG-3 diuretic renogram is performed, and
management is individualized. It is incumbent on
clinicians caring for these infants to have a good
understanding of the vagaries of the diuretic
renogram and to monitor infants with a suspected
UPJ obstruction closely. In addition, a review of
the radiologic studies (not just the radiology
report) by the pediatric urologist is important.
There has been significant progress in the
development of minimally invasive techniques
in pediatric pyeloplasty, even in infants. Although
infant pyeloplasty can be performed through a
small incision (lumbotomy or flank muscle-­
splitting) in many centers, some pediatric urolo-
gists are performing the procedure with
traditional laparoscopic techniques or laparo-
scopically with the da Vinci robot. Success rates
have been compared to series of open surgical
repair. Hospital stay and narcotic use are less
with the minimally invasive approach. Generally,
a transperitoneal approach has been used, with
mobilization of the colon. However, on the left
side, the pyeloplasty can be performed with a
transmesenteric approach.
90.6.2 Multicystic Dysplastic Kidney
A multicystic dysplastic kidney is composed of
multiple noncommunicating cysts of varying
sizes with a stromal component with dysplastic
elements. These kidneys do not function. Although
multicystic kidney is the most common cause of
an abdominal mass in neonates, the vast majority
of multicystic kidneys are detected by prenatal
sonography. Some clinicians incorrectly assume
that multicystic kidney and polycystic kidney are
synonymous terms. Polycystic kidney disease is
an inherited disorder and has an ‘adult form’
(autosomal dominant) and an ‘infantile form’
(autosomal recessive) and affects both kidneys. In
contrast, a multicystic kidney is almost always
J. S. Elder
Fig. 90.4  Renal US in newborn with right multicystic
dysplastic kidney, showing kidney replaced by multiple
noncommunicating cysts of various sizes with minimal
dysplastic parenchyma
unilateral and is usually not an inherited disorder.
Sonography of multicystic kidneys is often diag-
nostic, demonstrating multiple echolucent cysts
of varying sizes with no discernible cortex
(Fig. 90.4). Occasionally, the cysts may resemble
a severe UPJ obstruction with minimal paren-
chyma, termed the ‘hydronephrotic variant’. The
contralateral kidney is abnormal in 5–10% of
cases. Renal scintigraphy (MAG-3 or DMSA
[dimercaptosuccinic acid] scan) shows non-func-
tion, but with current US techniques, renal scin-
tigraphy generally is unnecessary for confirmation.
On occasion, there is a segmental multicystic kid-
ney, in which there is a complete duplication
anomaly of the upper urinary tract, with the upper
pole being multicystic. Previously, a VCUG often
was ordered because as many as 15% have contra-
lateral VUR, but currently, it seems unnecessary
unless there is contralateral hydronephrosis.
The management of a multicystic kidney is
becoming less controversial. If an abdominal
mass that is symptomatic is present, early
nephrectomy is indicated. However, left
untreated, most multicystic kidneys become
smaller. Potential complications include malig-
nancy and hypertension. In a review of 26 clinical
series, no cases of Wilms’ tumor were reported
among 1041 children, and the maximum esti-
mated risk is 0.07% (Chang et al. 2018). Tumors
arise from the stromal, not the cystic, component
of multicystic kidneys. Consequently, even if the
90  Management of Antenatal Hydronephrosis
cysts regress completely, the likelihood that the
kidney could develop a neoplasm is not altered.
With regard to hypertension, the risk is 3.2%.
Generally, a follow-up sonogram is recom-
mended at 6 months of age. If the cysts enlarge,
the stromal core increases in size, or hyperten-
sion develops, laparoscopic simple nephrectomy
is recommended. However, further follow-up
sonography is unnecessary unless there is con-
cern regarding the contralateral kidney, because
finding a Wilms’ tumor incidentally would be
extremely rare. Because of the occult nature of
hypertension, annual blood pressure measure-
ment is recommended, and if hypertension
occurs, nephrectomy should be considered.
90.6.3 Primary Megaureter
(Non-refluxing)
A megaureter refers to a wide ureter and may be
(1) primary or secondary, (2) obstructive or non-
obstructive, and (3) refluxing or non-refluxing.
Non-refluxing megaureter results from an aperi-
staltic segment of the distal ureter that does not
allow normal propulsion of urine. In this condi-
tion, sonography shows a dilated ureter and renal
pelvis with variable renal parenchymal atrophy.
VCUG shows no VUR in most cases. Before the
antenatal sonography era, most patients with this
condition presented with flank pain, flank mass,
pyelonephritis, hematuria, or stone disease.
Surgical correction consists of excision of the
aperistaltic segment, tailoring (also known as
a b
Fig. 90.5  Female newborn with non-refluxing left mega-
ureter discovered by antenatal US (a and b). Ultrasound
shows grade 4 hydronephrosis and dilated ureter. Diuretic
renogram, 8 weeks of age, showed 50% differential renal
function in left kidney. Obstructive drainage pattern was
1261
tapering) of the ureter, and reimplantation of the
ureter into the bladder. Although severe hydrone-
phrosis may be present, often there is gradual
reduction in hydronephrosis over a period of sev-
eral years (Fig. 90.5). Braga et al. (2016) reported
that the majority of their patients demonstrated
resolution of hydronephrosis in a median of 17
months, although those with a mean ureteral
diameter >17 mm required surgical intervention.
This group found that continuous antibiotic pro-
phylaxis and circumcision reduced the risk of
UTI. DiRenzo et al. (2015) found that resolution
or improvement of hydronephrosis occurs in all
cases of mild postnatal dilation and 60% of those
with moderate or severe upper urinary tract dila-
tation. The British Association of Paediatric
Urologists recommends initial nonoperative
management of primary obstructive megaureter.
In follow-up, surgical intervention is recom-
mended for UTI, pain, worsening hydronephro-
sis, differential renal function <40%, or a
significant drop in renal function. Consequently,
most of these patients may be followed non-­
operatively on antibiotic prophylaxis and serial
monitoring of renal function and drainage.
In neonates with an antenatal diagnosis of
hydroureteronephrosis, a renal sonogram and
VCUG should be obtained. Early management is
identical to that of neonates with suspected UPJ
obstruction. If an abdominal mass, solitary kid-
ney, or bilateral hydroureteronephrosis is present,
then a MAG-3 diuretic renogram should be
obtained promptly. Otherwise, the study is
deferred until 6–8 weeks of age. In some centers,
c
noted following administration of furosemide. Patient was
managed non-operatively. (c) Ultrasound 6 months slater
shows grade 1 hydronephrosis. Diuretic renogram later
showed normal drainage pattern
1262
the diuretic-stimulated drainage from the renal
pelvis and ureter are measured separately. If the
differential renal function is at least 40%, the
child generally is managed non-operatively, even
with grade 4 hydronephrosis, and a follow-up
renal sonogram or diuretic renogram is obtained
every 3–6 months. These infants should receive
prophylactic antibiotics while stasis is present in
the upper ureter and kidney.
Early repair of megaureter has a higher com-
plication rate, including VUR and ureteral
obstruction, than in older children. There are
three other treatment options if surgical repair
seems necessary in the neonate or young infant.
The first is to perform a temporary cutaneous ure-
terostomy, allowing the ureter to decompress
over a period of 12–18 months. The ureterostomy
can be either an end cutaneous ureterostomy or
loop ureterostomy. It should be placed over the
inguinal skin crease. If there is reduced renal
function, it is unlikely to improve following uri-
nary diversion. Subsequently, the ureterostomy
can be taken down and ureteral reimplantation
with or without tapering is performed. Another
option is temporary refluxing ureteral reimplan-
tation, in which the ureter is anastomosed end-to-­
side to the bladder and performing traditional
ureteral reimplantation when the child is older.
The other option is to dilate the ureterovesical
junction and insert a double J ureteral stent, leav-
ing it interposed between the bladder and the kid-
ney. Farrugia et  al reported a series of 19
megaureters that were stented at a median age of
6 months, 1/3 endoscopically. Nearly 1/3 had
stent-related complications, but 50% had
improvement in upper tract drainage following
stent removal.
In summary, if differential renal function
remains normal and the child is asymptomatic, it
seems safe to follow these patients with renal
sonograms and diuretic renography to monitor
hydronephrosis, and renal function and drainage.
If renal functional deterioration, slowing of upper
urinary tract drainage, or UTI occurs, ureteral
reimplantation is recommended.
J. S. Elder
90.6.4 Ureterocele and Ectopic
Ureter
A ureterocele is a cystic dilatation of the distal
end of the ureter and is usually obstructive. In
children, they usually extend through the bladder
neck, termed ‘ectopic’, but may remain entirely
within the bladder, termed ‘intravesical’ or
‘orthotopic’. Ectopic ureteroceles and ectopic
ureters occur more commonly in girls than in
boys and usually are associated with the upper
pole of a completely duplicated collecting sys-
tem. In boys with a ureterocele, however, 40%
drain a single collecting system. Prenatal sonog-
raphy typically shows either hydroureterone-
phrosis or upper pole hydronephrosis with a
dilated ureter. These conditions are bilateral in
10–15% of patients.
Early evaluation consists of:
–– Sonography shows a hydronephrotic upper
pole connected to a dilated ureter; the uretero-
cele typically is visualized in the bladder
(Fig. 90.6a, b).
–– VCUG often visualizes the ureterocele and
demonstrates whether VUR is present, either
into the lower pole moiety or contralateral col-
lecting system; an ectopic ureter that inserts
into the bladder neck also typically refluxes
into the upper pole obstructed ureter.
–– DMSA or MAG-3 renal scan shows whether
the moiety drained by the ectopic system
functions; this study may be done at 1–2
weeks of age because the result does not
change with functional maturity (Fig. 90.6c).
Alternatively, a CT or MRU demonstrates the
anatomy adequately, although it may not pro-
vide functional information regarding the
obstructed upper pole.
The management of neonates with a uretero-
cele is highly individualized. The least invasive
initial form of therapy is transurethral incision
(TUI), which can be performed either with a 3-F
Bugbee electrode or the holmium:YAG laser.
90  Management of Antenatal Hydronephrosis
a b
Fig. 90.6  Newborn with antenatal left hydronephrosis in
a duplex kidney. (a) Dilated left upper pole, normal lower
pole. The distal left upper pole ureter was dilated. (b)
The ureterocele is punctured several times at its
junction with the bladder mucosa. If the uretero-
cele is ectopic, it must be punctured both in the
bladder and in the urethra. TUI provides satis-
factory upper tract decompression with a single
procedure in >90% of cases. However, there is a
significant risk of post-operative VUR through
the ureterocele into the upper pole moiety,
which may require subsequent definitive treat-
ment. If the ureterocele is orthotopic, approxi-
mately 30% show VUR following TUI, whereas
if it is ectopic, 75% have VUR following
TUI. TUI is often the only procedure necessary.
1263
Ureterocele in bladder. (c) DMSA scan, left kidney on
left. Normal right kidney and left lower pole, and severely
reduced function in left upper pole
TUI of a ureterocele draining a nonfunctioning
moiety will not result in the development of any
significant degree of function. In recent years,
many centers have been performing minimally
invasive (laparoscopic) upper pole hemine-
phrectomy. These procedures have been per-
formed either with a retroperitoneal or
transperitoneal approach. Common complica-
tions include perirenal urinoma and, in some
cases, devascularization of the lower pole moi-
ety. The latter complication is most common in
infants. Another option is to perform transperi-
toneal laparoscopic partial nephrectomy with
1264
robotic assistance. Consequently, if the hydro-
nephrotic upper pole does not function, this
author maintains the infant on antibiotic pro-
phylaxis and laparoscopic upper pole hemine-
phrectomy (or nephrectomy, if the entire kidney
does not function) with or without robotic assis-
tance performed electively at 6 months of age,
assuming the upper pole system remains hydro-
nephrotic. If the renal scan shows significant
upper pole function, however, ureteropyelos-
tomy or ureteroureterostomy, in which the upper
pole ureter is anastomosed to the lower pole
renal pelvis or ureter, is recommended. This
procedure can be performed either at the level of
the kidney, with the removal of part of the
redundant distal ureter, or low, through an ingui-
nal incision. Total urinary tract reconstruction in
neonates and infants is not recommended
because of the high complication rate caused by
the small size of the infant’s bladder.
90.6.5 Posterior Urethral Valves
The most common cause of severe obstructive
uropathy in children is posterior urethral valves
(PUV), which are tissue leaflets fanning distally
from the prostatic urethra to the external urinary
sphincter (Fig.  90.7). The incidence of this
abnormality is 1:8000. Typically, the leaflets are
separated by a slit-like opening. Approximately
one-third ultimately develop chronic renal fail-
ure or severe renal insufficiency. Prognosis is
significantly better if the antenatal sonogram
before 24 weeks’ gestation was normal. In 1
study, 9 of 17 patients with PUV whose hydro-
nephrosis was discovered before 24 weeks’ ges-
tation developed renal failure, whereas only one
of 14 recognized after 24 weeks’ gestation
developed end-stage renal disease. Favorable
prognostic factors include a serum creatinine
level of less than 0.8–1.0  mg% after bladder
decompression, unilateral VUR into a nonfunc-
tioning kidney (‘VURD syndrome’), ascites, and
identification of the corticomedullary junction
on renal sonography. Early delivery of infants
with an antenatal diagnosis of suspected PUV is
not recommended, unless there is oligohydram-
J. S. Elder
nios. If severe bilateral renal dysplasia is present,
pulmonary hypoplasia is often also present, and
problems with ventilation may result. Initially, a
small feeding tube should be passed into the
bladder for urinary drainage until electrolyte
imbalances can be corrected. A Foley catheter is
not recommended because the balloon may
cause significant bladder spasm and impede
upper tract drainage. Care should be taken
­passing the catheter, as the prostatic urethra is
dilated and there is bladder neck hypertrophy;
the feeding tube may coil in the prostatic urethra
and not drain the bladder. In this setting, catheter
irrigation typically results in the fluid coming
out of the urethra next to the catheter. AVCUG
should be obtained to confirm the diagnosis, and
a renal scan should be performed to evaluate the
upper tract differential renal function. In new-
borns, alternative treatments include transure-
thral endoscopic ablation of PUV, cutaneous
vesicostomy, and high diversion (cutaneous
pyelostomy). The ideal initial treatment is valve
ablation with a small Bugbee electrode, as is
used with TUI, or the holmium:YAG laser. In
small neonates, the 8 or 9 Fr resectoscope may
be too large for the urethra, and a temporary ves-
icostomy may be necessary. A vesicostomy also
should be considered for those with a serum cre-
atinine level that remains significantly elevated
after bladder decompression. Cutaneous pyelos-
tomy rarely affords better drainage compared
with cutaneous vesicostomy and diverts urine
away from the bladder, which may prevent nor-
mal bladder growth. However, in selected cases
the Sober-­en-­T cutaneous ureterostomy is use-
ful. In this procedure, the upper ureter is brought
out to the abdomen and transected, and the distal
segment is anastomosed to the renal pelvis; this
option allows urine to drain both through the
ureterostomy, as well as to the bladder.
90.6.6 Vesicoureteral Reflux
Some neonates with medium- and high-grade
VUR are detected following the finding of ANH
(Figs.  90.8 and 90.9). Approximately 80% of
such patients are boys. In the most severe cases of
90  Management of Antenatal Hydronephrosis
a b
c d
Fig. 90.7  Male newborn with posterior urethral valves.
(a, b) Right (a) and left (b) enlarged kidneys with bilateral
grade 3–4 hydronephrosis with echogenic parenchyma.
The distal ureters were dilated. (c, d) VCUG shows severe
massive VUR, the bladder may also become dis-
tended from aberrant micturition into the upper
urinary tract. In the AUA Reflux Guidelines anal-
ysis, reflux-related renal scarring was present in
47.9% of those with grades IV–V VUR, but only
6.2% of those with grades I–II VUR. Consequently,
in neonates with grades III–V VUR, a DMSA
scan is recommended to determine whether
1265
trabeculated bladder with diverticula, dilated prostatic
urethra and grade V left vesicoureteral reflux. Patient
underwent transurethral ablation of posterior urethral
valves
reflux-related renal scarring is present. Initially,
neonates with prenatally diagnosed VUR are
managed medically. Most are placed on amoxi-
cillin prophylaxis for 2 months, followed by
nitrofurantoin or trimethoprim- sulfamethoxa-
zole prophylaxis, and circumcision is recom-
mended for male neonates to decrease the risk for
UTIs.
1266 J. S. Elder

Fig. 90.8  Male newborn with

antenatal and postnatal bilateral
hydronephrosis. VCUG shows
bilateral grade IV vesicoureteral
reflux. The urethra was normal

b c

Fig. 90.9  Male newborn with antenatal left hydrone- grade V reflux into lower pole, left kidney. (c) DMSA
phrosis in a duplex kidney. No ectopic ureter or uretero- renal scan, left kidney on left. Differential function left
cele. (a) Dilated left lower pole, normal upper pole. The 42% right 58%. Normal right kidney and left upper pole,
distal left lower pole ureter was dilated. (b) VCUG shows reduced function left lower pole
90  Management of Antenatal Hydronephrosis
Neonates with VUR are more likely to show
spontaneous resolution than are older children.
Indeed, 20–35% of ureters with grade IV or V
VUR have reflux resolution within 2 years; how-
ever, a significant proportion develop a break-
through UTI, and anti-reflux surgery is
recommended in these cases. VUR that occurs
only when the bladder is quite full or during void-
ing is more likely to resolve spontaneously the
reflux that appears during bladder filling. The
success rate for open surgical correction of VUR
in infants can be as high as in older children.
Another option is subureteral injection of dextra-
nomer microspheres/hyaluronic acid into the ure-
terovesical junction, in which the success rate is
70–80% with a single injection.
90.7 Conclusions
Approximately 1–2% of newborns have an ante-
natal diagnosis of hydronephrosis or significant
renal pelvic dilation. Hydronephrosis is often
caused by nonobstructive conditions. The likeli-
hood of significant urologic pathology is directly
related to the size of the fetal renal pelvis, and 90%
with an anteroposterior diameter >2 cm need sur-
gical intervention or long-term urologic follow-up.
Following delivery, antibiotic prophylaxis should
be administered if there is severe hydronephrosis
or hydroureteronephrosis. Circumcision should be
offered to affected boys. A renal sonogram is gen-
erally obtained at 3–4 weeks. A VCUG should be
obtained if VUR or posterior urethral valves is sus-
pected. If there is grade 3 or 4 hydronephrosis and
the VCUG is negative, a diuretic renogram is rec-
ommended. Pediatric urologic or pediatric nephro-
logic consultation is helpful in planning evaluation
and treatment. Antenatal recognition of hydrone-
phrosis allows postnatal diagnosis and treatment
of urologic pathology, preventing complications of
pyelonephritis and obstruction. In the past decade,
significant progress has been made in the develop-
ment of minimally invasive treatment options.
1267
References
Braga LH, Ruzhynky V, Pemberton J et  al (2014)
Evaluating practice patterns in postnatal manage-
ment of antenatal hydronephrosis: a national survey
of Canadian pediatric urologists and nephrologists.
Urology 83:909–914
Braga LH, Farrokhyar F, D’Cruz J et al (2015) Risk fac-
tors for febrile urinary tract infection in children with
prenatal hydronephrosis: a prospective study. J Urol
193:1766–1771
Braga LH, D’Cruz J, Rickard M et al (2016) The fate of
primary nonrefluxing megaureter: a prospective out-
come analysis of the rate of urinary tract infections,
surgical indications and time to resolution. J Urol
195:1300–1305
Chang A, Sivananthan D, Nataraja RM et  al (2018)
Evidence-based treatment of multicystic dysplastic
kidney: a systematic review. J Pediatr Urol 14:510–519
DiRenzo D, Persico A, DiNocola M et  al (2015)
Conservative management of primary non-refluxing
megaureter during the first year of life: a longitudinal
observational study. J Pediatr Urol 11:226.e1–6
Elder JS, Stansbrey R, Dahms BB et al (1995) Renal his-
tologic changes secondary to ureteropelvic junction
obstruction. J Urol 154:719–722
Herz D, Merguerian P, McQuiston L (2014) Continuous
antibiotic prophylaxis reduces the risk of febrile UTI
in children with asymptomatic antenatal hydronephro-
sis with either ureteral dilation, high-grade vesicoure-
teral reflux, or ureterovesical junction obstruction. J
Pediatr Urol 10:650–654
Hodhod A, Capolicchio JP, Jednak R et  al (2016)
Evaluation of urinary tract dilation classification
system for grading postnatal hydronephrosis. J Urol
195:725–730
Nguyen HT, Herndon CD, Cooper C et  al (2010) The
Society for Fetal Urology consensus statement on the
evaluation and management of antenatal hydronephro-
sis. J Pediatr Urol 6:212–231
Nguyen HT, Benson CB, Bromley B et  al (2014)
Multidisciplinary consensus on the classification of
prenatal and postnatal urinary tract dilation (UTD
classification system). J Pediatr Urol 10:982–998
Reuss A, Wladimiroff JW, Steward PA et  al (1988)
Noninvasive management of fetal obstructive uropa-
thy. Lancet 2:949
Sinha A, Bagga A, Krishna A et al (2013) Revised guide-
lines on management of antenatal hydronephrosis.
Indian J Nephrol 23:83–97
Zaraba P, Lorenzo AJ, Braga LH (2014) Risk factors for
febrile urinary tract infection in infants with prenatal
hydronephrosis: comprehensive single center analysis.
J Urol 191:1614–1618
 Upper Urinary Tract Obstructions
Leon Chertin and Boris Chertin
91.1 Pelviureteric Junction
Obstruction
91.1.1 Historical Overview
The detection of renal abnormalities during pre-
natal ultrasonography was first reported in the
beginning of the seventeenth century (Garrett
et al. 1970). Since then, the routine use of ultraso-
nography for the detection of congenital anoma-
lies has become a part of routine care during the
antenatal period. Currently, it is estimated that
genitourinary anomalies comprise nearly 20% of
all prenatally detected fetal anomalies. Amongst
these, hydronephrosis is one of the most com-
monly detected anomalies, seen in approximately
1–5% of all pregnancies and occurs due to vari-
ous causes. Thus, we have an increasing number
of patients who are presenting to the clinician
with a presumptive diagnosis, rather than a symp-
tom, and sometimes before they are even born.
Although the initial reports of surgical outcome
of the correction of the neonatal PUJ obstruction
L. Chertin
Sakler School of Medicine, Tel Aviv University,
Tel Aviv, Israel
B. Chertin (*)
Department of Urology, Hebrew University,
Jerusalem, Israel
e-mail: chertinb@szmc.org.il
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_91
91
were excellent (King et  al. 1984), following
observations regarding renal function preserva-
tion during conservative treatment have started a
new era in the treatment of antenatal hydrone-
phrosis (Ransley et  al. 1990). In our earlier
reports, we clearly demonstrated that prenatal
diagnosis of hydronephrosis with close follow-up
after delivery is much superior, in terms of renal
function preservation, compared to those chil-
dren who were diagnosed to have PUJ obstruc-
tion due to clinical symptoms (Chertin et  al.
1999). Multiple reports demonstrated that
approximately 30% of children will require sur-
gery during surveillance, therefore expectant
management will spare the majority of children
from surgery (Koff and Campbell 1992; Koff
2000; Ulman et  al. 2000; Onen et  al. 2002;
Chertin et al. 2002).
91.1.2 Incidence
The overall incidence of neonatal hydronephro-
sis, which leads to the diagnosis of PUJ obstruc-
tion, approximates 1  in 500 births. The ratio of
males to females is 2:1  in the neonatal period,
with left-sided lesions occurring in 60% of cases.
In the newborn period, a unilateral process is
most common, but bilateral PUJ obstruction was
found in 10–49% of neonates in some reported
series (Gokce et al. 2012).
1269
1270
91.1.3 Etiopathogenesis
PUJ obstruction is classified as intrinsic, extrin-
sic, or secondary. Intrinsic obstruction results
from the failure of transmission of the peristaltic
waves across the PUJ, with the failure of urine to
be propulsed from the renal pelvis into the ureter,
which results in multiple ineffective peristaltic
waves that eventually causes hydronephrosis by
incompletely emptying the pelvic contents.
Extrinsic mechanical factors include aberrant
renal vessels, bands, adventitial tissues, and
adhesions that cause angulation, kinking, or com-
pression of the PUJ.  Extrinsic obstruction may
occur alone but usually coexists with intrinsic
ureteropelvic junction pathology. Secondary PUJ
obstruction may develop as a consequence of
concomitant severe vesicoureteric reflux (VUR),
which occurs in 15–30% of children who have
ipsilateral PUJ obstruction, in which a tortuous
ureter may kink proximally.
91.1.4 Pathophysiology
Renal morphogenesis is a complex, temporally
and spatially regulated process by which precur-
sor cells develop into a structurally and function-
ally normal kidney. Abnormal or dysregulated
renal development results in a wide range of renal
abnormalities, collectively known as congenital
anomalies of the kidney and urinary tract
(CAKUT), which compose the most common
cause of end-stage renal disease (ESRD) in chil-
dren. In humans, kidney and urinary tract devel-
opment begins at approximately 3 weeks
gestation, with the formation of the initial urinary
excretory precursor, the pronephros, which
undergoes complete involution (Little and
McMahon 2012; Combes et al. 2015; Short and
Smyth 2016). By 34–36 weeks gestation, nephro-
genesis is complete, and the structural and func-
tional relationship of each nephron segment is
fully developed. On average, 1 million (range
0.2–2.7 million) individual nephrons in each kid-
ney arise from the embryonic precursor cells.
Concurrent with the morphogenesis of the kid-
ney, functional development of the fetal kidney
L. Chertin and B. Chertin
also progresses with increasing gestational age.
Prenatally, the placenta controls fluid and electro-
lyte homeostasis, and the primary function of the
fetal kidney is the production of urine to maintain
amniotic fluid volume. In the later stages of ges-
tation, urine output, tubular function, and glo-
merular filtration increase with gestational age.
After birth, the neonatal kidney undergoes
physiologic changes to adapt to the extrauterine
environment; rapid changes occur over the first
several weeks and continue until reaching adult
levels at 1–2 years of life. Changes in hydrostatic
pressure in the renal pelvis and individual neph-
rons are critical in determining the effects of
obstruction on Glomerular Filtration Rate (GFR).
Normal proximal intratubular pressure of
12 mmHg increases in direct correlation with ris-
ing intrapelvic pressure with a maximum of
approximately 40  mmHg. Higher intrapelvic
pressure of 50–70 mmHg is not transmitted to the
proximal tubules, perhaps because of compres-
sion of the renal papilla. Continuation of obstruc-
tion increases in preglomerular vascular
resistance (afferent arteriolar vasoconstriction),
which in turn could lead to the increased accumu-
lation of fibrotic tissue and appearance of renal
dysplasia. Furthermore, chronic partial obstruc-
tion leads to the gradual decrease in GFR and is
accompanied by an increase in the fractional
excretion of filtered sodium, indicating decreased
tubular reabsorption. Leaving the obstruction
untreated might lead to the irreversible tubular
and glomerular injury with eventual renal loss.
91.1.5 Diagnosis
91.1.5.1 Prenatal Diagnosis
The bladder is visualized by 14 weeks gestation.
The ureters are usually not seen in the absence of
distal obstruction or reflux. The fetal kidney may
be visualized at the same time as the bladder. If
not, they are always visualized by 16th weeks
gestation. However, it is not until 20–24 weeks
gestation, when the fetal kidney is surrounded by
fat, that the internal renal structures appear dis-
tinct. Renal growth can then be assessed easily.
Beyond 20 weeks, fetal urine production is the
91  Upper Urinary Tract Obstructions
main source of amniotic fluid. Therefore, major
abnormalities of the urinary tract may result in
oligohydramnios. Because of the distinct urine
tissue interface, hydronephrosis can be detected
as early as 16 weeks gestation. An obstructive
anomaly is recognized by demonstrating dilated
renal calyces and pelvis. A multitude of measure-
ments and different gestational age cut-off points
have been recommended in the assessment of
fetal obstructive uropathy.
Routine estimation of anteroposterior (AP)
diameter of renal pelvis in the fetus with
­hydronephrosis is considered a useful marker for
classification of renal dilatation and possible
obstruction. AP renal pelvis threshold values
ranged between 2.3 and 10 mm. Positive predic-
tive values for pathological dilatation confirmed
in the neonate ranged between 2.3 and >40% for
AP renal measurements of 2–3 mm and 10 mm,
respectively. One study, which included more
than 46,000 screening patients, published the
standards regarding renal pelvic measurement.
This study clearly demonstrated that only fetuses
exhibiting third-trimester AP renal pelvis dilata-
tions >10 mm would merit postnatal assessment.
In order to standardize postnatal evaluation of
prenatal hydronephrosis a grading system of
postnatal hydronephrosis was implemented in
1993 by the Society for Fetal Urology (SFU). In
SFU system, the status of calices is paramount
while the size of the pelvis is less important. In
SFU grading of hydronephrosis, there is no
hydronephrosis in Grade 0. At Grade 1, the renal
pelvis is only visualized. Grade 2 of hydrone-
phrosis is diagnosed when a few (but not all)
renal calices are identified in addition to the renal
pelvis. Grade 3 hydronephrosis requires that vir-
tually all calices are depicted. Grade 4 hydrone-
phrotic kidneys will exhibit similar caliceal status
with the involved kidney exhibiting parenchymal
thinning. Often this classification is applied also
on prenatal hydronephrosis. We have published
our data regarding prenatal findings with the spe-
cial emphasis on the natural history of hydrone-
phrosis during the postnatal period. Our data
shows that SFU grade of prenatal hydronephrosis
is not a significant predictive factor for surgery in
unilateral hydronephrosis. However, SFU Grades
1271
3–4 prenatal bilateral hydronephrosis indicates
that the majority of the children will require sur-
gical correction during the postnatal period.
In 2014, the “urinary tract dilation (UTD)”
classification system was introduced to replace
the SFU system and other grading systems. It
consists of 6 parameters, namely APD of the
renal pelvis, urinary tract dilation, parenchymal
thickness, parenchymal appearance, ureteral sta-
tus, and bladder status, furthermore distinguish-
ing whether these parameters are antenatal (“A”)
or postnatal (“P”). There are two antenatal and
three postnatal categories of risk: A1 (low risk) or
A2–3 (intermediate/high risk) for antenatal UTD;
and P1 (low), P2 (intermediate), P3 (high risk)
for postnatal UTD.  Persistent UTD A1 or UTD
A2 to A3 warrants postnatal evaluation (Nguyen
et al. 2014).
In the case of severe prenatal bilateral hydro-
nephrosis, severe hydroureteronephrosis, or
severe impairment of the solitary kidney, fetal
bladder aspiration for urinary proteins and elec-
trolytes is recommended from 17 weeks of gesta-
tion in some reports in order to predict the renal
injury secondary to obstructive uropathy. Fetal
urinary sodium level less than 100 mmol/L, chlo-
ride level of less than 90 mmol/L and an osmolal-
ity of less than 210 mOsm/kg are considered as
prognostic features for good renal function.
91.1.5.2 Clinical Presentation
The most common presentation is abdominal
flank mass. Fifty percent of abdominal masses in
newborns are of renal origin with 40% being sec-
ondary to PUJ obstruction. Other clinical presen-
tations include urinary tract infection irritability,
vomiting, and failure to thrive. Ten to 35% of
PUJ obstructions are bilateral and associated
abnormalities of the urinary tract are seen in
about 30%. PUJ problems are often associated
with other congenital anomalies, including
imperforated anus, contralateral dysplastic kid-
ney, congenital heart disease, VATER syndrome,
and esophageal atresia.
91.1.5.3 Differential Diagnosis
With the increasing number of antenatally diag-
nosed hydronephrosis it is difficult to interpret
1272 L. Chertin and B. Chertin

the underlying pathology and its significance.
Severe obstructive uropathies are detrimental to
renal function. However, on the other hand,
hydronephrosis without ureteral or lower tract
anomaly is common. The important aspect of
postnatal investigations is to identify the group of
patients who will benefit from early intervention
and those who need to be carefully followed.
Ultrasound: Follow-up ultrasound examina-
tion is necessary in the postnatal period in ante-
natally detected hydronephrosis. If bilateral
hydronephrosis is diagnosed in utero in a male
infant, postnatal evaluation should be carried out
within 24 h primarily because of the possibility
of posterior urethral valves. If the ultrasound scan
is negative in the first 24–48 h in any patient with
unilateral or bilateral hydronephrosis, a repeat
scan should be performed after 5–10 days, recog-
nizing that neonatal oliguria may mask a moder-
ately obstructive lesion.
If hydronephrosis is confirmed on the postna-
tal scan, further careful scan of the kidney, ureter,
bladder, and in boys, the posterior urethra is
essential (Fig. 91.1a).

Fig. 91.1 (a) A sagittal

a
plane scan through the
obstructed right kidney
confirms obstruction at
the level of the
pelviureteric junction.
(b) 99Tc MAG3 scan in
the above patient.
Clearance curve for right
kidney confirming the
high-grade obstruction
on this side

b
91  Upper Urinary Tract Obstructions
Radionuclide Scans: DTPA is completely fil-
tered by the kidneys at maximum concentration
of 5% being reached in 5  min, falling to 2% at
15 min. Recently, it has been reported that use of
tracers that rely on tubular extraction such as 123
I-Hippuran and 99Tc MAG3 (Fig.  91.1b) may
improve diagnostic accuracy. The kidney of the
young infant is immature; renal clearance, even
when corrected for body surface, progressively
increases until approximately 2 years of age.
Therefore, the renal uptake of tracer is particu-
larly low in infants, and there is a high back-
ground activity. Thus, the traces with a high
extraction rate, such as 123 I-Hippuran and 99Tc
MAG3, provide reasonable images enabling the
estimation of the differential kidney function dur-
ing the first few weeks of life. It is also helpful in
assessing the size, shape, location, and function
of the kidney. Diuretic augmented renogram is a
provocative test and is intended to demonstrate or
exclude obstructive hydronephrosis by stressing
an upper urinary tract with a high urine flow.
Obstruction usually is defined as a failure of
tracer washout after diuretic stimulation. If
unequivocal, it eliminates the need for further
investigations. In equivocal cases, F15 in which
furosemide is given 15  min before the test pro-
vides a better assessment of the drainage of upper
urinary tract. Forced hydration prior to a scan
increases THE predictive value of non-obstructed
patterns by up to 94%. Since glomerular filtration
and glomerular blood flow are still low in the
newborn, the handling of isotype is unpredictable
and can be misleading.
Functional magnetic resonance urography
(MRU) has been recently proposed by many
study groups as an alternative technique to evalu-
ate the drainage curve and split renal function
(SRF) in obstructive uropathy. This method
allows the precise understanding of the kidney
anatomy while providing information regarding
renal functioning without radiation exposure
obviating the need to use contrast media. When
surgical correction is planned, MRU aids in
clearly identifying anatomically crossing vessels
and obstructive pathology.
Pressure-Flow Study: In the equivocal cases
and in the presence of impaired function, the
1273
pressure-flow study (Whitaker Test) and ante-
grade pyelography may be necessary to confirm
or exclude obstruction. Whitaker Test is based on
the hypothesis that if the dilated upper urinary
tract can transport 10 ml/min without an inordi-
nate increase in pressure, the hydrostatic pressure
under physiological conditions should not cause
impairment of renal function and the degree of
obstruction if present is insignificant. However, it
is an invasive test and is seldom required.
Antegrade pyelography may be performed with
ultrasound guidance in patients where diagnosis
is difficult. Retrograde pyelography is seldom
required to determine the status of ureters. The
disadvantages include difficulty in ureteral cath-
eterization in neonates, trauma, and edema that
may change partial obstruction to the complete
one. In patients where diagnosis is equivocal,
serial examinations may be necessary. Routine
use of micturating cystourethrogram (MCUG) in
patients with antenatal unilateral hydronephrosis
is controversial. Some authors advocate regular
use of MCUG as a part of postnatal evaluation
citing 15–30% of incidence of concomitant VUR
either uni- or contralateral. Others recommend
only performing MCUG in patients with SFU Gr
III and IV hydronephrosis. We have abounded to
perform MCUG routinely in children with unilat-
eral antenatal hydronephrosis based on the fact
that even if the reflux exists usually it is of low
grade and does not require any treatment. We
reserve MCUG only for patients with bilateral
hydronephrosis or for those whose ureter was
seen at any stage of antenatal or postnatal
follow-up.
91.1.6 Management
A considerable controversy exists regarding the
management of newborn urinary tract obstruc-
tions. Some authors advocate early surgical
intervention to prevent damage to maturing
nephrons, while others feel that early surgery
carries no specific benefit. During late prenatal
and early postnatal life, there is progressive
increase in glomerular filtration rate.
Additionally, this transition is associated with an
1274
abrupt decline in urine output from what appears
to be a quite high in utero output to a rather low
early neonatal level of urine production. These
physiological observations may explain the com-
mon observation of hydronephrosis detected
antenatally, which on postnatal follow-up reverts
to an unobstructed pattern. Surgery is usually
undertaken in infants whose renal function dete-
riorates during the observation period. We have
analyzed our database of 343 children (260
males and 83 females) with an antenatal diagno-
sis of hydronephrosis, which led to the postnatal
diagnosis of PUJ obstruction, who were deliber-
ately followed up conservatively at our depart-
ment over a 16-year period, in order to define
which factors lead to surgery (Chertin et  al.
2006). One hundred and seventy-nine children
(52.2%) required surgical correction in the
course of conservative management. Average
age at surgery was 10.6 months (range 1 month–7
years). Of these, 50% underwent surgery during
the first 2 years of life and majority of the
remaining patients underwent surgery between
the age of 2 and 4. Only two patients required
surgery later on. Univariate analysis revealed
that child sex, side of hydronephrosis are not sig-
nificant predictive factors for surgery. However,
SFU Grades 3–4 of postnatal hydronephrosis
(p  <  0.0001, Odds ratio 0.06281), RRF <40%
(p < 0.0001, Odds ratio 0.1022) were significant
independent risk factors for surgery. A number
of different operations have been described for
surgical correction of PUJ obstruction. The clas-
sical traditional approach is an extraperitoneal
approach via lateral flank incision. In cases
where suspicion of an obstruction distal to the
PUJ arise, antegrade or retrograde study of the
ureter upon or during surgery are recommended.
In some cases, posterior lumbotomy may be
applied. The use of muscle splitting rather than
muscle cutting makes it almost a minimally inva-
sive procedure. The location of the incision just
under and parallel to the 12th rib has a cosmetic
advantage. The bilateral procedure is possible if
indicated under the same anesthesia without
position changes. This approach should not be
used in older children or significantly obese.
L. Chertin and B. Chertin
The various techniques of pyeloplasty are
divided into dismembered and non-dismembered
pyeloplasty. The basic principle of these opera-
tions is an excision of the PUJ, with a subsequent
oval-shaped anastomosis between the ureter and
lower part of the pelvis. Different types of stents
are placed for drainage usually for 6 weeks. The
most popular are Double J Paediatric Stents or
Pipi-Salle Stent nephrostomy (Cook, USA).
In the last decade, minimal invasive approaches
have replaced the classic gold standard open dis-
membered Anderson-Hynes pyeloplasty. Since it
was first described (Schuessler et al. 1993), lapa-
roscopic pyeloplasty has developed into a suc-
cessful approach. Introduction of robotic
technology has helped overcome some of the
limitations of complex laparoscopic procedures
such as offering magnified three-dimensional
vision, tremor reduction, motion scaling,
extended range of motion for the surgical arms,
and better ergonomics (Andolfi et  al. 2020).
These features render robotic surgery ideal for
many reconstructive and ablative procedures.
Robot-assisted laparoscopic pyeloplasty (RALP)
can be performed by a trans or retroperitoneal
approach. Suturing is done with a 6-0 monofila-
ment absorbable suture, but one can utilize any
5-0 or 6-0 suture depending on the size of the
patient. Currently, it appears that nothing larger
than 6-0 for small children and infants is recom-
mended. RALP is associated with excellent suc-
cess rates, noninferior to those with the open
approach, without increased complications
(Kafka et  al. 2019). Recent studies have found
that RALP decreased length of hospitalization,
decreased postoperative narcotic requirement,
improved cosmetic results, and decreased human
capital losses to the parents.
Although antegrade and retrograde endopy-
elotomy has been shown to be effective in chil-
dren, this approach has not been recommended in
neonates, infants, or young children. However, it
should be considered in older children or in those
with failed primary dismembered pyeloplasty.
Bilateral Pelviureteric Obstruction: Surgical
correction of the symptomatic side or side with
better function should take precedence. If a
91  Upper Urinary Tract Obstructions
nephrectomy is considered on one side, the
pyeloplasty should precede this.
Postoperative Complications: infection, adhe-
sive obstruction (transperitoneal approach), tem-
porary obstruction at the anastomosis resulting in
excessive urine leakage and failures due to post-
operative stricture at anastomotic sites. An over-
all reoperation rate of 8.2% was reported in the
early series. However, in the latest series, when
temporally double-J stents were utilized, the
reoperation rate was negligible.
Follow-up and Results: Follow-up ultrasound
may be performed 3–6 months after operation
when maximum improvement can be seen. A
follow-up radionuclide scan should be done 6–8
months following pyeloplasty, in order to evalu-
ate an improvement in the renal function and
drainage. Pyeloplasty in the neonatal period
when indicated gives excellent results. Moreover,
successful pyeloplasty after the prenatal diagno-
sis of PUJ obstruction is associated with improved
renal function throughout puberty even in chil-
dren with initial poor renal function.
Conclusions: The majority of children with
antenatal diagnosed hydronephrosis which led to
the diagnosis of PUJ obstruction may and should
be followed conservatively. When surgery is indi-
cated, SFU 3&4 Grade of hydronephrosis and
initial renal function less than 40% seems to
serve as a positive predictive factor for surgical
intervention. Utilization of RALP is increasing
rapidly, suggesting that the robotic approach may
be the new gold standard for minimally invasive
pyeloplasty. RALP is associated with excellent
success rates compared with the open approach
without increased complications.
91.2 Megaureter, Ureterovesical
Junction Obstruction
91.2.1 Historical Overview
Megaureter involves a ureter that is dilated out of
proportion to the rest of the urinary tract and
above the norms. Originally coined by Caulk in
1923 as megaloureter, other synonyms in use
include wide ureter and hydroureter. Cussen
1275
(1971) and later Hellstrom et  al. in 1985 have
established the normal measurement of the ure-
teral diameter in infants and children from 30
weeks of gestation to 12 years of age. Normal
ureteral diameter in children is rarely greater than
5 mm, and ureters larger than 7 mm can be con-
sidered megaureters.
Classification: The Paediatric Urology Society
in 1976 adopted a standard nomenclature for cat-
egorizing megaureters, which is a useful guide
for management (Smith et  al. 1977). There are
three types described:
1. Refluxing ureter which may be primary or
secondary to distal obstruction or pathology
2. Obstructive: which may be primary and
include intrinsic obstruction, or secondary
due to distal obstruction or extrinsic causes.
3. Non-refluxing, non-obstructed which may be
primary-idiopathic type or secondary to dia-
betes insipidus or infection.
In 1980, King subsequently modified this
classification by adding a fourth group consisting
of the refluxing, obstructed megaureters (King
1980).
91.2.2 Incidence
The true incidence of primary obstructive mega-
ureter (POM) is not known but it is considered to
be a cause of 10–23% of antenatally detected
upper urinary tract dilations (Gokce et al. 2012).
Primary obstructive megaureter is more common
in males than females, and the left ureter is more
likely to be involved than the right. Seventeen to
34% patients have bilateral megaureters, which is
associated with contralateral dysplasia or obstruc-
tion in 10–15% of cases. Most POMs are detected
by antenatal ultrasound screening with the vast
majority being asymptomatic.
91.2.3 Etiopathogenesis
The most common etiologic causes of obstruc-
tive megeureters are: (1) alteration in muscular
1276
orientation, (2) muscular hypoplasia with fibro-
sis or excessive collagen deposition resulting in
a discontinuity of muscular coordination, and
finally (3) disturbance in the electric syncytium
along with the nexus injury causing pathological
innervation. Prenatal Diagnosis: Currently the
vast majority of obstructive megaureters are dis-
covered on prenatal ultrasound. Usually ureter is
not seen in fetal scans. Visualization of dilated
ureter to the level of vesicoureteric junction
without abnormal bladder may suggest obstruc-
tion or reflux. However, this may be a transient
phenomenon. Fetal urine flow is 4–6 times
greater before birth than after and is due to dif-
ferences in renal vascular resistance, glomerular
filtration, and concentrating ability. This high
outflow contributes to ureteral dilatation.
Another contributing factor includes increased
compliance of the fetal ureter.
91.2.4 Pathophysiology
The pathophysiology of primary obstructed
megaureters (POMs) has not been fully under-
standable yet. However, there is general agree-
ment that the presence of an adynamic distal
ureteral segment is the most common cause of
primary obstructive megaureter. The presence of
narrowed terminal portion of ureter will not con-
vey the peristaltic wave or dilate enough to permit
free passage of urine. This results in excess
boluses of urine which coalesce and cause ure-
teral dilatation. The contraction waves become
smaller and are unable to coopt the walls of dilated
ureters. This along with infection could damage
the renal parenchyma and could lead to irrevers-
ible renal damage similar to PUJ obstruction.
91.2.5 Diagnosis
91.2.5.1 Prenatal Diagnosis
The most intriguing question on prenatal evalua-
tion is which children will require follow-up only
and who will need surgery in order to rescue the
renal function (Farrugia et al. 2014). In our data-
L. Chertin and B. Chertin
base, we have analyzed 79 children with antena-
tal diagnosis of obstructive megaureter (Chertin
et  al. 2008). Antenatal SFU grade of hydrone-
phrosis had no predictive value for the surgery in
postnatal follow-up. However, those children
who had ureteric diameter of more than 1.4 cm
are more likely to require surgery postnatally.
91.2.5.2 Differential Diagnosis
Antenatally diagnosed ureteral dilatation needs
further evaluation to confirm or exclude obstruc-
tion, reflux, or both. The clinician should con-
front the common clinical dilemma in pediatric
antenatal hydronephrosis and must distinguish
between those patients who will have deteriora-
tion in renal function while on surveillance proto-
col and therefore will benefit from early surgery
and those who have nonobstructive hydronephro-
sis. In antenatally detected cases, ultrasonogra-
phy should be performed between 3 and 5 days
after birth to confirm antenatal findings. If no
dilatation is seen, a repeat ultrasound should be
performed after a few weeks as neonatal oliguria
can mask dilatation. If dilatation persists on a
repeat ultrasound, radionuclide scan is performed
at the age of 4–6 weeks. In male infants with
bilateral ureterohydronephrosis, MCUG should
be performed for 72 h since birth to rule out pos-
terior urethral valve. As stated previously, MRU
has the ability to provide detailed information
regarding anatomy, as well as renal function and
drainage in a single study without the use of ion-
izing radiation.
Ultrasonography classically shows hydroure-
ter and variable hydronephrosis, with hyperperi-
stalsis of a lower ureter that terminates shortly
above the bladder in a narrow, adynamic segment
(Fig. 91.2a). However, the narrow segment may
not always be visualized and therefore MCUG is
necessary to exclude VUR.  Contrast-enhanced
voiding urosonography (CEVUS) is a dynamic
imaging technique that enables the morphologic
and functional evaluation of the entire urinary
tract by introducing an ultrasonographic contrast
agent into the bladder. CEVUS is most often
indicated to detect VUR.  It is highly sensitive
while avoiding exposure to ionizing radiation.
91  Upper Urinary Tract Obstructions
a 91.2.6 Management
b vative follow-up (Chertin et al. 2008). Univariant
Fig. 91.2 (a) Longitudinal renal scan demonstrates
severe left ureterohydronephrosis. (b) Intravenous uro-
gram shows severe left ureterohydronephrosis
Intravenous Urography (IVP) may be neces-
sary in equivocal cases to establish the diagnosis
(Fig.  91.2b). It delineates the anatomy showing
dilated, obstructed ureter. However, it is better to
wait for a few weeks for renal maturation to allow
concentration of contrast reliability. Occasionally,
Whitaker Test and antegrade pyelography may be
required to establish the diagnosis.
1277
It is being increasingly recognized that many
antenatal and neonatal ureteral dilatations improve
with time. Surgery is indicated in patients with
progressive ureteral dilatation and deterioration in
renal function (Di Renzo et al. 2013). In our series
of the 79 children with antenatal diagnosis of
hydronephrosis who led to postnatal confirmation
of UJL obstruction, only 25 (31%) children
required surgical correction over 16-year conser-
analysis did not reveal statistical significance in
those children who required surgical correction
and in those who followed conservatively neither
from the side of obstruction nor sex of the patient.
However, those children who required surgery
had ureteric diameter more than 1.4  cm, renal
function less than 30%, and SFU Grades 3 and 4
of hydronephrosis.
Operation: There are various techniques of
reimplanting the ureter in a non-refluxing manner
after excision of adynamic, narrow segment. The
initial approach to the ureter can be either intra-
vesical, extravesical or combined. The most com-
monly used techniques for intravesical approach
are Cohen’s transtrigonal reimplantation and
Politano Leadbetter operation. The basic princi-
ple of these operations is an excision of ady-
namic, obstructive terminal narrow portion of the
ureter and reimplantation of the remaining ureter
into the bladder in an anti-refluxing fashion. In
those cases, a significant dilated tapering of the
ureter is required.
Non-excisional techniques such as Folding
(Kalikinski) or Plication (Star) of ureteral wall
are also available. Although these techniques
have the advantage of avoiding a suture line with
potential urinary leakage, they are inappropriate
for very dilated ureter as it reduces diameter by
only 50% and in neonates it can become too
bulky for the tunnel. In these cases, excisional
tapering technique utilizing Hendren clamps
where part of the ureteral wall is excised by using
a knife and scissors is recommended.
1278
Robot-assisted laparoscopic (RAL) surgery is
a safe, minimally invasive technique that has
become more widely used in pediatric urology
over recent decades (Boysen et al. 2018; Andolfi
et al. 2020).
With several advantages over standard lapa-
roscopy, robotic surgery is particularly well-­
suited to reconstructive surgery involving delicate
structures like the ureter. A robotic approach pro-
vides excellent access to and visualization of the
ureter at all levels. We have recently published
our experience with a novel surgical technique in
which we performed robotic-assisted laparo-
scopic dismembered extravesical cross trigonal
ureteral reimplantation (RADECUR). When
reviewing the literature and in comparison to
open and laparoscopic reimplantation for obstruc-
tive megaureter, success rate, operating time, and
complications are comparable when considering
RADECUR (Neheman et al. 2020).
In newborns and children younger than 1 year,
ureter reimplantation could be very technically
demanding and potentially threatening for blad-
der functional development, with high complica-
tion rate. With the advent of minimally invasive
surgery back in 1998, the first report of endo-
scopic balloon dilatation for POM in children was
published (Desgrandchamps 2001). Advantages
of the endoscopic approach include no surgical
incision, no violation of the bladder, no manipula-
tion of the distal ureteral blood supply, and no
need for prolonged catheterization. Endoscopic
techniques include the use of guidewires, balloon
catheters, and incisional uretrotomy.
However, the recent meta-analysis of the endo-
scopic management for persistent or progressive
POM in children older than 12 months of age dem-
onstrated modest success rates. In infants, it may
best be utilized as a temporizing procedure.
Approximately one-third of patients require surgi-
cal reintervention (Doudt et al. 2018).
91.2.7 Postoperative Course
In cases of the intravesical approach, the drain is
removed after 24–48 h. Stents are removed after
7–10 days followed by suprapubic catheter.
Ureteral reimplantation using extravesical
L. Chertin and B. Chertin
approach can be performed without stenting. An
indwelling urethral catheter is placed usually for
3–5 days to avoid urinary retention. Recently, we
have started to remove the indwelling catheter on
the following day after robotic surgery. Also, dur-
ing open surgery while utilizing intravasical
approach, there is no need to leave behind the
drains and suprapubic catheter. We have learned
that placement of JJ stents for the period of 2
weeks with indwelling urethral catheter for 3
days is safe and efficient and could significantly
shorten hospital admission without compromis-
ing child safety and success of the surgery.
Complications: Wound infection, vesicoure-
teral reflux due to short tunnel with no effective
flap valve mechanism, or obstruction due to a
fibrotic distal end secondary to ischemia espe-
cially in children who underwent excisional type
of ureteric remodeling.
Follow-up and Results: ultrasound is performed
3 months following surgery with radiologic stud-
ies at 6 months after repair in order to assess renal
function and drainage. MCUG is performed at the
time of radionuclide scan to rule out denovo reflux
into reimplanted previously obstructed ureter. We
recently published our data on long-term follow-
up of renal function following ureter reimplanta-
tion. We demonstrated that successful ureteral
reimplantation following antenatal diagnosis of
ureterovesical junction (UVJ) obstruction showed
an increase in relative renal function not only dur-
ing short- and midterm follow-up, but also allows
preserving the renal function throughout the
puberty period (Neeman et al. 2020).
91.3 Conclusions
The majority of the children with antenatal diag-
nosed primary obstructive megaureters do not
require surgical correction. When surgical inter-
vention is required it appear that initial renal
function <30%, ureteral diameter >1.2  cm, and
SFU 3–4 are positive predictive factors for sur-
gery. Surgery when indicated improves and
maintains renal function in the long-term follow-
­up. Recent data demonstrate that robotic reim-
plantation is an emerging modality for UVJ
obstruction correction.
91  Upper Urinary Tract Obstructions 1279

References Kafka IZ, Kocherov S, Jaber J, Chertin B (2019) Pediatric

robotic-assisted laparoscopic pyeloplasty (RALP):
does weight matter? Pediatr Surg Int 35(3):391–396
Andolfi C, Adamic B, Oommen J, Gundeti MS (2020)
King LR (1980) Megaloureter: Definition, diagnosis and
Robot-assisted laparoscopic pyeloplasty in infants and
management. J Urol 123:222–223
children: is it superior to conventional laparoscopy?
King LR, Coughlin PW, Bloch EC, Bowie JD, Ansong K,
World J Urol 38(8):1827–1833
Hanna MK (1984) The case for immediate pyeloplasty
Boysen WR, Akhavan A, Ko J, Lendvay TS, Huang J,
in the neonate with ureteropelvic junction obstruction.
Garcia-Roig M et  al (2018) Prospective multicenter
J Urol 132:725–728
study on robot-assisted laparoscopic extravesical ure-
Koff SA (2000) Postnatal management of antenatal hydro-
teral reimplantation (RALUR-EV): Outcomes and
nephrosis using an observation approach. Urology
complications. J Pediatr Urol 14(3):262.e1–262.e6
55:609–612
Chertin B, Fridmans A, Knizhnik M, Hadas-Halpern I,
Koff SA, Campbell K (1992) Nonoperative management
Hain D, Farkas A (1999) Does early detection of ure-
of unilateral neonatal hydronephrosis: natural history
teroplevic junction obstruction improve surgical out-
of poorly functioning kidneys. J Urol 152:593–597
come in terms of renal function? J Urol 162:1037–1041
Little MH, McMahon AP (2012) Mammalian kidney
Chertin B, Rolle U, Farkas A, Puri P (2002) Does delay-
development: principles, progress, and projections.
ing pyeloplasty affect renal function in children? BJU
Cold Spring Harb Perspect Biol 4(5):a008300
Int 90(1):72–75
Neeman BB, Jaber J, Kocherov S, Farkas A, Chertin B
Chertin B, Pollack A, Koulikov D, Rabinowitz R, Hain
(2020) Does renal function remain stable after puberty
D, Hadas-Halpren I et  al (2006) Conservative treat-
in children who underwent ureteral reimplantation due
ment of uretero-pelvic junction obstruction in chil-
to ureterovesical junction obstruction? Eur J Pediatr
dren with antenatal diagnosis of hydronephrosis:
Surg. https://doi.org/10.1055/s-­0040-­1712172
Lessons learned after 16 years of follow up. Eur Urol
Neheman A, Kord E, Koucherov S, Kafka I, Gaber J, Noh
49(4):734–739
PH, Zisman A, Chertin B (2020) A novel surgical
Chertin B, Pollack A, Koulikov RR, Shen O, Hain D et al
technique for obstructed megaureter: robot-assisted
(2008) Long-term follow up of antenatal diagnosed
laparoscopic dismembered extravesical cross-­trigonal
megaureters. J Ped Urol 4:188–191
ureteral reimplantation-short-term assessment. J
Combes AN, Davies JA, Little MH (2015) Cell-cell inter-
Endourol 34(3):249–254
actions driving kidney morphogenesis. Curr Top Dev
Nguyen HT, Benson CB, Bromley B, Campbell JB, Chow
Biol 112:467–508
J, Coleman B, Cooper C et al (2014) Multidisciplinary
Cussen LJ (1971) The morphology of congenital dilata-
consensus on the classification of prenatal and postna-
tion of ureter: intrinsic ureteral lesions. Aust NZJ Surg
tal urinary tract dilation (UTD classification system). J
41:185–194
Pediatr Urol 10(6):982–998
Desgrandchamps F (2001) Endoscopic and surgical repair
Onen A, Jayanthi VR, Koff SA (2002) Long-term fol-
of the ureter. Curr Opin Urol 11:271–274
lowup of prenatally detected severe bilateral newborn
Di Renzo D, Aguiar L, Cascini V, Di Nicola M, McCarten
hydronephrosis initially managed nonoperatively. J
KM, Ellsworth PI et  al (2013) Long-term fol-
Urol 168(3):1118–1120
low up of primary nonrefluxing megaureter. J Urol
Ransley PG, Dhillon HK, Gordon I, Duffy PG, Dillon
190(3):1021–1026
MJ, Barratt TM (1990) The postnatal management of
Doudt AD, Pusateri CR, Christman MS (2018)
hydronephrosis diagnosed by prenatal ultrasound. J
Endoscopic management of primary obstruc-
Urol 144:584–587
tive megaureter: a systematic review. J Endourol
Schuessler WW, Grune MT, Tecuanhuey LV, Preminger
32(6):482–448
GM (1993) Laparoscopic dismembered pyeloplasty. J
Farrugia MK, Hitchcock R, Radford A, Burki T, Robb
Urol 150(6):1795–1799
A, Murphy F (2014) British Association of Paediatric
Short KM, Smyth IM (2016) The contribution of branch-
Urologists consensus statement on the management
ing morphogenesis to kidney development and dis-
of the primary obstructive megaureter. J Pediatr Urol
ease. Nat Rev Nephrol 12(12):754–767
10(1):26–33
Smith ED, Cussen LJ, Glenn JF, Hendren WH, Jeffs RD,
Garrett WJ, Grunwald G, Robinson DE (1970) Prenatal
Johnson JH et al (1977) Report of the working party
diagnosis of fetal polycystic kidney by ultrasound.
to establish an international nomenclature for the large
Aust N Z J Obstet Gynaecol 10:7–9
ureter. Birth Defects Orig Artic Ser 13:3–8
Gokce I, Biyikli N, Tugtepe H, Tarcan T, Alpay H (2012)
Ulman I, Jayanthi VR, Koff SA (2000) The long-term
Clinical spectrum of antenatally detected urinary tract
follow up of newborns with severe unilateral hydro-
abnormalities with respect to hydronephrosis at post-
nephrosis initially treated nonoperatively. J Urol
natal ultrasound scan. Pediatr Surg Int 28(5):543–552
164:1101–1105
Hellstrom M, Hajlmas K, Jacobsson B, Jodal U, Odén A
(1985) Normal ureteral diameter in infancy and child-
hood. Acta Radiol 26:433–435
 Ureteric Duplication Anomalies
Thambipillai Sri Paran and Prem Puri
92.1 Introduction
Incomplete and complete duplication anomalies
of the upper urinary tract are estimated to be
present in up to 0.8% of the normal population.
The majority of these anomalies have normally
developed renal moieties and cause no functional
problems. When detected during antenatal scans,
it is of paramount importance that the clinician
keeps in mind that the vast majority of these
anomalies are benign, incidental findings.
Overall, ureteric duplications are more com-
mon in females than in males. In 15% of
patients, the ureteric duplication is bilateral
(Privett et al. 1976). There is an increased inci-
dence of up to 12% within some families, sug-
T. S. Paran (*)
Children’s Health Ireland at Crumlin, Dublin, Ireland
and Trinity College Dublin and University College
Dublin, Dublin, Ireland
P. Puri
Department of Pediatric Surgery, Beacon Hospital,
and University College Dublin, Dublin, Ireland
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_92
92
gestive of an autosomal dominant inheritance
(Atwell et al. 1974).
However, a percentage of ureteric duplica-
tions are associated with ureterocoele, ectopic
ureter, VUR, and pelviureteric junction obstruc-
tion (PUJO). One or more of these abnormalities
are frequently found in the presence of a compli-
cated complete duplication (Fig.  92.2), where
two separate ureters are seen to insert separately
into the bladder or an ectopic opening. In incom-
plete duplication (Fig. 92.1), the two bifid ureters
may join at any level from the ureteropelvic junc-
tion to the bladder and open as a single entity into
the bladder; though rare, these can also be associ-
ated with VUR and PUJO as much as with a non-­
duplicated ureter.
1281
1282 T. S. Paran and P. Puri

anomaly are VUR and PUJO. The management

of these complications will be discussed later in
this chapter.

92.3 Complete Duplication

This anomaly is thought to result when two ure-

teral buds arise from the Wolffian duct instead of
one (Campbell and Walsh 1992). The ureteral
bud that gives rise to the upper pole ureter is more
closely associated with the Wolffian duct and is
carried medially and caudally along with the
Wolffian duct (Rasouly and Lu 2013). This is
thought to result in the upper pole ureter opening
more medially and inferiorly than the lower pole
ureter into the bladder (Weigert-Mayer law;
Fig. 92.2). Sometimes, this upper pole ureter has

Fig. 92.1  Incomplete duplication. The two ureters unite

well above the bladder and open into the bladder with a
single opening

92.2 Incomplete Duplication

At 5 weeks of gestation, the ureteric bud appears

from the Wolffian duct. Premature division of a
single ureteral bud before it reaches the meta-
nephric blastema is thought to result in incom-
plete duplication of the ureter (Kozlov and
Schedl 2020). The premature division may
occur at variable distance from the kidney, but
there is only one ureteric orifice present on the
affected side (Fig. 92.1). Incomplete duplication
is three times more common than complete
duplication. However, cadaveric studies have
estimated the incidence to be even higher than Fig. 92.2  Complete duplication with intravesical ure-
terocoele associated with the upper pole ureter. The lower
current estimates (Arumugam 2020). The com- pole normal ureter inserts superiorly and laterally with a
monest complications seen with this type of short intramural tunnel
92  Ureteric Duplication Anomalies 1283

an abnormally prolonged or close attachment to

the Wolffian duct, which may result in it opening
at an ectopic location such as the urethra, seminal
vesicles, vas deferens, or even epididymis.
The upper pole ureters are associated with
ureterocoeles and ectopic insertions, while the
lower pole ureters are associated with VUR and
PUJO. These pathologies and their clinical mani-
festations, together with their management, are
discussed below.

92.4 Investigations

92.4.1 Renal Ultrasound

Following on from either antenatal scan or clini-

cal history suggestive of renal pathology, a renal
ultrasound is the first line of investigation. It may
identify ureteric duplication and associated renal
pelvic dilatation or ureteric dilatation clearly.
Usually, the kidney in ureteric duplication is
larger in size than a normal kidney, and the upper
pole has fewer numbers of calyces (usually one Fig. 92.3 Voiding cystourethrogram showing gross
third of the total). When VUR is present, lower reflux into the lower pole ureter (Drooping Lilly sign)
pole hydronephrosis and ureteric dilatation will
be seen. When the renal parenchyma is This is the study of choice to demonstrate
hypoechoic, then renal dysplasia should be sus- VUR and its severity (Fig. 92.3). It is important
pected. Absence of ureteric jet on ultrasound in to appreciate that an ectopic ureter, opening at the
conjunction with VUR on voiding cystourethro- bladder neck, can show reflux and have features
gram prompts to the diagnosis of refluxing type of suggestive of obstruction as well. It may be nec-
obstructed megaureter (Santhalia et  al. 2018). essary to repeat the study to demonstrate this
Ultrasound can also identify a ureterocoele well. complex phenomenon (Santhalia et al. 2018).
Ectopic dilated ureters may mimic a ureterocoele
during ultrasound study.
92.4.3 Intravenous Pyelogram (IVP)

92.4.2 Voiding Cystourethrogram
(VCUG)
This study is best performed with a feeding tube
rather than with a foley catheter, which has a bal-
loon that may confuse the diagnosis of an ure-
terocoele. The bladder should not be overfilled,
as it may compress the ureterocoele. It is also
important to take an oblique view to demonstrate
the ureterocoele, which is posteriorly placed.
Though the use of this investigation has become
very infrequent in pediatric urology with the
introduction of CT and MR urograms, when
confronted with a dilemma of demonstrating dif-
ficult ectopic ureters, IVP still has a role to play.
During the latter stages of this study, when the
bladder is filled with contrast, an ureterocoele
could also be clearly demonstrated. A “drooping
lily sign,” which is the inferior and lateral dis-
placement of lower renal moiety by the nonfunc-
1284
tioning upper renal moiety, is a good sign of
occult duplex kidney with an ectopic ureter
(Figs. 92.4 and 92.5).
Fig. 92.4 Intravenous pyelogram demonstrating PUJ
obstruction in the lower pole ureter
Fig. 92.5  Intravenous pyelogram demonstrating a large
ureterocoele on the left side with no uptake of contrast
seen within the upper renal moiety (dysplastic moiety)
T. S. Paran and P. Puri
92.4.4 DMSA (99mTc
Dimercpatosuccinic
Acid) Scan
Unlike the MAG3 scans, DMSA scans result in
renal tubular labeling and are unaffected by
obstruction to drainage. Therefore, they provide
more accurate assessment of function and also
highlight any scarring within the parenchyma.
Since the newborn kidneys may not uptake the
radionuclides properly, all such studies should be
deferred until at least 6 weeks after delivery to
allow for maturation of renal function.
92.4.5 MAG3 (Mercaptoacetyltryglyc
erine) Scan
This study combines information regarding the
relative renal function with that of any degree of
obstruction to drainage. As mentioned above, in
the presence of obstruction, the relative function
values may be somewhat overestimated. As with
DMSA scans, MAG3 should also be postponed
to allow maturation of neonatal kidneys.
92.4.6 Computed Tomography (CT)
and Magnetic Resonance
Imaging (MRI) Scans with or
Without Urogram
Over the last few years, CT and MR urograms
have slowly replaced the need for IVP. These pro-
vide clear images of renal parenchyma, any mass
lesions and that of urothelium (Kawashima et al.
2004a, b). In addition, they also provide detailed
images of the intra-abdominal organs in relation
to the urinary tract, highlighting any external
compression on the pelvis and ureters. MRI is
particularly useful in cases where occult dysplas-
tic renal moieties, ectopic ureters, and uretero-
coeles are suspected and not clearly defined by
all other modalities of investigation. MR urogra-
92  Ureteric Duplication Anomalies
phy is highly accurate in the assessment of ecto-
pic ureters. In incontinent girls, MR urography
should be the method of choice for depicting or
ruling out ectopic ureter (Figueroa et al. 2014).
92.5 Vesicoureteric Reflux (VUR)
VUR is the commonest problem associated with
both complete and incomplete ureteric duplica-
tions. It is more common in girls than in boys. In
complete duplications, it is mostly seen in the
lower pole ureter, which inserts superiorly and
laterally, resulting in a short intramural tunnel.
However, following ureterocoele puncture or
excision and reimplantation, VUR may also be
seen in the upper pole ureter.
The management of VUR in ureteric duplica-
tion is the same as in normal nonduplicated sys-
tems. Studies have shown that the rate of
resolution of minor grades of reflux is similar to
that seen in nonduplicated systems. In the new-
born period, it is essential to start chemoprophy-
laxis, until all the problems associated with
Fig. 92.6  The dilated upper pole ureter should be cut at its insertion with the lower pole ureter. Following repair of the
lower pole ureteric defect, the upper pole ureter is anastamosed end-to-side to lower pole renal pelvis
1285
ureteric duplication are identified. High grades of
reflux, associated breakthrough infections in
spite of adequate chemoprophylaxis, and pro-
gressive renal scarring are all indications for an
antireflux surgical intervention.
Surgical options include endoscopic subure-
teric Deflux injection, reimplantation of the
ureter(s), ureteroureterostomy (Fig.  92.6), or
heminephrectomy (Fig. 92.7), when the function
within the associated renal moiety is poor.
Reimplantation of the refluxing ureter could only
be undertaken in isolation, if it can be safely iso-
lated from the normal ureter without causing
ischemic injury. If this is not possible or when
there is documented reflux into both ureters on
the same side, then common sheath reimplanta-
tion (where both ureters are reimplanted without
being separated) could be done (Fig.  92.8).
Recurrence of reflux is well reported within these
reimplanted ureters, and may necessitate further
surgery.
Endoscopic correction of VUR (with submu-
cosal injection of Deflux) has been shown to be
highly effective for high-grade reflux in single
1286
Fig. 92.7  Through a subcostal loin incision and retro-
peritoneal approach, the duplex kidney is exposed. The
entire hilar vessels, including that of the lower pole should
be clearly identified, before the upper pole vessels could
Fig. 92.8  The two ureters, which are bound closely, are
dissected out together and reimplanted en bloc to avoid
ischemic injury to the lower segments of the ureters
nonduplicated systems by several authors. Puri
et  al. have shown that this technique could be
safely and effectively adopted in ureteric duplica-
tion as well. The injection is made under the
T. S. Paran and P. Puri
be tied off. The upper pole is usually dysplastic with a
clear demarcation from the healthy lower pole. The defect
following the heminephrectomy is closed to achieve
hemostasis
bladder mucosa, 2–3  mm below the refluxing
ureteric opening (which is usually the one belong-
ing to the lower renal moiety that opens superi-
orly and laterally in the bladder) at the 6 o’clock
position. The needle needs to be advanced fully
and enough material is injected until a “volcanic”
bulge of the mucosa is seen around both ureteric
orifices, resulting in a slit-like appearance of both
orifices.
Ureteroureterostomy (Fig.  92.6), where the
markedly dilated ureter is anastamosed to the
side of the normal ureter, was initially used to
overcome obstructed ureters secondary to ure-
terocoeles. However, this approach has been suc-
cessfully used in dilated ureters secondary to
severe VUR as well. But, de novo ipsilateral
VUR of the normal ureter has been noted follow-
ing ureteroureterostomy. The advantages of this
approach are that it could be performed laparo-
scopically with minimal morbidity, shorter hos-
pital stay, and better cosmetic results.
When the refluxing ureter is associated with a
poorly functioning or nonfunctioning dysplastic
renal moiety, then heminephrectomy would be
the best option (Fig. 92.7). At times, both upper
and lower renal moieties may have very poor
function and nephroureterectomy will become
the option of choice. Most reports show that a
second incision to remove the residual ureteric
stump is not necessary in most children and could
92  Ureteric Duplication Anomalies
be safely undertaken in the minority of children
who develop problems related to residual ureteric
stump (De Caluwe et al. 2002). However, when
heminephrectomy or nephrectomy is undertaken
laparoscopically, complete removal of the
ureter(s) could safely be carried out at the same
time.
92.6 Ureterocoele
Ureterocoele is a cystic dilatation of the terminal
intramural segment of the distal ureter. It is usu-
ally associated with the dilatation of the ureter
and calyces with a dysplastic poorly functioning
renal moiety. Ureterocoeles can be classified into
intravesical and extravesical based on the posi-
tion of their opening. In the intravesical uretero-
coele, the opening of the ureterocoele is located
between the normal position of the ureteric ori-
fice and the bladder neck. The extravesical ure-
terocoele opens ectopically at the bladder neck or
urethra, and the opening is usually proximal to
the external sphincter. These ectopic uretero-
coeles are associated with significant obstruction
and dysplastic upper renal moieties. A large ure-
terocoele could be seen prolapsing through the
vestibule (Fig. 92.9).
Ureterocoeles have been reported to be 4–8
times more common in females than in males. In
females, 95% of the ureterocoeles are associated
Fig. 92.9  Large ureterocoele seen to prolapse through
the vestibule
1287
with complete duplication of ureters, while in
males only two thirds are associated with com-
plete duplication; the remainder is associated with
single system normal kidneys. The cystic swelling
of the ureterocoele and the dilated ureters associ-
ated with them are easily identified in the antena-
tal scans, after 20 weeks of gestation. Most
intravesical ureterocoeles are treated with endo-
scopic puncture. In neonates with an uninfected
ureterocoele discovered by antenatal scans, endo-
scopic puncture of the ureterocoele alone may be
sufficient. If a dysplastic renal moiety is identified
in these asymptomatic children following endo-
scopic puncture or deroofing, on functional imag-
ing studies, one could adopt a wait-and-see policy
or elect to carry out heminephrectomy at a conve-
nient time. Of late, the consensus has shifted to
minimal intervention in these antenatally diag-
nosed asymptomatic children during the 1st year
of life and subsequent evaluation as to what surgi-
cal reconstruction, if any, is required.
In those infants with infected ureterocoele,
early endoscopic puncture of the ureterocoele
must be undertaken. Imaging studies, such as
DMSA scans, should be delayed until 6–8 weeks,
following the endoscopic deroofing, to attain
accurate values. The overall incidence of VUR
following endoscopic puncturing of the uretero-
coele is much less than with open surgical proce-
dures. Studies have shown that almost 90% of
intravesical and 50% of extravesical ureterocoele
renal moieties show useful function, if drained in
early life. Therefore, a full anatomic and func-
tional assessment should be delayed till 1 year of
age, and appropriate reconstructive surgery must
then be undertaken.
In almost 50% of extravesical and 15% of
intravesical ureterocoeles, further reconstructive
surgery will be required. A dysplastic nonfunc-
tioning upper moiety, prone to recurrent infection
is better removed (heminephroureterectomy;
Fig. 92.7). When the upper moieties have reason-
able function, then ureteroureterostomy
(Fig.  92.6) or excision of ureterocoele followed
by ureteric reimplantation can be carried out
(Fig. 92.10). The final management must be tai-
lored according to the functional and anatomical
findings in each child, and demands a flexible
approach.
1288
Fig. 92.10  Excision of ureterocoele and reimplantation of duplex ureters
Studies have shown that in up to 50% of those
who were treated with heminephrectomy and
extended ureterectomy, a second bladder surgery
is necessary to deal with VUR of the lower pole
ureter or rarely for bladder outlet obstruction.
However, the management of VUR that only
occurs following heminephrectomy could be
guided by the same principles as discussed above.
92.7 Ectopic Ureters
Ectopic ureters are nearly three times more com-
mon in females than in males. As with uretero-
coeles, the vast majority of the ureters are
associated with duplication in females. In males,
the majority of ectopic ureters are associated
with single system kidneys. The com-monest
sites of ectopic ureteral openings in males are
posterior urethra and prostatic urethra; a small
percentage is seen to be associated with seminal
vesicles and epididymis. In females, the com-
T. S. Paran and P. Puri
monest sites of ectopic ureteric openings are the
urethra, vestibule, and vagina.
When ureteric duplication is detected in ante-
natal scans, a thorough evaluation must include a
search for ectopic openings. However, when
present later in life, the females tend to present
with a history of normal voiding pattern, with
damp underwear day and night. In males, the
symptoms may be one of urgency and frequency
or epididymo-orchitis (Warchol et al. 2014). The
most important factor in diagnosing ectopic ure-
ters is high index of clinical suspicion.
The ectopic ureters are usually associated
with the upper renal moiety, and may be seen eas-
ily with the initial ultrasound scanning. However,
at times, demonstrating an ectopic ureteric open-
ing with imaging studies can be highly challeng-
ing. Intravenous pyelography, contrast-enhanced
delayed CT scan, and MRI scan have all proven
useful in the diagnosis of an ectopic ureteric
opening and the associated nonfunctioning, non-
dilated renal moiety.
92  Ureteric Duplication Anomalies
When diagnosed later in life, there has already
been a period of infection and dysplasia in the
associated renal moiety, and heminephrectomy is
the treatment of choice in these children. Of late,
due to early detection following antenatal diag-
nosis, the upper moieties could be saved by ure-
teropyelostomy, ureteroureterostomy, or ureteric
reimplantation, when good relative function is
confirmed with imaging studies.
At times, when continued vaginal discharge or
local infection is noted, a second surgery may be
necessary to deal with the residual distal segment
of ureter following the initial surgery.
92.8 Pelviureteric Junction
Obstruction (PUJO)
Though PUJO in ureteric duplication is relatively
rare, when present, it is mostly seen in association
with the lower renal moiety. On renal ultrasound, a
dilated lower pole renal pelvis, with no dilatation
of the ureter is seen. MAG3 scintigraphy is neces-
sary to confirm the degree of obstruction and the
relative function of the affected renal moiety.
Depending on the exact anatomy, a standard
pyeloplasty or a pyeloureterostomy (anastomos-
ing the lower pole pelvis to the upper pole ureter)
could be carried out. During pyeloureterostomy,
the remainder of the distal ureter should be
removed to avoid VUR into this remnant ureteric
stump and associated infections in future.
However, if the function in the associated renal
moiety is poor, then heminephrectomy will
become the treatment of choice.
92.9 Conclusion
With the increased referrals of antenatally diag-
nosed renal anomalies, including that of duplica-
tion of ureters, it is essential for a clinician to
remember that only a small percentage of this
group will need therapeutic interventions. Vast
majority of incomplete duplications are asymp-
1289
tomatic and need no more than ultrasound fol-
low-­up. However, those presenting with urinary
tract infections and/or hydronephrosis need
appropriate investigations and interventions to
preserve ipsilateral and contra lateral nephrons
from further damage. When children present with
persistent urinary tract infections and/or urinary
leakage, high index of suspicion is necessary to
look for occult duplication anomalies.
References
Arumugam S, Subbiah NK, Senthiappan AM (2020)
Double ureter: incidence, types and its applied signifi-
cance—a cadaveric study. Cureus 12(4):e7760
Atwell JD, Cook PL, Howell CJ, Hyde I, Parker BC
(1974) Familial incidence of bifid and double ureters.
Arch Dis Child 49(5):390–393
Campbell MF, Walsh PC (1992) Anomalies of the upper
urinary tract, vol 2. Campbell’s urology. WB Saunders,
Philadelphia, pp 1376–1381
De Caluwe D, Chertin B, Puri P (2002) Fate of the retained
ureteral stump after upper pole heminephrectomy in
duplex kidneys. J Urol 168(2):679–680
Figueroa VH, Chavhan GB, Oudjhane K, Farhat W (2014)
Utility of MR urography in children suspected of hav-
ing ectopic ureter. Pediatr Radiol 44(8):956–962
Kawashima A, Vrtiska TJ, LeRoy AJ, Hartman RP,
McCollough CH, King BF Jr (2004a) CT urography.
Radiographics 24:S35–S54
Kawashima A, Sandler CM, Wasserman NF, LeRoy AJ,
King BF, Goldman SM (2004b) Imaging of urethral
disease: a pictorial review. Radiographics 24(Suppl
1):S195–S216
Kozlov VM, Schedl A (2020) Duplex kidney formation:
developmental mechanisms and genetic predisposi-
tion. F1000Res (9):F1000
Privett JT, Jeans WD, Roylance J (1976) The incidence
and importance of renal duplication. Clin Radiol
27:521–530
Rasouly HM, Lu W (2013) Lower urinary tract develop-
ment and disease. Wiley Interdiscip Rev Syst Biol
Med 5(3):307–342
Santhalia PK, Mittal D, Gupta AK, Jana M (2018) Utility
of ureteric jet; A simple yet useful toll for the evalu-
ation of complex urogenital anomaly. Indian J Radiol
Imaging 28(1):45–48
Warchol TD, Szmigielska A, Kirzmien G, Warchol S
(2014) Ectopic ureter, renal dysplasia and recurrent
epididymitis in an infant: case report and review of the
literature. Clin Case Rep 2(1):7–9
 Vesicoureteral Reflux
Prem Puri and Balazs Kutasy
93.1 Introduction
Primary vesicoureteral reflux (VUR), the retro-
grade flow of urine from the bladder into the ure-
ter and kidneys, is the most common urological
anomaly in children. It occurs in 1–2% of the
pediatric population and in 30–50% of children
who present with urinary tract infection (UTI)
(Mattoo 2011). VUR is more common in girls
than in boys with a ratio of 4:1 in children older
than 1 year (Gorelick and Shaw 1999). In infants
diagnosed with VUR <1 year of age, on the other
hand, there is a clear predominance of males
(Sjostrom et  al. 2009). Boys also tend to have
higher grades of VUR diagnosed at younger ages,
although their VUR is more likely to resolve with
time (Hannula et  al. 2010; Menezes and Puri
2009; Alsaywid et al. 2010).
The association of VUR, UTI, and renal dam-
age is well-known (Fanos and Cataldi 2004).
Parenchymal injury in VUR occurs early, before
age 3 years in most patients. Most renal scars are
present when reflux is discovered at initial evalu-
ation for UTI (Peters and Rushton 2010).
P. Puri (*)
Department of Pediatric Surgery, Beacon Hospital,
and University College Dublin, Dublin, Ireland
B. Kutasy
Astrid Lindgren’s Children’s Hospital, Karolinska
University Hospital, Stockholm, Sweden
e-mail: balazs.kutasy@sll.se
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_93
93
The hereditary and familial nature of VUR is
now well recognized and several studies have
shown that siblings of children with VUR have a
much higher incidence of reflux than the general
pediatric population. Prevalence rates of 27%–
51% in siblings of children with VUR and a 66%
rate of VUR in offspring of parents with previ-
ously diagnosed reflux have been reported
(Chertin and Puri 2003; Menezes and Puri 2009;
Puri et al. 2011). The risk and severity of VUR in
siblings depend on the sex of the affected child.
Brothers of index male patients have a higher
grade of reflux and a higher rate of associated
duplex systems (Pirker et al. 2006b). The devel-
opment of mild renal scarring in familial VUR
seems to mainly depend on UTIs, while moderate
and severe scarring are associated with high-­
grade reflux and male gender (Pirker et al. 2006a).
Furthermore, there is an increased risk of reflux-­
related morbidity among the first-, second-, and
third-degree relatives of index patients with VUR
(Hunziker and Puri 2012). VUR can also occur
secondary to neurogenic bladder and posterior
urethral valves, high intravesical pressures result-
ing in reflux.
93.2 Etiopathogenesis
The ureterovesical junction (UVJ) acts as a valve
and closes during micturition or when the blad-
der contracts. The UVJ is structurally and func-
1291
1292
tionally adapted to allow the intermittent passage
of urine and prevent the reflux of urine in the
bladder (Roshani et al. 1996). The main defect in
patients with VUR is believed to involve the mal-
formation of the UVJ, in part due to shortening of
the submucosal ureteric segment due to congeni-
tal lateral ectopia of the ureteric orifice. Since
VUR primarily involves abnormalities of the ure-
ter and ureteric orifice, it has been suggested that
the timing and positioning of branching of the
bud from the Wolffian duct may be related to
VUR. For this reason, in the ­complete duplica-
tion of ureters, since the ureter draining the lower
pole has a proximal position in the bladder with
respect to other ureters, it is more likely to have a
chance of reflux because of the short tunnel
(Subramaniam et al. 2015).
Lower urinary tract disorder (LUTD) is an
entity in VUR etiology, which has been more
understood in recent years (Koff et al. 1998; Ural
et al. 2008). LUTD is a common finding in chil-
dren evaluated for UTIs (Sillen et  al. 2010;
Snodgrass 1998). In LUTD with no neurological
abnormality, there are two mechanisms that lead
to problems: failure to inhibit the detrusor reflex
during the storage phase; and overcompensation
by the external sphincter during the voiding
phase (Koff et al. 1998). LUTD results in volun-
tary contractions of the external sphincter by the
child in an attempt to stop leakage of urine by
involuntary contractions of the detrusor. This
often causes higher intravesical pressures both
during the storage and voiding phase. High intra-
vesical pressure may cause reflux of urine, espe-
cially during voiding when the intravesical
pressures are the highest. Not all LUTD bladders
will develop reflux despite very high intravesical
pressure. If LUTD does cause reflux, it probably
occurs due to the compromise of a borderline
ureterotrigonal structure or anatomical decom-
pensation of the bladder caused by abnormal
bladder dynamics (Koff et al. 1998).
Familial clustering of VUR implies that
genetic factors have an important role in its
pathogenesis, but no single major locus or gene
for VUR has yet been identified and most
researchers now acknowledge that VUR is genet-
ically heterogeneous (Puri et  al. 2011). Darlow
P. Puri and B. Kutasy
et  al. (2017) identified, through genome-wide
parametric and nonparametric linkage analysis, a
region on 10q26 as a major genetic contributor to
VUR in European populations. Verbitsky et  al.
(2021), recently reported the largest VUR copy
number variant analysis and genome-wide asso-
ciation study to date, accounting for multiple
modes of inheritance and sex-specific effects in
VUR, and found that 6% of patients carried high-­
risk genotypes.
93.2.1 Mechanism of Renal Scarring
The association between VUR and renal scarring
is now widely recognized. Ten to forty percent of
children with symptomatic VUR have evidence
of renal scarring (Tekgul et al. 2012). Scarring is
directly related to the severity of reflux (Tekgul
et  al. 2012). Belman and Skoog assessed renal
scarring in 804 refluxing units and found renal
scars in 5% of those with grade I reflux, 6% of
those with grade II reflux, 17% of those with
grade III Reflux, 25% of those with grade IV
reflux, and 50% of those with grade V reflux
(Belman and Skoog 1989). Follow-up studies
have shown that 10–20% of children with reflux
nephropathy develop hypertension or end-stage
renal disease (Blumenthal 2006).
The mechanism by which reflux produces
renal scars is still not clear. It is essential to dis-
tinguish between the commonly acquired seg-
mental scarring associated with VUR and
infection, and the primary scarring seen congeni-
tally, in which the etiology is very different and
linked to abnormal metanephric development
(Peters and Rushton 2010).
There is no doubt that bacterial pyelonephri-
tis produces renal scars experimentally and clin-
ically. Most scarring tends to occur at the renal
poles, where the anatomy of the renal papillae
permits the backflow of urine into collecting
ducts. This is called intrarenal reflux and gives
pathogenic bacteria access to the renal tubules.
The subsequent cascade of inflammation, with
release of superoxide and other mediators,
results in  local tissue ischemia and fibrosis.
When enough renal parenchyma is affected,
93  Vesicoureteral Reflux
hypertension, renal insufficiency, and renal fail-
ure can result (Subramaniam et  al. 2015).
However, only half of patients with acute pyelo-
nephritic will have such a scar. What converts an
acute inflammatory process into a scar in some
patients and not in others is not clearly
understood.
Some of the worst examples of renal injury
associated with VUR are those that are present at
birth. As renal damage at that time cannot be the
consequence of infection, such injury is assumed
to be developmental in origin which is then called
congenital nephropathy (Peters and Rushton
2010).
The three mechanisms considered potential
etiologies for renal scar formation are (1) reflux
of infected urine with interstitial inflammation
and damage, (2) sterile, usually high-grade reflux,
which may damage the kidney through a mechan-
ical or immunological mechanism, and (3) abnor-
mal embryological development with subsequent
renal dysplasia (Chertin and Puri 2003). Patients
in the latter group may also have UTI in the post-
natal period, resulting in extensive parenchymal
damage. It is well recognized that in the first two
groups of renal parenchymal damage, it is essen-
tial to discover reflux early before damage can be
initiated. In the third group, it is clear that con-
genital damage currently cannot be prevented.
However, in these patients, it is mandatory to dis-
cover reflux at the early stages to prevent expo-
sure to UTI and avoid the possible progression of
renal parenchymal damage (Chertin and Puri
2003).
93.3 Diagnosis
93.3.1 Clinical Presentation
It is obviously important to diagnose VUR at the
earliest possible age, preferably in infancy. There
are a number of clinical presentations, which
should raise the suspicion of VUR in an infant.
As antenatal ultrasound becomes increasingly
routine, many cases will be suspected before
birth and should be investigated within the first
month of life (Grazioli et al. 2010). Infants with
1293
antenatally detected VUR show a male prepon-
derance (Blumenthal 2006). Infants with a poor
urinary stream as in posterior urethral valves or
infants with spina bifida have a high incidence of
VUR, while early investigations are indicated in
the first-degree relatives of index patients with
high-grade VUR (Menezes and Puri 2009; Sillen
et al. 2010).
In most cases, VUR is discovered clinically
after investigation of a UTI.  The incidence of
VUR in children with UTI is 30–50%, with even
higher incidence in infants (Hannula et al. 2010).
Among all infants with UTIs, boys are more
likely to have VUR than girls (Hannula et  al.
2010; Alsaywid et  al. 2010). Girls with VUR
seem to have a continuous risk of new infections,
even after potty training (Coulthard et al. 1997).
Children with VUR may also present with symp-
toms of voiding dysfunction, such as frequency,
urgency, and incontinence.
93.3.2 Radiological Investigations
93.3.2.1 Ultrasound
Since ultrasound is a noninvasive and non-­
radiating examination, it has a special role in the
radiological investigation of VUR.  Sonography
should be performed in any infant with suspicion
of VUR (Tekgul et al. 2012; Peters et al. 2010).
The kidneys and upper ureters should be exam-
ined both in B-mode and real time. The bladder
and lower ureters are assessed by real-time exam-
ination at each UVJ for dilatation, configuration,
peristalsis, and continuity with the bladder base.
VUR is suspected in the presence of a dilated pel-
vicaliceal system, upper or lower ureter, unequal
renal size, or cortical loss and increased echo-
genicity (Fig.  93.1). Sonography is not suffi-
ciently sensitive or specific for diagnosing VUR
(Logvinenko et al. 2015). However, screening or
monitoring of VUR with careful ultrasound can
avoid unnecessary invasive or irradiating
examinations.
Recently, the contrast-enhanced voiding uro-
sonography (ce-VUS) is gaining popularity
(Ntoulia et  al. 2018). It is relatively novel
contrast-­enhanced ultrasonographic technique
1294
Fig. 93.1  Ultrasound in an infant with grade V VUR
showing hydronephrosis and hydroureter
utilizing microbubbles to detect VUR. During the
exam, the ultrasound contrast agent is adminis-
tered intravesically via the urinary catheter.
Afterward, both kidneys and the bladder are
examined using a low mechanical index contrast
ultrasound preset, frequently alternating between
them during the filling and voiding phases. VUR
can be confirmed by the detection of echogenic
microbubbles in the upper urinary tract/pelvical-
yceal system (Ntoulia et al. 2018). The main ben-
efit of ce-VUS is the absence of ionizing radiation.
Recent studies have demonstrated good results in
the detection of VUR, however, the anatomic
detail depicted is inferior (Mane et al. 2018).
93.3.2.2  Voiding Cystourethrography
VUR is a dynamic process. Bladder filling and
voiding are necessary for its elucidation which
requires catheterization for adequate documenta-
tion. Despite the concerns about ionizing radia-
tion and its invasive nature, conventional voiding
cystourethrography (VCUG) remains the gold
standard for diagnosing VUR.  It allows better
determination of the grade of VUR (in a single or
a duplicated kidney) and better assessment of
P. Puri and B. Kutasy
Fig. 93.2 Voiding cystography showing grade V left
reflux in a male infant
bladder and urethral configuration (Tekgul et al.
2012) (Fig.  93.2). A clinician should use the
well-established grading system developed by
the International Reflux Study Committee
(Table 93.1) to minimize probable interobserver
deviations (Lebowitz et al. 1985).
Some investigators employ nuclear cystogra-
phy for diagnosing VUR (Fettich et al. 2003). This
can be either direct or indirect using technetium-­
labeled diaminotetra-ethyl-­
pentaacetic acid
(DTPA). In direct nuclear cystography, DTPA is
instilled into the bladder by urethral catheter or
suprapubic injection and the ureters and kidneys
are observed on camera during bladder filling and
voiding. In indirect nuclear cystography, DTPA is
injected intravenously. After the bladder is filled,
the patient is instructed to void, and the counts
93  Vesicoureteral Reflux
taken over the ureters and kidneys are used to
assess the presence of VUR. Indirect nuclear cys-
tography requires a cooperative patient and there-
fore is of no value in infants. The main disadvantage
of nuclear cystography is that it does not give ana-
tomical detail and VUR cannot be graded accord-
ing to international classification (Stefanidis and
Siomou 2007).
93.3.2.3 DMSA Scan
The evaluation of renal parenchyma by dimer-
captosuccinic acid Tc99m (DMSA) is crucial.
DMSA is the best nuclear agent for visualizing
the cortical tissue and differential function
between both kidneys (Piepsz et  al. 2001). It is
the most sensitive imaging modality for detecting
renal scarring. DMSA is taken by proximal renal
Table 93.1  Grading system for VUR on VCUG accord-
ing to the International Reflux Study Committee
(Lebowitz et al. 1985)
Grade I Reflux does not reach the renal pelvis;
Grade II Reflux reaches the renal pelvis; no
dilatation of the collecting system; normal
fornices.
Grade III Mild or moderate dilatation of the ureter,
with or without kinking; moderate
dilatation of the collecting system; normal
or minimally deformed fornices.
Grade IV Moderate dilatation of the ureter with or
without kinking; moderate dilatation of the
collecting system; blunt fornices, but
impressions of the papillae still visible.
Grade V Gross dilatation and kinking of the ureter,
marked dilatation of the collecting system;
papillary impressions no longer visible;
intraparenchymal reflux.
Fig. 93.3  DMSA scan in an infant with VUR shows small scarred right kidney
1295
tubular cells and is a good indicator of renal
parenchyma function. In areas of acute inflam-
mation or scarring, DMSA uptake is poor and
appears as cold spots. DMSA scans are therefore
used to detect and monitor renal scarring (Piepsz
et al. 2001) (Fig. 93.3). A baseline DMSA scan at
the time of diagnosis can be used for comparison
with successive scans later during follow-up
(Tekgul et al. 2012). The degree of renal damage
has major importance during follow-up. DMSA
can also be used as a diagnostic tool during sus-
pected episodes of acute pyelonephritis (Lee
et al. 2009).
93.3.2.4 Diagnostic Workup
The diagnostic workup should aim to evaluate the
overall health and development of the child, the
presence of UTIs, renal status, the presence of
VUR, and lower urinary tract function. A basic
diagnostic workup comprises a detailed medical
history (including family history and screening
for LUTD), a physical examination including
blood pressure measurement, urinanalysis
(assessing proteinuria), urine culture as indicated,
and measurement of serum creatinine level in
patients with bilateral renal parenchymal abnor-
malities (Tekgul et al. 2012; Peters et al. 2010).
Ultrasonography of the kidney and the bladder
is the first standard evaluation tool for children
with prenatally diagnosed hydronephrosis
(Tekgul et al. 2012). Ultrasonography should be
delayed until after the first week after birth
because of early oliguria in the neonate. The first
two ultrasonography scans within the first
1–2 months of life are highly accurate for defin-
1296
ing the presence or absence of renal pathology.
VUR is rare in infants with two normal succes-
sive postnatal ultrasonography examination and
if present, is likely to be low grade (Coplen et al.
2008; Hafez et al. 2002). The presence of cortical
abnormalities on ultrasonography warrants the
use of VCUG for detecting VUR (Skoog et  al.
2010). The use of VCUG is recommended in
patients with ultrasonographic findings of bilat-
eral high-grade hydronephrosis, duplex kidneys
with hydronephrosis, ureterocele, ureteral dilata-
tion, and abnormal bladders, because the likeli-
hood of VUR is much higher (Tekgul et al. 2012).
When infants who were diagnosed with prenatal
hydronephrosis become symptomatic with UTI,
further evaluation with VCUG should be consid-
ered (Tekgul et al. 2012; Skoog et al. 2010).
The prevalence of VUR in siblings and off-
spring of reflux patients is high (Menezes and
Puri 2009). Therefore, the early identification of
children with VUR can prevent episodes of UTI
and renal scaring (Menezes and Puri 2009).
Siblings and offspring should be screened by
renal ultrasonography. VCUG is recommended if
there is evidence of renal scarring on ultrasound
or a history of UTI (Tekgul et al. 2012).
VCUG is recommended at age 0–2 years after
the first proven febrile UTI (Tekgul et al. 2012).
If reflux is diagnosed, further evaluation consists
of DMSA scan. This conventional evaluation is
called bottom-up approach. An alternative top-­
down approach is also an option. This approach
is carried out a DMSA scan first, close to the time
of a febrile UTI, to determine the presence of
pyelonephritis, which is then followed by VCUG
if the DMSA scan reveals kidney involvement. A
normal DMSA scan with no subsequent VCUG
will fail to detect VUR in 5–27%, with the cases
of missed VUR being presumably less signifi-
cant. In contrast, a normal DMSA scan with no
VCUG will avoid unnecessary VCUG in >50%
of individuals screened (Hansson et  al. 2004;
Herz et al. 2010; Quirino et al. 2011; Preda et al.
2007).
Detection of LUTD is essential in treating
children with VUR (Skoog et  al. 2010; Tekgul
et al. 2012). It has been suggested that reflux with
LUTD will resolve faster after LUTD correction
P. Puri and B. Kutasy
and that patients with LUTD are at higher risk for
developing UTIs and renal scaring (Colen et al.
2006). If there are symptoms suggestive of LUTD
(urgency, wetting, constipation, or holding
maneuvers), an extensive history and examina-
tion, including voiding charts, uroflowmetry, and
residual urine determination, will reliably diag-
nose underlying LUTD (Sillen et  al. 2010). In
LUTD, VUR is often low-grade and ultrasonog-
raphy findings are often normal (Sillen et  al.
2010). There is no indication for performing
VCUG in all children with LUTD.  Any child
who fails standard therapy for LUTD should
undergo urodynamic investigation (Tekgul et al.
2012).
93.4 Management
The goals of management of the child with VUR
are to prevent recurrent febrile UTIs, prevent renal
injury, and minimize the morbidity of the treatment
(Peters etal, 2010; Tekgul et  al. 2012; Hunziker
et al. 2012) The two main options available for the
treatment of VUR are medical or surgical.
93.4.1 Medical Management
This strategy is based on three important assump-
tions: (1) Sterile low-grade VUR is not harmful
to the kidneys in most cases and has no relevant
effect on kidney function, (2) Children can out-
grow VUR, at least the lower grades, and (3)
Continuous low-dose antibiotic prophylaxis
(CAP) can prevent infection for many years
while VUR is still present (Elder et  al. 1997;
Craig et al. 2009; Williams et al. 2006; Wheeler
et al. 2004).
The conservative approach includes watchful
waiting, intermittent or continuous antibiotic
prophylaxis, and bladder rehabilitation in patients
with LUTD (Colen et al. 2006; Craig et al. 2009;
Williams et  al. 2006; Wheeler et  al. 2004).
Circumcision during early infancy may be con-
sidered as part of the conservative approach,
because it is effective in reducing the risk of UTI
(Singh-Grewal et al. 2005).
93  Vesicoureteral Reflux 1297

Previous studies have been shown that there is vesical ureter. These procedures, although effec-
no or low evidence that CAP will prevent UTI or tive, involve open surgery, prolonged in-hospital
renal damage for low-grade or asymptomatic stay, and are not free of complications, even in
VUR (Peters et  al. 2010; Hari et  al. 2015; the best hands. The rate of obstruction after ure-
Roussey-Kesler et  al. 2008). It has been also teral reimplantation needing reoperation reported
reported that CAP is superior to surveillance con- by the American Urological Association was
cerning prevention of UTI in patients with high-­ 0.3–9.1% (Duckett et al. 1992).
grade VUR (Peters et al. 2010; Craig et al. 2009; If VUR resolution is the main goal, open sur-
Brandstrom et  al. 2010a). The Swedish reflux gery can achieve a success rate of >97% in grade
study has demonstrated a significantly higher risk II–IV VUR, and >80% in grade V (Routh et al.
to develop renal scarring without treatment 2012; Garcia-Aparicio et  al. 2013; ‘Prospective
(Brandstrom et al. 2010b). This study has further trial of operative versus non-operative treatment
shown that there is a significant difference of severe vesicoureteric reflux in children:
between girls and boys. Girls have a higher risk 5 years’ observation. Birmingham Reflux Study
of UTI without prophylaxis. Therefore, it is rec- Group 1987). The International reflux study has
ommended to stop CAP at 1 year of age in boys demonstrated that surgery was superior to medi-
and continue the CAP until finishing of toilet cal treatment concerning recurrent febrile UTI,
training in girls (Brandstrom et al. 2010a). however there was no difference in the renal out-
On the other hand, there is a growing concern come, even in the 10-year long-term follow-up
with CAP (Kutasy et al. 2017). It has been shown (Piepsz et al. 1998; Jodal et al. 1992, 2006).
that CAP can increase the risk of infection with Minimal invasive antireflux techniques, such
antibiotic-resistant bacteria (Nagler et al. 2011). as laparoscopic or robotic-assisted surgery, are
The compliance of the treatment can be poor, and increasing in popularity. It has been shown that
therefore it can lower the effectiveness of the these approaches are feasible and safe. However,
treatment by 40% (Copp et al. 2010). Furthermore, these techniques are more invasive than endo-
a Cochrane report from 2011 calculated that 33 scopic treatment, and the advantages over open
children would need long-term antibiotics treat- surgery are still debated. Therefore, at present, a
ment to prevent one more child developing kid- laparoscopic approach cannot be recommended
ney damage over the course of 2–3 years (Nagler as a routine procedure (Tekgul et al. 2012).
et al. 2011).
93.4.2.2 Endoscopic Treatment
of VUR
93.4.2 Surgical Treatment The introduction of endoscopic treatment of
VUR was a radical departure from the standard
93.4.2.1 Antireflux Procedures surgical intervention and observational manage-
Various intra- and extra-vesical techniques have ment of VUR (Puri and O’Donnell 1984;
been described for the surgical correction of O’Donnell and Puri 1984; Puri 1990). Since the
VUR. All techniques share the basic principle of U.S.  Food and Drug Administration (FDA)
lengthening the intramural part of the ureter by approval of dextranomer/hyaluronic acid
submucosal embedding of the ureter (Tekgul (Deflux®) as a biocompatible agent for submu-
et al. 2012). The majority of the open antireflux cosal injection into the refluxing ureteric orifice
procedures entail opening the bladder and per- in 2001, endoscopic treatment of VUR has gained
forming a variety of procedures on the ureters worldwide popularity and is today an integral
such as transvesical reimplantation (Politano-­ part of the American Urological Association
Leadbetter technique) and transtrigonal advance- guidelines (Peters et  al. 2010) and also the
ment of the ureters (Cohen technique). The main European Association of Urology guidelines
extravesical technique is Lich-Gregoir where a (Tekgül et al. 2012) for the management of pri-
submucosal groove is created around the juxta- mary reflux in children.
1298 P. Puri and B. Kutasy

Endoscopic treatment is a well-established uous injection, the needle should then gradually
method in VUR management (Tekgul et  al. be withdrawn until a “volcanic” bulge of implant
2012). This technique involves a bulking agent is seen (Figs. 93.4, 93.5, 93.6, and 93.7). It is rec-
injected under the submucosal portion of ommended that the needle is kept in position for
­refluxing ureter and the achievement of ureteric 30 s after administration to minimize extrusion of
coaptation is aimed. The injected bulking agent the injected product. Depending on the used
elevates the ureteral orifice and the distal ureter, bulking agent, most refluxing ureters require not
so that coaptation is increased. This results in more than 0.4–1.0 mL to correct VUR with this
narrowing of the lumen, which prevents reflux of technique. In general, a correctly placed injection
urine into the ureter, while still allowing its ante- creates the appearance of a nipple, on top of
grade flow. It is a minimally invasive method and
it can be performed on an outpatient basis (Tekgul
et  al. 2012). Several bulking agents have been
used over the past three decades. The most widely
used tissue augmenting substance is dextrano-
mer/hyaluronic acid (Deflux) (Tekgul et  al.
2012). Over the years, several injection tech-
niques have been described for the endoscopic
treatment of VUR including subureteric injection
(STING) (O’Donnell and Puri 1984), intraure-
teric hydrodistention implantation technique
(HIT) (Kirsch et  al. 2004), combined HIT/
STING, and double HIT (i.e., proximal and distal
intraluminal submucosal injections) (Kirsch and
Arlen 2014).

Endoscopic Injection Technique

The disposable Puri flexible catheter for injection
(STORZ) is a 4F catheter onto which is swaged a Fig. 93.4  Endoscopic appearance of a grade V refluxing
21- gauge with 1  cm of the needle protruding ureteral orifice prior to injection
from the catheter. Alternatively, a rigid needle
can be used. A 1-mL syringe filled with Deflux is
attached to the catheter. The needle is introduced
2–3  mm below the refluxing orifice at the 6
o’clock position into the bladder mucosa and
advanced for approximately 4–5  mm into the
submucosal portion of the distal ureter. However,
in children with grade IV and V reflux with a
wide ureteric orifice, the needle should be
inserted directly into the affected orifice to
increase the length of the intravesical ureter. It is
important to introduce the needle with pinpoint
accuracy before starting the injection as any per-
foration of the mucosa or the ureter may allow
the injected material to escape, which may result
in failure. As the tissue augmenting substance is
Fig. 93.5  In high-grade reflux, the needle for injection is
administrated, a bulge slowly starts to appear at inserted under the mucosa directly inside the affected ure-
the floor of the submucosal ureter. Under contin- teral orifice
93  Vesicoureteral Reflux
Fig. 93.6  During the injection, the needle is slowly with-
drawn until a “volcanic” bulge of paste is seen
Fig. 93.7  At the end of the injection, the ureteral orifice
has a slit-like appearance
which sits a slit-like or inverted crescentic orifice.
If the bulge appears in an incorrect place, e.g., at
the side of the ureter or proximal to it, the needle
should not be withdrawn. Instead, it should be
moved so that the point is in a more favorable
position. The non-injected ureteric roof usually
retains its compliance while preventing reflux.
For the HIT, the bladder is semi-filled to
allow for good visualization of the affected ori-
fice and to avoid tension within the submucosal
layer of the ureter secondary to overdistension.
1299
Hydrodistention of the refluxing orifice is initi-
ated to define the injection site within the ure-
teric tunnel. The needle is inserted approximately
4 mm into the submucosa of the intramural ure-
ter at the 6 o’clock position. At this point, irriga-
tion should be stopped while the tissue
augmenting substance is injected.
Postoperative Care
Postoperative urinary catheterization is not nec-
essary and patients are usually able to void
­without any problems after recovery from gen-
eral anesthesia. The vast majority of infants and
children with VUR can be treated as day cases.
Antibiotic prophylaxis should be continued until
VCUG and renal ultrasonography are performed
3 months after the procedure. Antibiotics can be
stopped if the VCUG demonstrates resolution of
reflux (Figs.  93.8 and 93.9). Nevertheless, a
patient should be considered for further endo-
scopic treatment if VUR has not resolved. In gen-
eral, it is recommended that following successful
endoscopic correction of VUR, renal and bladder
ultrasound is repeated at 1  year, and in 2-year
intervals thereafter to monitor the appearance of
the urinary tract, as well as the site and size of the
injected implant.
Results of Endoscopic Treatment
Puri et al. (2012) reported results in 2341 reflux-
ing ureters in 1551 children with an overall suc-
cess rate of 87% after one injection. This success
or failure to a greater degree is dependent on the
precise injection technique (Moore and Bolduc
2014; Oswald et  al. 2002; Akin et  al. 2014;
Friedmacher et al. 2018). The success rate is sig-
nificantly lower for complete duplicated ureters
(68%) (Hunziker et al. 2013) and for neuropathic
bladder (62%) (Elder et  al. 2006). The result
depends on the grade of VUR, with lower effi-
ciency in grade V and bilateral VUR cases (Elder
et  al. 2006; Routh et  al. 2012). Recently,
Friedmacher et  al. (2018) reported results of
endoscopic treatment in grades IV and V reflux.
VUR resolution rate after the first endoscopic
injection was 70% for grade IV and 62% for
grade V cases. The Swedish reflux trial compar-
1300
Fig. 93.8  Voiding cystourethrogram showing grade V
reflux
ing three treatment arms (endoscopic injection,
antibiotic prophylaxis, surveillance without anti-
biotic prophylaxis) has demonstrated that the
endoscopic treatment gives the highest resolution
rate of VUR compared to antibiotic prophylaxis
(71% vs 39%) at 2 years of follow-up. However,
the new scar formation is higher with endoscopic
injection (7%) compared with antibiotic prophy-
laxis (0%). They reported a 20% recurrence rate
at 2 years after endoscopic procedure (Holmdahl
et al. 2010).
The only significant problem with endoscopic
correction of VUR has been a failure in some
patients and who needed a repeat injection. A
meta-analysis including 5527 patients with VUR
calculated resolution rates of 78.5% for grades I
and II, 72% for grade III, 63% for grade IV, and
51% for grade V after endoscopic therapy with an
aggregate success rate of 85% following one or
more injections (Elder et  al. 2006). Another
review of 47 articles reported similar results
P. Puri and B. Kutasy
Fig. 93.9  Voiding cystourethrogram in the same child
showing resolution of VUR 3  months after the STING
procedure
(Routh et  al. 2010a). A single-center study of
1551 children (2341 refluxing ureters) with inter-
mittent and high-grade reflux that underwent
endoscopic Deflux® injection demonstrated that
VUR persisted in less than 13% of cases and was
subsequently eradicated following a second or
third injection (Puri et al. 2012). These findings
suggest that results can be improved by simply
repeating the procedure, thus being as effective
as ureteral reimplantation as shown by a random-
ized clinical trial (Garcia-Aparicio et  al. 2013).
However, a higher incidence of recurrent febrile
UTIs has frequently been noted in girls than in
boys after successful reflux correction
(Brandstrom et  al. 2010a, b). Hunziker et  al.
(2012) conducted a long-term follow-up study of
incidence of febrile UTIs in 1271 children after
successful endoscopic correction of VUR.  The
authors found that 5.7% of children developed
febrile UTIs after correction of VUR and pre-
93  Vesicoureteral Reflux 1301

dominantly were girls, the majority of whom had renal injury, while minimizing the morbidity of
bladder bowel dysfunction. investigations and therapy, and avoid overtreat-
ment (Peters et al. 2010; Tekgul et al. 2012).
Complications of Endoscopic Treatment Regardless of the grade of VUR or the pres-
Endoscopic injection of bulking agents for the ence of renal scars or symptoms, all patients
treatment of VUR has become a therapeutic diagnosed within the first year of life should be
­alternative to antibiotic prophylaxis and ureteral treated initially with CAP.  During early child-
reimplantation. Although considered a safe and hood, the kidneys are at higher risk of develop-
efficient procedure, several studies have reported ing new scars. Immediate antibiotic treatment
cases of UVJ obstruction after endoscopic cor- should be initiated for febrile breakthrough
rection of VUR. UTI.  Definitive endoscopic or surgical correc-
A recent systematic review by Friedmacher tion is the preferred treatment in patients with
and Puri (2019) evaluated the incidence of UVJ recurrent UTI or persistent high-grade VUR. In
obstruction following endoscopic injection of patients with lower grade of VUR and no symp-
different tissue augmenting substances. Twenty-­ toms, close surveillance without CAP may be an
five publications were identified that provided option. If LUTD is detected, the initial treatment
detailed information on 95 children who devel- should be the management of LUTD. If parents
oped UVJ obstruction after endoscopic treatment prefer definitive therapy to conservative manage-
of VUR.  They reported that the timing of UVJ ment, endoscopic/surgical correction may be
obstruction varied from immediately after the considered (Peters et  al. 2010; Tekgul et  al.
procedure to 63 months. Interestingly, over half 2012; Friedmacher et al. 2018).
of the patients in this review showed asymptom-
atic hydroureteronephrosis on follow-up imag-
ing, whereas the remaining presented with 93.4.4 Follow-Up
symptoms of UVJ obstruction. The authors of
this review reported that UVJ obstruction is a rare Ongoing monitoring of a child’s overall health
complication after endoscopic correction of including monitoring of blood pressure, height,
VUR, generally reported in less than 1% of and weight, and urinalysis for proteinuria and
treated cases, which appears to be independent of bacteriuria are necessary annually through ado-
the injected substance, volume, and technique. lescence (Peters et al. 2010; Tekgul et al. 2012).
However, long-term follow-up is recommended Specific investigations relating to VUR will
as asymptomatic or delayed UVJ obstruction can depend on the clinical situation and risk factors.
occur, potentially leading to deterioration of renal A high-risk patient will need more closer moni-
function. toring and more aggressive multidisciplinary
follow-up. Ultrasonography is recommended
annually-biannually (Tekgul et  al. 2012; Puri
93.4.3 Treatment Strategy et  al. 2012). Following surgical or endoscopic
procedures for VUR a renal ultrasound should
Management of VUR has been controversial. be obtained to assess for obstruction (Tekgul
There is no uniform treatment of VUR. The choice et al. 2012). Traditionally, postoperative VCUG
of management is dependent on the presence of following endoscopic injection for VUR was
renal scars, the clinical course, the grade of VUR, recommended (Puri and O’Donnell 1987; Puri
ipsilateral renal function, bilaterality, bladder et al. 2012; Friedmacher et al. 2018). Currently,
function, associated anomalies of the urinary due to high ionizing radiation and traumatic
tract, age, gender, compliance, side effects of the experience for the child during VCUG, many
medicine, antimicrobial resistance and parental centers recommend an ultrasound 2–3  months
preference (Tekgul et al. 2012). The management after endoscopic correction of VUR and stop
should minimize the risk of acute infection and antibiotic prophylaxis. The VCUG is only
1302
p­ erformed in high-risk patients or after recurrent
febrile UTIs. DMSA imaging is advised when a
renal ultrasound is abnormal, when there is a
greater concern for scarring, or if there is an ele-
vated serum creatinine (Peters et  al. 2010). If
symptomatic breakthrough UTI occurs, a new
evaluation for bladder dysfunction or recurrent
VUR is recommended (Peters et al. 2010). It is
recommended that the long-term concerns of
hypertension (particularly during pregnancy),
renal functional loss, recurrent UTI, and familial
VUR in the child’s siblings and offspring are dis-
cussed with the family (Peters et  al. 2010;
Hunziker et al. 2012, 2014).
93.5 Conclusion
Primary VUR is the most common urological
anomaly in children. The association of VUR,
UTI, and renal damage is well-known. The
choice of management is dependent on the
presence of renal scars, the clinical course, the
grade of VUR, ipsilateral renal function, bilat-
erality, bladder function, associated anomalies
of the urinary tract, age, gender, compliance,
side effects of the medicine, antimicrobial
resistance and parental preference. The man-
agement should minimize the risk of acute
infection and renal parenchymal damage, while
minimizing the morbidity of investigations and
therapy.
Acknowledgments Special thanks to the Radiological
Department at Karolinska University Hospital, Sweden,
for the DMSA images.
References
Akin M, Erginel B, Karadag CA, Yildiz A, Ozcelik GS,
Sever N, Genc NM, Dokucu AI (2014) A comparison
of the double hydrodistention implantation technique
(HIT) and the HIT with a polyacrylate/polyalco-
hol copolymer (PPC) for the endoscopic treatment
of primary vesicoureteral reflux. Int Urol Nephrol
46:2057–2061
Alsaywid BS, Saleh H, Deshpande A, Howman-Giles
R, Smith GH (2010) High grade primary vesicoure-
teral reflux in boys: long-term results of a prospective
cohort study. J Urol 184:1598–1603
P. Puri and B. Kutasy
Belman AB, Skoog SJ (1989) Nonsurgical approach to
the management of vesicoureteral reflux in children.
Pediatr Infect Dis J 8:556–559
Birmingham Reflux Study Group (1987) Prospective trial
of operative versus non-operative treatment of severe
vesicoureteric reflux in children: five years’ observa-
tion. Br Med J (Clin Res Ed) 295:237–241
Blumenthal I (2006) Vesicoureteric reflux and urinary
tract infection in children. Postgrad Med J 82:31–35
Brandstrom P, Esbjorner E, Herthelius M, Swerkersson S,
Jodal U, Hansson S (2010a) The Swedish reflux trial
in children: III. Urinary tract infection pattern. J Urol
184:286–291
Brandstrom P, Neveus T, Sixt R, Stokland E, Jodal U,
Hansson S (2010b) The Swedish reflux trial in chil-
dren: IV. Renal damage. J Urol 184:292–297
Chertin B, Puri P (2003) Familial vesicoureteral reflux. J
Urol 169:1804–1808
Colen J, Docimo SG, Stanitski K, Sweeney DD, Wise B,
Brandt P, Wu HY (2006) Dysfunctional elimination
syndrome is a negative predictor for vesicoureteral
reflux. J Pediatr Urol 2:312–315
Coplen DE, Austin PF, Yan Y, Dicke JM (2008) Correlation
of prenatal and postnatal ultrasound findings with the
incidence of vesicoureteral reflux in children with
fetal renal pelvic dilatation. J Urol 180:1631–1634.
discussion 34
Copp HL, Nelson CP, Shortliffe LD, Lai J, Saigal CS,
Kennedy WA, Project Urologic Diseases in America
(2010) Compliance with antibiotic prophylaxis in chil-
dren with vesicoureteral reflux: results from a national
pharmacy claims database. J Urol 183:1994–1999
Coulthard MG, Lambert HJ, Keir MJ (1997) Occurrence
of renal scars in children after their first referral for
urinary tract infection. BMJ 315:918–919
Craig JC, Simpson JM, Williams GJ, Lowe A, Reynolds GJ,
McTaggart SJ, Hodson EM, Carapetis JR, Cranswick
NE, Smith G, Irwig LM, Caldwell PH, Hamilton S,
Roy LP, Reflux Prevention of Recurrent Urinary
Tract Infection in Children with Vesicoureteric, and
Investigators Normal Renal Tracts (2009) Antibiotic
prophylaxis and recurrent urinary tract infection in
children. N Engl J Med 361:1748–1759
Darlow JM, Darlay R, Dobson MG, Stewart A, Charoen
P, Southgate J, Baker SC, Xu Y, Hunziker M, Lambert
HJ, Green AJ, Santibanez-Koref M, Sayer JA,
Goodship THJ, Puri P, Woolf AS, Kenda RB, Barton
DE, Cordell HJ (2017) Genome-wide linkage and
association study implicates the 10q26 region as a
major genetic contributor to primary nonsyndromic
vesicoureteric reflux. Sci Rep 7(1):14595
Duckett JW, Walker RD, Weiss R (1992) Surgical results:
international reflux study in children--United States
branch. J Urol 148:1674–1675
Elder JS, Peters CA, Arant BS Jr, Ewalt DH, Hawtrey CE,
Hurwitz RS, Parrott TS, Snyder HM 3rd, Weiss RA,
Woolf SH, Hasselblad V (1997) Pediatric vesicoure-
teral reflux guidelines panel summary report on the
management of primary vesicoureteral reflux in chil-
dren. J Urol 157:1846–1851
93  Vesicoureteral Reflux
Elder JS, Diaz M, Caldamone AA, Cendron M, Greenfield
S, Hurwitz R, Kirsch A, Koyle MA, Pope J, Shapiro E
(2006) Endoscopic therapy for vesicoureteral reflux:
a meta-analysis. I. Reflux resolution and urinary tract
infection. J Urol 175:716–722
Fanos V, Cataldi L (2004) Antibiotics or surgery for vesi-
coureteric reflux in children. Lancet 364:1720–1722
Fettich J, Colarinha P, Fischer S, Frokier J, Gordon I,
Hahn K, Kabasakal L, Mann M, Mitjavila M, Olivier
P, Piepsz A, Porn U, Roca I, Sixt R, van Velzen J
(2003) Guidelines for direct radionuclide cystography
in children. Eur J Nucl Med Mol Imaging 30:B39–B44
Friedmacher F, Puri P (2019) Ureteral obstruction after
endoscopic treatment of vesicoureteral reflux: does the
type of injected bulking agent matter? Curr Urol Rep
20(9):49
Friedmacher F, Colhoun E, Puri P (2018) Endoscopic
injection of dextranomer/hyaluronic acid as first line
treatment in 851 consecutive Children with high grade
vesicoureteral reflux: efficacy and long-term results. J
Urol 200(3):650–655
Garcia-Aparicio L, Rovira J, Blazquez-Gomez E, Garcia-­
Garcia L, Gimenez-Llort A, Rodo J, Morales L (2013)
Randomized clinical trial comparing endoscopic treat-
ment with dextranomer hyaluronic acid copolymer
and Cohen’s ureteral reimplantation for vesicoureteral
reflux: long-term results. J Pediatr Urol 9:483–487
Gorelick MH, Shaw KN (1999) Screening tests for uri-
nary tract infection in children: a meta-analysis.
Pediatrics 104:e54
Grazioli S, Parvex P, Merlini L, Combescure C, Girardin
E (2010) Antenatal and postnatal ultrasound in the
evaluation of the risk of vesicoureteral reflux. Pediatr
Nephrol 25:1687–1692
Hafez AT, McLorie G, Bagli D, Khoury A (2002) Analysis
of trends on serial ultrasound for high grade neonatal
hydronephrosis. J Urol 168:1518–1521
Hannula A, Venhola M, Renko M, Pokka T, Huttunen NP,
Uhari M (2010) Vesicoureteral reflux in children with
suspected and proven urinary tract infection. Pediatr
Nephrol 25:1463–1469
Hansson S, Dhamey M, Sigstrom O, Sixt R, Stokland E,
Wennerstrom M, Jodal U (2004) Dimercapto-succinic
acid scintigraphy instead of voiding cystourethrog-
raphy for infants with urinary tract infection. J Urol
172:1071–1073. discussion 73–4
Hari P, Hari S, Sinha A, Kumar R, Kapil A, Pandey
RM, Bagga A (2015) Antibiotic prophylaxis in the
management of vesicoureteric reflux: a randomized
double-­blind placebo-controlled trial. Pediatr Nephrol
30:479–486
Herz D, Merguerian P, McQuiston L, Danielson C, Gheen
M, Brenfleck L (2010) 5-year prospective results of
dimercapto-succinic acid imaging in children with
febrile urinary tract infection: proof that the top-down
approach works. J Urol 184:1703–1709
Holmdahl G, Brandstrom P, Lackgren G, Sillen U,
Stokland E, Jodal U, Hansson S (2010) The Swedish
reflux trial in children: II.  Vesicoureteral reflux out-
come. J Urol 184:280–285
1303
Hunziker M, Puri P (2012) Familial vesicoureteral reflux
and reflux related morbidity in relatives of index
patients with high grade vesicoureteral reflux. J Urol
188(4):1463–1466
Hunziker M, Mohanan N, D’Asta F, Puri P (2012)
Incidence of febrile urinary tract infections in chil-
dren after successful endoscopic treatment of vesi-
coureteral reflux: a long-term follow-up. J Pediatr
160(6):1015–1020
Hunziker M, Mohanan N, Puri P (2013) Dextranomer/
hyaluronic acid endoscopic injection is effective in the
treatment of intermediate and high grade vesicoure-
teral reflux in patients with complete duplex systems.
J Urol 189(5):1876–1881
Hunziker M, Colhoun E, Puri P (2014) Renal cortical
abnormalities in siblings of index patients with vesi-
coureteral reflux. Pediatrics 133(4):e933–e937
Jodal U, Koskimies O, Hanson E, Lohr G, Olbing H,
Smellie J, Tamminen-Mobius T (1992) Infection pat-
tern in children with vesicoureteral reflux randomly
allocated to operation or long-term antibacterial pro-
phylaxis. The International Reflux Study in Children.
J Urol 148:1650–1652
Jodal U, Smellie JM, Lax H, Hoyer PF (2006) Ten-
year results of randomized treatment of children
with severe vesicoureteral reflux. Final report of
the International Reflux Study in Children. Pediatr
Nephrol 21:785–792
Kirsch AJ, Arlen AM (2014) Evaluation of new deflux
administration techniques: intraureteric HIT and dou-
ble HIT for the endoscopic correction of vesicoure-
teral reflux. Expert Rev Med Devices 11:439–446
Kirsch AJ, Perez-Brayfield M, Smith EA, Scherz HC
(2004) The modiified sting procedure to correct vesi-
coureteral reflux: improved results with submuco-
sal implantation within the intramural ureter. J Urol
171:2413-6
Koff SA, Wagner TT, Jayanthi VR (1998) The relation-
ship among dysfunctional elimination syndromes, pri-
mary vesicoureteral reflux and urinary tract infections
in children. J Urol 160:1019–1022
Kutasy B, Coyle D, Fossum M (2017) Urinary tract infec-
tion in children: management in the era of antibiotic
resistance-A pediatric urologist’s view. Eur Urol
Focus 3:207–211
Lebowitz RL, Olbing H, Parkkulainen KV, Smellie JM,
Tamminen-Mobius TE (1985) International system
of radiographic grading of vesicoureteric reflux.
International reflux study in children. Pediatr Radiol
15:105–109
Lee MD, Lin CC, Huang FY, Tsai TC, Huang CT, Tsai JD
(2009) Screening young children with a first febrile
urinary tract infection for high-grade vesicoureteral
reflux with renal ultrasound scanning and technetium-­
99m-­ labeled dimercaptosuccinic acid scanning. J
Pediatr 154:797–802
Logvinenko T, Chow JS, Nelson CP (2015) Predictive
value of specific ultrasound findings when used as a
screening test for abnormalities on VCUG.  J Pediatr
Urol 11(176):e1–e7
1304
Mane N, Sharma A, Patil A, Gadekar C, Andankar M,
Pathak H (2018) Comparison of contrast-enhanced
voiding urosonography with voiding cystourethrog-
raphy in pediatric vesicoureteral reflux. Turk J Urol
44:261–267
Mattoo TK (2011) Vesicoureteral reflux and reflux
nephropathy. Adv Chronic Kidney Dis 18:348–354
Menezes M, Puri P (2009) Familial vesicoureteral reflux—
is screening beneficial? J Urol 182:1673–1677
Moore K, Bolduc S (2014) Prospective study of
polydimethylsiloxane vs dextranomer/hyaluronic acid
injection for treatment of vesicoureteral reflux. J Urol
192:1794–1799
Nagler EV, Williams G, Hodson EM, Craig JC (2011)
Interventions for primary vesicoureteric reflux.
Cochrane Database Syst Rev 3:CD001532
Ntoulia A, Back SJ, Shellikeri S, Poznick L, Morgan T,
Kerwood J, Christopher Edgar J, Bellah RD, Reid JR,
Jaramillo D, Canning DA, Darge K (2018) Contrast-­
enhanced voiding urosonography (ceVUS) with the
intravesical administration of the ultrasound contrast
agent Optison for vesicoureteral reflux detection in
children: a prospective clinical trial. Pediatr Radiol
48:216–226
O’Donnell B, Puri P (1984) Treatment of vesicoureteric
reflux by endoscopic injection of Teflon. Br Med J
(Clin Res Ed) 289(6436):7–9
Oswald J, Riccabona M, Lusuardi L, Bartsch G, Radmayr
C (2002) Prospective comparison and 1-year follow-
­up of a single endoscopic subureteral polydimethylsi-
loxane versus dextranomer/hyaluronic acid copolymer
injection for treatment of vesicoureteral reflux in chil-
dren. Urology 60:894–897. discussion 98
Peters C, Rushton HG (2010) Vesicoureteral reflux associ-
ated renal damage: congenital reflux nephropathy and
acquired renal scarring. J Urol 184:265–273
Peters CA, Skoog SJ, Arant BS Jr, Copp HL, Elder JS,
Hudson RG, Khoury AE, Lorenzo AJ, Pohl HG,
Shapiro E, Snodgrass WT, Diaz M (2010) Summary
of the AUA guideline on management of primary vesi-
coureteral reflux in children. J Urol 184:1134–1144
Piepsz A, Tamminen-Mobius T, Reiners C, Heikkila J,
Kivisaari A, Nilsson NJ, Sixt R, Risdon RA, Smellie
JM, Soderborg B (1998) Five-year study of medical
or surgical treatment in children with severe vesico-­
ureteral reflux dimercaptosuccinic acid findings.
International Reflux Study Group in Europe. Eur J
Pediatr 157:753–758
Piepsz A, Colarinha P, Gordon I, Hahn K, Olivier P,
Roca I, Sixt R, van Velzen J, Medicine Paediatric
Committee of the European Association of Nuclear
(2001) Guidelines for 99mTc-DMSA scintigraphy in
children. Eur J Nucl Med 28:BP37–BP41
Pirker ME, Colhoun E, Puri P (2006a) Renal scarring in
familial vesicoureteral reflux: is prevention possible? J
Urol 176:1842–1846. discussion 46
Pirker ME, Mohanan N, Colhoun E, Barton D, Green A,
Puri P (2006b) Familial vesicoureteral reflux: influ-
ence of sex on prevalence and expression. J Urol
176:1776–1780
P. Puri and B. Kutasy
Preda I, Jodal U, Sixt R, Stokland E, Hansson S (2007)
Normal dimercaptosuccinic acid scintigraphy makes
voiding cystourethrography unnecessary after urinary
tract infection. J Pediatr 151: 581-4:84 e1
Puri P (1990) Endoscopic correction of primary vesico-
ureteric reflux by subureteric injection of polytetra-
fluoroethylene. Lancet 335:1320–1322
Puri P, O’Donnell B (1984) Correction of experimentally
produced vesicoureteric reflux in the piglet by intra-
vesical injection of Teflon. Br Med J (Clin Res Ed)
289:5–7
Puri P, O’Donnell B (1987) Endoscopic correction of
grades IV and V primary vesicoureteric reflux: six
to 30 month follow-up in 42 ureters. J Pediatr Surg
22:1087–1091
Puri P, Gosemann JH, Darlow J, Barton DE (2011)
Genetics of vesicoureteral reflux. Nat Rev Urol
8(10):539–552
Puri P, Kutasy B, Colhoun E, Hunziker M (2012) Single
center experience with endoscopic subureteral dextra-
nomer/hyaluronic acid injection as first line treatment
in 1,551 children with intermediate and high grade
vesicoureteral reflux. J Urol 188:1485–1489
Quirino IG, Silva JM, Diniz JS, Lima EM, Rocha AC,
Simoes e Silva AC, Oliveira EA (2011) Combined use
of late phase dimercapto-succinic acid renal scintigra-
phy and ultrasound as first line screening after urinary
tract infection in children. J Urol 185:258–263
Roshani H, Dabhoiwala NF, Verbeek FJ, Lamers WH
(1996) Functional anatomy of the human ureterovesi-
cal junction. Anat Rec 245:645–651
Roussey-Kesler G, Gadjos V, Idres N, Horen B, Ichay
L, Leclair MD, Raymond F, Grellier A, Hazart I,
de Parscau L, Salomon R, Champion G, Leroy V,
Guigonis V, Siret D, Palcoux JB, Taque S, Lemoigne
A, Nguyen JM, Guyot C (2008) Antibiotic prophy-
laxis for the prevention of recurrent urinary tract infec-
tion in children with low grade vesicoureteral reflux:
results from a prospective randomized study. J Urol
179:674–679. discussion 79
Routh JC, Bogaert GA, Kaefer M, Manzoni G, Park JM,
Retik AB, Rushton HG, Snodgrass WT, Wilcox DT
(2012) Vesicoureteral reflux: current trends in diag-
nosis, screening, and treatment. Eur Urol 61:773–782
Routh JC, Inman BA, Reinberg Y (2010) Dextanomer/
hylluronic acid for pediatric vesicoureteral
reflux:Systamatic review. Pediatrics: 125:1010–9
Sillen U, Brandstrom P, Jodal U, Holmdahl G, Sandin A,
Sjoberg I, Hansson S (2010) The Swedish reflux trial in
children: V. Bladder dysfunction. J Urol 184:298–304
Singh-Grewal D, Macdessi J, Craig J (2005) Circumcision
for the prevention of urinary tract infection in boys: a
systematic review of randomised trials and observa-
tional studies. Arch Dis Child 90:853–858
Sjostrom S, Jodal U, Sixt R, Bachelard M, Sillen U (2009)
Longitudinal development of renal damage and renal
function in infants with high grade vesicoureteral
reflux. J Urol 181:2277–2283
Skoog SJ, Peters CA, Arant BS Jr, Copp HL, Elder JS,
Hudson RG, Khoury AE, Lorenzo AJ, Pohl HG,
93  Vesicoureteral Reflux
Shapiro E, Snodgrass WT, Diaz M (2010) Pediatric
vesicoureteral reflux guidelines panel summary
report: clinical practice guidelines for screening sib-
lings of children with vesicoureteral reflux and neo-
nates/infants with prenatal hydronephrosis. J Urol
184:1145–1151
Snodgrass W (1998) The impact of treated dysfunctional
voiding on the nonsurgical management of vesicoure-
teral reflux. J Urol 160:1823–1825
Stefanidis CJ, Siomou E (2007) Imaging strategies for
vesicoureteral reflux diagnosis. Pediatr Nephrol
22:937–947
Subramaniam R, Radmayer C, Leclair MD, Kocvara R,
ESPU (2015) Pediatric urology web book. Routledge,
New York
Tekgul S, Riedmiller H, Hoebeke P, Kocvara R, Nijman
RJ, Radmayr C, Stein R, Dogan HS, European
Association of Urology (2012) EAU guidelines on
vesicoureteral reflux in children. Eur Urol 62:534–542
1305
Ural Z, Ulman I, Avanoglu A (2008) Bladder dynamics
and vesicoureteral reflux: factors associated with idio-
pathic lower urinary tract dysfunction in children. J
Urol 179:1564–1567
Verbitsky M, Krithivasan P, Batourina E et al (2021) Copy
number variant analysis and genome-wide association
study identify loci with large effect for vesicoureteral
reflux. J Am Soc Nephrol. https://doi.org/10.1681/
ASN.2020050681
Wheeler DM, Vimalachandra D, Hodson EM, Roy LP,
Smith GH, Craig JC (2004) Interventions for primary
vesicoureteric reflux. Cochrane Database Syst Rev
3:CD001532
Williams GJ, Wei L, Lee A, Craig JC (2006) Long-term
antibiotics for preventing recurrent urinary tract
infection in children. Cochrane Database Syst Rev
3:CD001534
 Posterior Urethral Valves
Salvatore Cascio, David Coyle, Simona Nappo,
and Paolo Caione
94.1 Introduction
Posterior urethral valves (PUV) are characterized
by the presence of obstructing transverse mem-
branous leaflets in the posterior urethra, which
cause severe obstructive uropathy in many
patients. Its incidence in live-born males is
approximately 1 in 4000 (Brownlee et al. 2019).
An additional burden of antenatally suspected
cases end in fetal demise or voluntary termina-
tion, with the rate of the latter varying between
jurisdictions. In the United States, a 20-year
review found that 46% of antenatally suspected
cases of PUV ended in voluntary termination
(Clayton and Brock 2014). Associated anomalies
or genetic syndromes are rare but children with
PUV have a higher incidence of both cryptorchi-
dism (16% vs 1% in general population) and
inguinal hernias (11% vs 1–5% in general popu-
lation) (Heikkila et al. 2008; Lee et al. 2020).
S. Cascio (*)
University College Dublin and Children’s Health
Ireland, Dublin, Ireland
e-mail: salvatorecascio@doctors.co.uk
D. Coyle
Children’s Health Ireland, Dublin, Ireland
S. Nappo · P. Caione
Division of Paediatric Urology, “Bambino Gesù”
Children’s Hospital, Rome, Italy
e-mail: caione@opbg.net
© Springer Nature Switzerland AG 2023
P. Puri, M. E. Höllwarth (eds.), Pediatric Surgery, https://doi.org/10.1007/978-3-030-81488-5_94
94
94.2 Embryology, Pathogenesis,
and Classification
Urethral development begins at the sixth week of
gestation in humans, with male urethral develop-
ment diverging from that of females at 9  weeks
and reaching completion at 14 weeks. During the
fifth week, the cloaca begins to be divided into the
rectum posteriorly and the urogenital (UG) sinus
anteriorly by the UG septum. At this point, the
cloacal membrane traverses the junction of the
hindgut and the ectoderm. In the sixth week, the
phallic portion of the UG sinus—which lies cau-
dally to the insertion of the mesonephric ducts—
extends distally onto the genital tubercle. The
resulting urethral plate of endodermal cells tubu-
larizes to form the penile urethra in an androgen-­
dependent manner (Krishnan et al. 2006).
The proximal urethra (prostatic and membra-
nous portions) are derived from the pelvic por-
tion of the UG sinus, immediately below the
bladder but above the insertion of the mesoneph-
ric ducts. During the seventh week, the meso-
nephric ducts migrate caudally within the sinus
to lie close together at the sinus tubercle (a.k.a.
Müller’s tubercle), itself formed by the fusion of
the caudal-most portions of the paramesonephric
ducts. Once the UG septum completes the sepa-
ration of the UG sinus and rectum, the cloacal
membrane obliterates, exposing the UG sinus
(Krishnan et  al. 2006, Standring 2016, Docimo
et al. 2017).
1307
1308
Fig. 94.1  Type 1 and Type 3 PUV according to Young’s classification
The pathogenesis of PUV has been debated
for over 150  years. The most widely accepted
model for most of the twentieth century was that
put forward by Young et  al. in 1919. They
described 3 types of PUV.  Type 1 valves are
bicuspid, extending from the verumontanum to
the anterior urethra and are thought to be over-
growths of normally present connective tissue
derived from that which invests the ejaculatory
ducts. Type 2 valves extend from the proximal
verumontanum to the bladder neck without any
obvious theory as to their pathogenesis. Type 3
valves are full circumference valves with a cen-
tral defect and were thought to arise from the fail-
ure of the cloacal membrane to fully obliterate
(Krishnan et al. 2006) (Fig. 94.1).
This model was further refined in the 1990s by
Dewan et al. based on the concept that Young’s
type 1 and type 3 valves were the same pathol-
ogy, with differences observed as a result of vari-
ation in endoscopic view and prior instrumentation
of the urethra, e.g. urethral catheterization. They
termed the unified pathology Congenitally
Obstructing Posterior Urethral Membrane
(COPUM), whereby a single membrane extends
obliquely in the sagittal plane from the distal ver-
umontanum to the roof of the urethra, the confor-
S. Cascio et al.
mation of which creates dual lumens and likely
accounts for the appearance of “Type 1” PUV at
endoscopy. The embryological basis for COPUM
is thought to be the persistence of tissue connect-
ing the verumontanum to the urethral roof from
14 weeks due to failure of apoptosis. Furthermore,
Type 2 valves were felt to be a distinct anatomi-
cal and embryological entity. Known as a Cobb’s
Collar, or Congenital Urethral Stricture, this
transverse membrane distal to the external ure-
thral sphincter is thought to arise due to failure of
complete obliteration of the cloacal membrane
(Krishnan et al. 2006, Docimo et al. 2017).
94.3 Pathophysiology
94.3.1 Antenatal
Antenatal infra-vesical lower urinary tract
obstruction results in a variety of abnormal find-
ings or varying severity depending on the degree
of obstruction. The developing kidney is espe-
cially sensitive to the effects of obstructive urop-
athy and secondary vesicoureteric reflux (VUR).
Elevated hydrostatic pressure as a result of per-
sistent hydronephrosis can lead to tubular atro-
94  Posterior Urethral Valves 1309

phy and a reduction in the number of nephrons

(hypoplasia). Progressive uropathy is associated
with inflammatory cell infiltration and fibrosis of
variable severity. It appears some patients have a
degree of primary renal dysplasia due to abnor-
malities in nephrogenesis (Jain and Chen 2019).
The impact of the high-pressure antenatal envi-
ronment created by PUV on the lower urinary
tract includes a dilated posterior urethra, a promi-
nent bladder neck, and thickened, trabeculated
bladder with or without pseudodiverticula
(Docimo et al. 2017).
A number of pathophysiological “pop-off”
mechanisms have been described, which may
function to reduce the severity of hydronephrosis
and protect renal development where present
(D’Oro et al. 2020). In patients with the “VURD”
syndrome, comprising PUV, Unilateral VUR and
renal Dysplasia, the unaffected kidney has been
shown to be normal (Fig. 94.2). This is present in
up to 15–20% of patients. Large congenital blad-
der diverticula can create additional reservoir
capacity to reduce the hydrostatic pressure in the
upper urinary tracts. A similar effect may occur
when urinary ascites develops as a result of uri-
nary extravasation, usually from forniceal rup-
ture. Fetal urine is a key component of amniotic Fig. 94.2  The “VURD Syndrome”: valves, unilateral
fluid; metanephric urine production begins at reflux, dysplasia
12  weeks gestation but fetal urine makes a sig-
nificant contribution to amniotic fluid after during cystoscopy, the bladder is found to be
18-week gestation. Amniotic fluid is essential to severely trabeculated with pseudodiverticula and
bronchial branch development in the lungs, in the ureteric orifice may be difficult to visualize.
particular during the canalicular phase between Histologically, there is detrusor hypertrophy,
16- and 28-week gestation; hence, oligohydram- smooth muscle hypertrophy with type III colla-
nios has a detrimental effect on the developing gen and elastin. A combination of these factors
lung resulting in pulmonary hypoplasia, respira- and altered neurotransmitter profile contribute to
tory distress, and increased risk of pneumothorax the clinical picture of a noncompliant, overactive,
(Fig. 94.3). Some patients will exhibit the charac- high-pressure bladder (Levin et al. 1995; Docimo
teristic facies and lower limb abnormalities asso- et al. 2017).
ciated with Potter syndrome (Levin et al. 1995;
Nasir et al. 2011).
94.3.3 Upper Urinary Tract

94.3.2 Lower Urinary Tract
Postnatally, the prostatic urethra above the
obstruction is dilated and elongated. The bladder
may be palpable as a suprapubic mass. Grossly,
Approximately 50% of patients with PUV have
VUR at the time of diagnosis due to either incom-
petence of the ureterovesical junction secondary
to high intravesical pressure or due to embryo-
logical lateral placement of the ureteric bud.
1310 S. Cascio et al.

Fig. 94.3  Timing of

renal and pulmonary Alvoolar

Surfactant
Glomorular
development and
tubular
maturation

Birth
39 weeks

Saccular
Glomerular
26 weeks formation

Fetal

Canalicular
urine
Lung
Development

13 weeks Kidny Formation:

Ureteral Budding
Penetrates
Metanophric
Mesoderm
6 1/2
weeks

Placental
transudate

0 weeks
Kidnoy Amniotic fluld Lung

Numerous publications have suggested that while
VURD syndrome may act as a “pop-off” mecha-
nism to protect the contralateral kidney, it is not
protective of renal function in the long term
(Narasimhan et  al. 2005). Renal dysplasia,
whether primary or secondary, is associated with
the presence of primitive ducts, disorganized tis-
sue architecture, foci of cartilage, and cortical
microcysts. These changes are irreversible and
renal dysplasia is considered an adverse prognos-
tic indicator of long-term renal function (Haecker
et al. 2002).
94.3.4 The “Valve Bladder
Syndrome”
The term “Valve Bladder Syndrome” was coined
by Mitchell in the early 1980s to describe the per-
sistent and progressive uretero-hydronephrosis
and loss of renal function in children who under-
went successful valve ablation. The proposed
mechanism for this is the presence of raised detru-
sor pressures during bladder filling and detrusor
overactivity, exacerbated by polyuria secondary
to lower-than-normal numbers of collecting ducts
94  Posterior Urethral Valves 1311

Bladder Outlet Obstruction

Polyuria: Detrusor Overactivity with High Intravesical

Reduced Number of Collecting Ducts + Pressure: Back Pressure on Upper Tracts
Nephrogenic Diabetes Insipitus

Antimuscarinic
Alpha-blockers
CIC
Desmopressin Reduced Bladder Compliance, Increased PVR:
Bladder Neck Incision
Nocturnal Bladder Increased Upper Tract Dilatation
Drainage

Myogenic failure-Detrusor Decompensation:

Overdistended, Hypercompliant Hypocontractile Bladder

Fig. 94.4  Pathophysiology of valve bladder syndrome

and an impaired ability of the kidney to respond to
antidiuretic hormone (nephrogenic diabetes insip-
idus). Myogenic failure, resulting from detrusor
decompensation, is characterized by a hypercom-
pliant hypocontractile bladder and is the final
component of the syndrome (Fig.  94.4). It is
widely thought that it represents a progression
from the other features. Impairment of physiolog-
ical bladder cycling plays an important role in the
pathophysiology of this condition and results
from polyuria, pseudo-­residuals from VUR, and
incomplete bladder emptying. This raises storage
pressures and worsens bladder dysfunction.
Urinary stasis increases the risk of urinary tract
infection. Detrusor overactivity and a poorly com-
pliant, never-empty bladder often manifest as uri-
nary incontinence, which, when still present by
5 years of age is an unfavorable indicator of the
likelihood of progression to chronic kidney dis-
ease (CKD) in boys with a history of PUV
(Glassberg and Combs 2016).
94.4 Diagnosis
94.4.1 Prenatal Diagnosis
Congenital anomalies of the genitourinary tract
are among the most common anomalies detected
by prenatal ultrasound with an incidence of 1 in
250 to 1 in 1000 pregnancies (Chitrit et al. 2016).
About one patient in 3 with PUV is diagnosed
antenatally (Brownlee et al. 2019). Lower urinary
tract obstruction (LUTO) is usually diagnosed
between 16 and 24  weeks of gestation and is
associated with a high perinatal mortality and
morbidity due to the high prevalence of pulmo-
nary hypoplasia and severe renal impairment.
Prenatal ultrasound has been shown to have a
high sensitivity (94%) and low specificity (43%)
for the postnatal diagnosis of PUV (Bernardes
et al. 2009). High sensitivity is essential for the
early detection and treatment of newborns with
infravesical obstruction. Different features of the
urinary tract are detected on an ultrasound scan,
indicating a possible LUTO, such as bladder size,
bladder wall thickness, posterior urethral dilata-
tion, bladder emptying, ureteral dilatation, hydro-
nephrosis, quality of the renal parenchyma
(corticomedullary differentiation, renal echo-
genicity), amniotic fluid volume, urinary ascites,
or perirenal urinoma.
The presence of an enlarged or distended
bladder, which is the defining feature of fetal
megacystis, is associated in 57% of cases with
PUV (Taghavi et  al. 2017). Different thresh-
olds of longitudinal bladder dimension (LBD)
have been used over the last 20  years to define
1312 S. Cascio et al.

megacystis that need to be correlated to ges- a

tational age and to bladder filling/emptying,
which is observed from 18–20-week gestation.
A threshold of LBD  >6, >7, and 8  mm has
been described in the first trimester and >10,
12, and 15  mm in the early second trimester
(Taghavi et  al. 2017). Fetal megacystitis is also
seen in urethral stenosis or atresia (7.4%), Prune
Belly Syndrome (3.8%), Megacystis-Microcolon
Intestinal Hypoperistalsis Syndrome (MMIHS)
(1.1%), and Cloacal anomalies (0.7%).
Additional sonographic features reported in b
association with a distended bladder include
bladder wall thickness (>3 mm) and the “keyhole
sign”, which represent a dilatation of the poste-
rior urethra in patients with posterior urethral
obstruction. The keyhole sign is not a reliable
predictor of the diagnosis of PUV, it has a high
sensitivity (94%) but low specificity (43%) for
the prenatal diagnosis of PUV (Bernardes et al.
2009). It is also present in approximately a third
of patients with other pathologies, such as VUR,
primary megaureter, ectopic obstructive ureter
Fig. 94.5  Postnatal ultrasound showing renal dysplasia
(Bernardes et al. 2009), and in a minority of cases (small hyperechoic kidneys) in a newborn with prenatal
of Prune Belly Syndrome and MMIHS (Osborne diagnosis of bilateral hydronephrosis and oligohydramnios
et al. 2011). A neurogenic bladder may also have
a keyhole appearance, most likely due to the
hypotonic bladder neck and dyssynergic vesical Table 94.1  Prenatal ultrasonographic features sugges-
tive of PUV
contraction.
Other prenatal sonographic variables have Bilateral uretero-hydronephrosis
Oligohydramnios
been shown to be poor prognostic factors and Abnormal renal parenchyma/cystic appearance/renal
being discriminative between PUV and VUR: hyperechogenicity
oligohydramnios, bilateral uretero-­ Distended bladder/fetal megacystis
hydronephrosis, and abnormal renal parenchyma Dilated posterior urethra/keyhole sign
Thick-walled bladder (>3 mm)
being more common in PUV patients (Fig. 94.5). Urinary ascites or perirenal urinoma
Moreover, anhydramnios and ascites or perirenal
urinoma are exclusively described in fetuses later
diagnosed with PUV and in none of the fetuses calcium <100 mg/dl) the better. However, a sys-
with VUR (Chitrit et al. 2016) (Table 94.1). tematic review by Morris et  al. evaluating the
Fetal urine analysis has been used to predict utility of fetal urine analysis in predicting post-
fetal renal outcome. The main function of the natal renal function has showed that elevated
fetal kidney is to clear free water, therefore the fetal urinary concentrations of sodium, calcium,
more dilute the urine (osmolality <200 mosm/L) and B2-microglobulin were not significantly
and the lower the levels of the urinary electro- predictive of poor renal function (Morris et  al.
lytes (Sodium <100 mg/dl, Chloride <90 mg/dl, 2007).
94  Posterior Urethral Valves
94.4.2 Neonatal Diagnosis
The large majority of patients with PUV are diag-
nosed soon after birth or in the first year of life,
following investigations for an antenatally
detected hydronephrosis or urinary tract i­ nfection.
Symptoms in the first year of life vary from fever,
vomiting, dehydration, failure to thrive to anuria
secondary to renal failure. A poor urinary stream
is an unreliable sign, as in some neonates, detru-
sor hypertrophy can overcome the urethral
obstruction allowing a good urinary stream.
Renal biochemistry is often abnormal at birth and
metabolic acidosis may be present. Plasma cre-
atinine measured at 1-month posttreatment has
been shown to be significantly higher in boys
with antenatally diagnosed bladder outlet
obstruction when compared to postnatal
(Brownlee et al. 2019). On clinical examination,
signs of severe bladder outlet obstruction can be
detected, such as palpable abdominal mass in the
suprapubic region from a distended bladder or in
the flank from an hydronephrotic kidney.
Abdominal distension can be due to urinary asci-
tes. Respiratory distress can also be the present-
ing sign soon after birth.
94.4.3 Delayed Diagnosis of PUV
Less than a quarter (23%) of patients with PUV
are diagnosed late, beyond 12  months of age
(Brownlee et al. 2019). Commonly, late present-
ing valves may show symptoms of urinary incon-
tinence, urinary tract infections, poor urinary
stream, dysuria/frequency, or renal impairment.
Traditionally, early presentation has been consid-
ered a poor prognostic factor suggestive of severe
obstruction, whereas delayed presentation has
been associated with a more favorable outcome.
However, several authors have reported increased
incidence of ESKD, dilated upper tract, and
increased serum creatinine in children with PUV
diagnosed after either the 1 or 2nd birthday
(Ansari et  al. 2008). The reasons why patients
with delayed presentation PUV do more poorly
1313
than those presenting early is not clear. The
authors speculate that children presenting late
may have more bladder dysfunction. This results
in loss of compliance and the development of a
small capacity high-pressure bladder.
Furthermore, persistent polyuria, resulting from
delayed presentation may worsen the bladder
dysfunction contributing to the deterioration of
the urinary tract.
94.5 Differential Diagnosis
In the prenatal age, differential diagnosis between
upper urinary tract dilatation due to PUV and
hydronephrosis due to ureteropelvic junction
obstruction (UPJO) or refluxing/obstructive
megaureter is not always possible. The presence
of other signs of infravesical obstruction (mega-
cysts, dilated urethra, oligohydramnios, renal
dysplasia) improves diagnostic accuracy. The dif-
ferential diagnosis between PUV in the neonatal
period is shown in Table 94.2.
In school-age children, other causes of blad-
der outlet obstruction must be ruled out, such as
foreskin pathology (balanitis xerotica obliterans),
meatal stenosis, urethral stricture, syringocele
(Fig.  94.6), anterior urethral valve, and neuro-
pathic bladder. A detailed history, careful inspec-
tion of the external genitalia, and a noninvasive
bladder assessment together with radiological
investigations (renal ultrasound and voiding cys-
tourethrogram) will lead to the correct
diagnosis.
Table 94.2  Differential Diagnosis of PUV in the neona-
tal period
Vesicoureteral reflux
Ureteropelvic junction obstruction
Obstructive megaureter
Ureterocele
Urethral atresia
Prune belly syndrome
Anterior urethral valves/syringocele
Megalourethra
Megacystis-microcolon hypoperistalsis syndrome
Neurogenic bladder
1314
Fig. 94.6  VCUG in a 11-month-old boy, showing a Cowpers’s syringocele (distension of the bulbourethral gland) and
left-sided VUR
94.6 Imaging
94.6.1 Ultrasound
After birth, in an uncatheterized patient, an ultra-
sound of the urinary tract will show a distended
bladder which may be associated with increased
bladder wall thickness and a dilated posterior
urethra (Fig.  94.7). Ultrasound will also give
information on the severity of unilateral or bilat-
eral uretero-hydronephrosis, tracing the dilated
ureters down to the level of the bladder, and
assess the renal parenchyma. Urinary ascites or
perirenal urinoma may be present, as they are the
S. Cascio et al.
consequence of the rupture of calyceal fornices
due to increased pressure.
94.6.2 Voiding Cystourethrogram
(VCUG)
VCUG is the gold standard for the diagnosis of
PUV (Fig. 94.8). Adequate antibiotic therapy and
the absence of urinary tract infection are manda-
tory to perform VCUG.  In newborns, transure-
thral catheterization is performed with size 6 or 8
feeding tube. Diluted contrast is injected slowly
to outline the bladder and assess its size and
94  Posterior Urethral Valves
Fig. 94.7  Postnatal ultrasound in a newborn with PUV,
showing a thick-walled bladder and bilateral ureteral
dilatation
Fig. 94.8  VCUG in a newborn, showing a dilated poste-
rior urethra, a trabeculated bladder, and multiple
pseudodiverticula
1315
shape (trabeculation, pseudodiverticulae). An AP
film is taken in the filling phase and the bladder is
filled up to capacity (Mean Bladder Capacity for
a full-term newborn: 24.6 ± 10.9 mls) (Wen et al.
2015). At bladder capacity, the urethral catheter
is slowly removed and the urethra evaluated dur-
ing micturition in the oblique views. In PUV
patients, the posterior urethra will typically
appear dilated and elongated, ending with a U
shape at the membranous urethra. Vesicoureteric
reflux is seen in less than half (43%) of boys with
PUV patients (Hennus et  al. 2012) (Fig.  94.9).
Reflux of contrast in the seminal vesicles may
occur.
94.6.3 Radioisotope Scan
Both DMSA and diuretic renogram (MAG3 or
DTPA) scans are essential investigations in the
management of PUV patients. The first is mostly
accurate in identifying small, poorly functioning
kidneys assessing split renal function and docu-
menting the progression of renal scarring
(Fig. 94.10). On the contrary, diuretic renogram
is more useful to assess drainage in cases of per-
sisting upper tract dilatation. The most com-
monly used isotope for diuretic renogram is now
99m
Tc MAG3 which relies on tubular extraction
rather than diethylenetriamine pentacetic acid
(DTPA) scan, which is excreted by filtration.
94.6.4 Urodynamics
Both noninvasive bladder assessment
(Uroflowmetry combined with EMG, Frequency/
Volume Chart) and invasive urodynamics alone
or in conjunction with fluoroscopy (video-­
urodynamics) are particular useful in the follow-
­up of patients with PUV especially after toilet
training.
1316 S. Cascio et al.

Fig. 94.9  VCUG in a newborn, showing VURD Syndrome: valves, unilateral VUR, and renal dysplasia

sterile technique, a pigtail catheter is introduced

percutaneously via a trocar ideally between the
fetal bladder (distal end of the shunt) and the
amniotic cavity (proximal end). In 2013, the
result of the only randomized controlled trial to
investigate the effectiveness of VAS in LUTO
was published (Morris 2013). The study showed
that perinatal and neonatal survival seemed to be
higher in the VAS group; however, overall clini-
cal outcomes at 1 and 2 years had a poor outlook
in both groups. Of the 31 allocated fetuses, only
10 (32%) children survived at 2 years of age (7 in
the VAS group and 3 in the conservative group),
Fig. 94.10  DMSA scan in a 6-month-old infant born 7 with moderate CKD, 1 ESKD, with only two
with PUV. It shows extensive scarring in the right kidney,
focal scarring in the lower and upper poles of left kidney. children having a normal renal function, both in
Uptake ratio measures 12% right, 88% left VAS group. Complications related to VA shunt-
ing occurred in 6 (40%) of the 15 shunted fetuses,
with 4 of the 15 (27%) pregnancies lost after
94.7 Management spontaneous rupture of membranes following
shunt insertion or dislodgment. The conclusions
94.7.1 Prenatal of the study supported the fact that the natural
pathogenesis of valve disease is severe and that
With the advent of fetal ultrasound in the early mortality and morbidity are substantial and inde-
1970s, fetal intervention has been proposed to pendent of treatment.
reduce mortality and limit or reverse renal injury The second modality, antegrade fetal cystos-
to these patients. There are two generally copy with laser ablation, has been described as a
accepted modalities for antenatal treatment of treatment option for severe LUTO. This modality
PUV.  The first one is the insertion of a vesico-­ is technically challenging but has the advantage
amniotic shunt (VAS). Under sedation and mater- of allowing visualization of the posterior urethra
nal local anesthetic, the fetus and the fetus helping in the determination of the causes of the
bladder are visualized by high-quality, high-­ infra-vesical obstruction. Under local or epidural
resolution, real-time 2D ultrasound, and with a maternal anesthesia and fetal anesthesia (fentanyl
94  Posterior Urethral Valves 131