Includes DVD

Fetal Cardiology
Embryology, Genetics, Physiology, Echocardiographic Evaluation, Diagnosis and Perinatal Management of Cardiac Diseases

Second Edition

Edited by

Simcha Yagel Norman H Silverman Ulrich Gembruch

Fetal Cardiology

SERIES IN MATERNAL-FETAL MEDICINE
Published in association with the Journal of Maternal-Fetal & Neonatal Medicine Editors-in-Chief: Gian Carlo Di Renzo and Dev Maulik

Available
1 Howard Carp, Recurrent Pregnancy Loss: Causes, Controversies and Treatment ISBN 9780415421300 2 Vincenzo Berghella, Obstetric Evidence Based Guidelines ISBN 9780415701884 3 Vincenzo Berghella, Maternal-Fetal Evidence Based Guidelines ISBN 9780415432818 4 Moshe Hod, Lois Jovanovic, Gian Carlo Di Renzo, Alberto de Leiva, Oded Langer, Textbook of Diabetes and Pregnancy, second edition ISBN 9780415426206

Of related interest
Joseph J Apuzzio, Anthony M Vintzelos, Leslie Iffy, Operative Obstetrics ISBN 9781842142844 Isaac Blickstein, Louis G Keith, Prenatal Assessment of Multiple Pregnancy ISBN 9780415384247 Tom Bourne, George Condous, Handbook of Early Pregnancy Care ISBN 9781842143230 Gian Carlo Di Renzo, Umberto Simeoni, The Prenate and Neonate: The Transition to Extrauterine Life ISBN 9781842140444 Asim Kurjak, Guillermo Azumendi, The Fetus in Three Dimensions: Imaging, Embryology and Fetoscopy ISBN 9780415375238 Asim Kurjak, Frank A Chervenak, Textbook of Perinatal Medicine, second edition ISBN 9781842143339 Catherine Nelson-Piercy, Handbook of Obstetric Medicine, third edition ISBN 9781841845807 Dario Paladini, Paolo Volpe, Ultrasound of Congenital Fetal Anomalies ISBN 9780415414449 Donald M Peebles, Leslie Myatt, Inflammation and Pregnancy ISBN 9781842142721 Felice Petraglia, Jerome F Strauss, Gerson Weiss, Steven G Gabbe, Preterm Birth: Mechanisms, Mediators, Prediction, Prevention and Interventions ISBN 9780415392273 Ruben A Quintero, Twin-Twin Transfusion Syndrome ISBN 9781842142981 Baskaran Thilaganathan, Shanthi Sairam, Aris T Papageorghiou, Amor Bhide, Problem Based Obstetric Ultrasound ISBN 9780415407281

Fetal Cardiology
Embryology, Genetics, Physiology, Echocardiographic Evaluation, Diagnosis and Perinatal Management of Cardiac Diseases
Second Edition
Edited by
Simcha Yagel MD Obstetrics and Gynecology Ultrasound Center Center for Human Placenta Research Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus and Department of Obstetrics and Gynecology Hebrew University – Hadassah Faculty of Medicine Jerusalem, Israel Norman H Silverman MD Dsc FACC Division of Pediatric Cardiology Lucile Packard Children’s Hospital Stanford University Medical Center Palo Alto, California, USA Ulrich Gembruch MD PhD Department of Obstetrics and Prenatal Medicine Center of Obstetrics and Gynecology University of Bonn Bonn, Germany Associate Editor Sarah Margalyt Cohen Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus Jerusalem, Israel

Informa Healthcare USA, Inc. 52 Vanderbilt Avenue New York, NY 10017 © 2009 by Informa Healthcare USA, Inc. Informa Healthcare is an Informa business No claim to original U.S. Government works Printed in the United States of America on acid-free paper 10 9 8 7 6 5 4 3 2 1 Library of Congress Cataloging-in-Publication Data Fetal cardiology : embryology, genetics, physiology, echocardiographic evaluation, diagnosis, and perinatal management of cardiac diseases / edited by Simcha Yagel, Norman H. Silverman, Ulrich Gembruch ; associate editor, Sarah Margalyt Cohen. – 2nd ed. p. ; cm. – (Series in maternal-fetal medicine) Includes bibliographical references and index. ISBN 978-0-415-43265-8 (hb : alk. paper) 1. Fetal heart–Diseases. 2. Fetal heart. I. Yagel, Simcha. II. Silverman, Norman H. III. Gembruch, Ulrich. IV. Series. [DNLM: 1. Heart Diseases–diagnosis. 2. Heart Diseases–therapy. 3. Fetal Heart– physiopathology. 4. Infant, Newborn. 5. Prenatal Diagnosis. WS 290 F4193 2009] RG618.F465 2009 618.3'261–dc22
2008049158

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequence of their use. No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. For Corporate Sales and Reprint Permissions call 212-520-2700 or write to: Sales Department, 52 Vanderbilt Avenue, 16th floor, New York, NY 10017. Visit the Informa Web site at www.informa.com and the Informa Healthcare Web site at www.informahealthcare.com

Composition by Exeter Premedia Services Pvt Ltd, Chennai, India Printed and bound in the United States of America.

To our wives, Neomie Yagel, Gabi Gembruch, and Heather Silverman, this volume is lovingly dedicated. ‘The only labor worth laboring for is a labor of love.’ —JS Gillette

Contents

List of video clips List of contributors Preface 1. Cardiovascular development Arnold CG Wenink 2. Cardiac morphogenesis Adriana C Gittenberger–de Groot and Robert E Poelmann 3. Cardiac anatomy and examination of specimens Cornelia Tennstedt 4. Placental implantation and development Simcha Yagel and Debra S Goldman-Wohl 5. Placental circulations Eric Jauniaux and Graham J Burton 6. The physics of ultrasound imaging Zvi Friedman 7. Technical advances in fetal echocardiography Boris Tutschek and David Sahn 8. Epidemiology of congenital heart disease: etiology, pathogenesis, and incidence Julien IE Hoffman 9. Indications for fetal echocardiography: screening in low- and high-risk populations Ulrich Gembruch and Annegret Geipel 10. 11. 12. 13. 14. 15. 16. Circulation in the normal fetus and cardiovascular adaptations to birth Abraham M Rudolph Development of fetal cardiac and extracardiac Doppler flows in early gestation Ahmet A Baschat and Ulrich Gembruch The examination of the normal fetal heart using two-dimensional echocardiography Rabih Chaoui First and early second trimester fetal heart screening Simcha Yagel, Sarah M Cohen, Baruch Messing, and Reuwen Achiron Four-dimensional ultrasound examination of the fetal heart by spatiotemporal image correlation Luís F Gonçalves, Jimmy Espinoza, Juan Pedro Kusanovic, Wesley Lee, and Roberto Romero Three- and four-dimensional ultrasound in fetal echocardiography: a new look at the fetal heart Simcha Yagel, Sarah M Cohen, Israel Shapiro, and Dan V Valsky Cardiac malpositions and syndromes with right or left atrial isomerism Rabih Chaoui

x xii xvi 1 9 19 27 41 57 83 101 111 131 153 173 185 197 219 239

viii

Contents

17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

Anomalies of the right heart Jean-Claude Fouron Pulmonary atresia with intact ventricular septum Julene S Carvalho Intracardiac shunt malformations Einat Birk and Norman H Silverman Left heart malformations Lindsey D Allan Ventricular outflow tract anomalies: so-called conotruncal anomalies Shi-Joon Yoo, Fraser Golding, and Edgar Jaeggi Aortic arch anomalies Shi-Joon Yoo, Timothy Bradley, and Edgar Jaeggi Developments in diagnosis of transposition of the great arteries Laurent Fermont Abnormal visceral and atrial situs and congenital heart disease Shi-Joon Yoo, Mark K Friedberg, and Edgar Jaeggi Diseases of the myocardium, endocardium, and pericardium during fetal life Paulo Zielinsky, Antonio Piccoli Jr, João Luiz Manica, and Luiz Henrique Nicoloso Cardiomyopathy in the fetus John M Simpson Ultrasound examination of the fetal coronary circulation Ahmet A Baschat and Ulrich Gembruch Fetal cardiac tumors Mary T Donofrio and Gerard R Martin The fetal venous system: normal embryology, anatomy, and physiology and the development and appearance of anomalies Simcha Yagel, Zvi Kivilevitch, Dan V Valsky, and Reuwen Achiron Extracardiac Doppler investigation in fetuses with congenital heart disease Annegret Geipel, Ulrich Gembruch, and Christoph Berg Electrophysiology for the perinatologist Edgar Jaeggi Fetal bradydysrhythmia Klaus G Schmidt Fetal tachyarrhythmia Ulrich Gembruch Cardiac diseases in association with hydrops fetalis Ulrich Gembruch and Wolfgang Holzgreve Mending the tiniest hearts: an overview Thomas Kohl Fetal cardiac function in normal and growth-restricted fetuses Giuseppe Rizzo, Alessandra Capponi, and Domenico Arduini Venous flow in intrauterine growth restriction and cardiac decompensation Torvid Kiserud Congestive heart failure in the fetus James C Huhta

251 269 281 291 305 329 343 347 363 375 385 401

413 427 435 449 461 483 515 531 547 561

30. 31. 32. 33. 34. 35. 36. 37. 38.

Contents

ix

39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50.

Congenital cardiovascular malformations and the fetal and neonatal circulation Abraham M Rudolph Twin–twin transfusion syndrome: impact on the cardiovascular system Jack Rychik Genetics and cardiac anomalies Eran Bornstein, David Seubert, and Mark I Evans Cardiac defects in chromosomally abnormal fetuses Jon Hyett and Alex Gooi Associated anomalies in congenital heart disease Christoph Berg, Ulrich Gembruch, and Annegret Geipel The neonate with congenital heart disease – medical and interventional management Ulrike Herberg Infants with congenital heart disease in the first year of life Andrew J Parry and Frank L Hanley Genetic counseling in families with congenital heart defects Klaus Zerres and Sabine Rudnik-Schöneborn Intrapartum evaluation of fetal well-being Yoram Sorokin and Sean C Blackwell Cardiac disease in pregnancy David Planer, Haim D Danenberg, and Chaim Lotan Maternal diseases and therapies affecting the fetal cardiovascular system Salim Kees and Eyal Schiff Congenital heart disease and the central nervous system: a perinatal perspective Amanda Shillingford and Jack Rychik Index

579 597 609 621 635 659 691 705 713 725 737 749

759

List of Video clips

Chapter 7 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 Dynamic colorization High-resolution real-time three-dimensional view of a beating normal fetal heart Multiplanar reconstruction of a virtual cardiac cycle Topographic imaging of a virtual cardiac cycle Topographic imaging of a normal heart and a heart with d-transposition of the great arteries Color tissue Doppler of a normal fetal heart at mid-trimester Pulsed-wave tissue Doppler study of a normal fetal heart Longitudinal strain in the fetal heart measured using speckle tracking Circumferential strain in the fetal heart measured using speckle tracking 87 87 88 89 89 93 93 95 96

Chapter 12 12.1 12.2 12.3 Two-dimensional echocardiography: a parallel sweep from the upper abdomen to the large thorax Apical four-chamber plane Sagittal view of aortic arch 173 177 180

Chapter 13 13.1 MPR screen of a volume data set acquired with STIC 190

Chapter 14 14.1 14.2 14.3 14.4 14.5 14.6 14.7 14.8 14.9 14.10 14.11 14.12 Two-dimensional ultrasound scan of the four-chamber view Standard planes of section Tomographic ultrasound imaging of a normal volume dataset Demonstration of technique to systematically visualize outflow tracts Demonstration of transposition of great arteries Demonstration of technique to visualize the aortic and ductal arches ‘Thick slice’ rendering of the atrioventricular valves of a normal fetus ‘Thick slice’ rendering of the atrioventricular valves in a fetus with Ebstein anomaly ‘Thick slice’ rendering in a case of absent pulmonary valve syndrome ‘Thick slice’ rendering of the right ventricle and pulmonary artery in a case of absent pulmonary valve syndrome Technique to obtain rendered images of the outflow tracts Crisscrossing of the outflow tracts 198 199 201 203 203 203 206 208 209 209 210 210

Video clips

xi

14.13 14.14

Technique to render the aortic and ductal arches Four-dimensional visualization of the aortic and ductal arches

210 211

Chapter 15 15.1 15.2 15.3 15.4 15.5 15.6 15.7 B-flow of normal heart and aortic arch Normal heart and great vessels STIC acquisition combined with IM and VOCAL analysis for fetal cardiac ventrical volumetry STIC acquisition in a case of TAPVC B-flow of the heart and great vessels in a fetus with interrupted inferior vena cava with azygos continuation IVS ‘virtual plane’ with color Doppler in evaluation of VSD B-flow modality showing the parallel great vessels in a case of transposition 220 221 227 229 229 230 231

Chapter 19 19.1 19.2 19.3 19.4 Four-chamber view of an atrioventricular septal defect Atrioventricular septal defect without attachment to interventricular septum Subarterial doubly committed ventricular septal defect Three-dimensional reconstruction of a perimembranous ventricular septal defect 285 286 288 288

Chapter 23 23.1 Inlet ventricular septal defect 343

Chapter 28 28.1 28.2 28.3 28.4 28.5 28.6 28.7 28.8 28.9 Two-dimensional echocardiographic imaging of a cardiac tumor Three-dimensional echocardiographic imaging of a cardiac tumor Two-dimensional echocardiographic imaging of a cardiac tumor Color Doppler imaging of a cardiac tumor Postnatal two-dimensional echocardiographic imaging of a cardiac tumor Postnatal color Doppler imaging of a cardiac tumor Postnatal three-dimensional ultrasound imaging of a cardiac tumor Two-dimensional echocardiographic imaging of an atrial tumor Postnatal two-dimensional echocardiographic imaging of an atrial tumor 401 401 402 402 402 402 404 405 405

Chapter 29 29.1 29.2 29.3 29.4 29.5 29.6 29.7 Fetal intra-abdominal umbilical vein varix B-flow image of fetal intra-abdominal umbilical vein varix Interrupted inferior vena cava with azygos continuation B-flow image of interrupted inferior vena cava with azygos continuation Three-dimensional ultrasound inversion mode of persistent right umbilical vein anomaly Four-dimensional ultrasound B-flow image of agenesis of ductus venosus Total anomalous pulmonary venous connection anomaly shown with high definition power Doppler and STIC acquisition 417 417 418 418 419 420 422

Contributors

Reuwen Achiron Department of Obstetrics and Gynecology Chaim Sheba Medical Center Tel Hashomer and Sackler School of Medicine Tel Aviv University Tel Aviv, Israel Lindsay D Allan Harris Birthright Centre for Fetal Medicine King’s College Hospital London, United Kingdom Domenico Arduini Department of Obstetrics and Gynecology Università Roma Tor Vergata Ospedale Fatebenefratelli Rome, Italy Ahmet A Baschat Department of Obstetrics, Gynecology, and Reproductive Sciences University of Maryland Baltimore, Maryland, USA Christoph Berg Department of Obstetrics and Perinatal Medicine University of Bonn Bonn, Germany Einat Birk Perioperative Pediatric Cardiology Heart Institute Schneider Children’s Medical Center Petach Tikva, Israel Sean C Blackwell Division of Maternal – Fetal Medicine Department of Obstetrics and Gynecology University of Texas Health Sciences Houston, Texas, USA

Eran Bornstein Department of Obstetrics and Gynecology Lenox Hill Hospital New York, New York, USA Timothy J Bradley University of Toronto Faculty of Medicine and Division of Cardiology Department of Paediatrics The Hospital for Sick Children Toronto, Ontario, Canada Graham J Burton Academic Department of Obstetrics and Gynaecology, Royal Free and University College London Medical School London and Department of Physiology, Development, and Neuroscience University of Cambridge Cambridge, United Kingdom Alessandra Capponi Department of Obstetrics and Gynecology Ospedale GB Grassi Rome, Italy Julene S Carvalho Brompton Fetal Cardiology Royal Brompton Hospital and Fetal Medicine Unit, St George’s Hospital London, United Kingdom Rabih Chaoui Center for Prenatal Diagnosis and Human Genetics Berlin, Germany Sarah Margalyt Cohen Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus Jerusalem, Israel Haim D Danenberg Heart Institute Hadassah – Hebrew University Medical Center Jerusalem, Israel

Contributors

xiii

Mary T Donofrio Children’s National Heart Institute Children’s National Medical Center Washington, DC, USA Jimmy Espinoza Wayne State University Department of Obstetrics and Gynecology Detroit, Michigan, USA Mark I Evans Institute for Genetics and Fetal Medicine St Luke’s Roosevelt Hospital Center and Department of Obstetrics and Gynecology and Genetics Columbia University New York, New York, USA Laurent Fermont Institut de Puériculture Paris, France Mark K Friedberg The Hospital for Sick Children Toronto, Ontario, Canada Jean-Claude Fouron Unité de Cardiologie Foetale, CHU Sainte Justine Université de Montréal Montréal, Québec, Canada Zvi Friedman GE Healthcare Inc. Haifa, Israel Annegret Geipel Department of Obstetrics and Prenatal Medicine University of Bonn Bonn, Germany Ulrich Gembruch Department of Obstetrics and Prenatal Medicine Center of Obstetrics and Gynecology University of Bonn Bonn, Germany Adriana C Gittenberger–de Groot Department of Anatomy and Embryology Leiden University Medical Center Leiden, The Netherlands Fraser Golding Echocardiography Laboratory Division of Cardiology Department of Paediatrics The Hospital for Sick Children Toronto, Ontario, Canada

Debra S Goldman-Wohl Center for Human Placenta Research Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus Jerusalem, Israel Luís F Gonçalves Perinatology Research Branch NICHD/NIH/DHHS Detroit, Michigan, USA Alex Gooi Prince Charles Hospital Brisbane, Queensland, Australia Frank L Hanley Division of Pediatric Cardiac Surgery University of California San Francisco, California, USA Ulrike Herberg Department of Pediatric Cardiology University of Bonn Bonn, Germany Julien IE Hoffman Cardiovascular Research Institute University of California San Francisco, California, USA Wolfgang Holzgreve Department of Obstetrics and Gynecology University of Basel Kantonspital Basel, Switzerland James C Huhta Perinatal Cardiology University of South Florida College of Medicine St Petersburg, Florida, USA John Hyett Royal Brisbane and Women’s Hospital Brisbane, Queensland, Australia Edgar T Jaeggi University of Toronto Faculty of Medicine and Fetal Cardiac Program Division of Cardiology Department of Paediatrics The Hospital for Sick Children Toronto, Ontario, Canada

xiv

Contributors

Eric Jauniaux Academic Department of Obstetrics and Gynaecology UCL EGA Institute for Women’s Health Royal Free and University College London (UCL Campus) London, United Kingdom Salim Kees Department of Obstetrics and Gynecology Chaim Sheba Medical Center Ramat Gan, Israel Torvid Kiserud Haukeland University Hospital Department of Clinical Medicine University of Bergen Bergen, Norway Zvi Kivilevitch Maccabi Health Services Women’s Health Center Ultrasound Center Beer Sheba, Israel Thomas Kohl Department of Obstetrics and Prenatal Medicine Bonn University Hospital Bonn, Germany Juan Pedro Kusanovic Perinatology Research Branch NICHD/NIH/DHHS Detroit, Michigan, USA Wesley Lee William Beaumont Hospital Royal Oak, Michigan, USA Chaim Lotan Heart Institute Hadassah – Hebrew University Medical Center Jerusalem, Israel João Luiz Manica Fetal Cardiology Unit Institute of Cardiology of Rio Grande do Sul Porto Alegre, Brazil Gerard R Martin Children’s National Medical Center Division of Cardiology Washington, DC, USA Baruch Messing Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus Jerusalem, Israel

Luiz Henrique Nicoloso Fetal Cardiology Unit Institute of Cardiology of Rio Grande do Sul Porto Alegre, Brazil Andrew J Parry Department of Paediatric Surgery Bristol Royal Hospital for Children Bristol, United Kingdom Antonio Piccoli Jr Fetal Cardiology Unit Institute of Cardiology of Rio Grande do Sul Porto Alegre, Brazil David Planer Heart Institute Hadassah – Hebrew University Medical Center Jerusalem, Israel Robert E Poelmann Department of Anatomy and Embryology Leiden University Medical Center Leiden, The Netherlands Giuseppe Rizzo Department of Obstetrics and Gynecology Università Roma Tor Vergata Ospedale Fatebenefratelli Rome, Italy Roberto Romero Perinatology Research Branch NICHD/NIH/DHHS Detroit, Michigan, and Bethesda, Maryland, USA Sabine Rudnik-Schöneborn Institute of Human Genetics University of Technology Aachen, Germany Abraham M Rudolph Cardiovascular Research Institute University of California San Francisco, California, USA Jack Rychik Fetal Heart Program Cardiac Center Children’s Hospital of Philadelphia Philadelphia, Pennsylvania, USA David J Sahn Oregon Health and Sciences University Portland, Oregon, USA

Contributors

xv

Eyal Schiff Department of Obstetrics and Gynecology Sheba Hospital Ramat Gan, Israel Klaus G Schmidt Division of Pediatric Cardiology and Pneumology Children’s Hospital University Medical Center Düsseldorf Düsseldorf, Germany David Seubert Department of Perinatology Bay State Medical Center Springfield, Massachusetts, USA Isabel Shapiro Bnai Zion Medical Center Haifa, Israel Amanda Shillingford Fetal Heart Program Cardiac Center Children’s Hospital of Philadelphia Philadelphia, Pennsylvania, USA Norman H Silverman Pediatric and Echocardiography Laboratory Division of Pediatric Cardiology Lucile Packard Children’s Hospital Stanford University Medical Center Palo Alto, California, USA John M Simpson Fetal and Paediatric Cardiology Department of Congenital Heart Disease Guy’s and St Thomas’ NHS Trust London, United Kingdom Yoram Sorokin Department of Obstetrics and Gynecology Maternal Fetal Medicine Fellowship Program Wayne State University School of Medicine Detroit, Michigan, USA Cornelia Tennstedt Department of Pathology Charité Medical Faculty of the Humboldt University Berlin, Germany

Boris Tutschek Department of Obstetrics and Gynecology University of Düsseldorf Düsseldorf, Germany Dan V Valsky Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus Jerusalem, Israel Arnold CG Wenink Department of Anatomy and Embryology Leiden University Medical Center Leiden, The Netherlands Simcha Yagel Obstetrics and Gynecology Ultrasound Center Center for Human Placenta Research Department of Obstetrics and Gynecology Hadassah – Hebrew University Medical Centers Mount Scopus and Department of Obstetrics and Gynecology Hebrew University – Hadassah Faculty of Medicine Jerusalem, Israel Shi-Joon Yoo Department of Medical Imaging Faculty of Medicine University of Toronto and Division of Cardiac Imaging Department of Diagnostic Imaging The Hospital for Sick Children Toronto, Ontario, Canada Klaus Zerres Institute of Human Genetics University of Technology Aachen, Germany Paulo Zielinsky Fetal Cardiology Unit Institute of Cardiology of Rio Grande do Sul Porto Alegre, Brazil

molecular biologists. which receives 50% of fetal blood volume. we believe. and is perhaps the main impetus for the second edition. and to professional training. and general cardiologists. and congestive heart failure. and the advances in three. commonly estimated to be 8:1000 live births. to extend the benefits of advanced prenatal cardiac screening to patients far from well-equipped centers. 3D/4D ultrasound applications have been shown to facilitate screening programs. as well as all life stages of heart disease. like the first. 3D/4D US has added virtual planes to fetal cardiac scanning: views of the fetal heart not generally accessible with a standard 2D approach. Diagnosis and Perinatal Management of Cardiac Diseases is. Many books have been devoted to detection of fetal anomalies.and four-dimensional (3D/4D) ultrasound (US) capabilities in fetal cardiac imaging. sonographers. and virtually all have been updated. will find it useful. We also discuss the often-ignored placenta. As many more are thought to die in utero or abort spontaneously. Echocardiographic Evaluation. from embryology and the genes involved in cardiac development. Motion gating technology has overcome much inherent difficulty in 3D fetal cardiac scanning. and counseling of families affected with CHD. The broad classification congenital heart disease includes the most common congenital malformations. Genetics. Physiology. and through offline networking capabilities to improve healthcare delivery systems. It is our hope that this volume will bridge among the specialties: obstetricians. pediatric cardiologists.Preface This second edition of Fetal Cardiology: Embryology. Anyone performing targeted organ scans or having an interest in the fetal heart. to parental counseling. and many to Doppler techniques. pediatric cardiologists. an indivisible part of the cardiovascular system. and is therefore in constant need of updating. and now allows near real-time 3D/4D heart examination. Over the very few years between the first and second editions. to the reproductive health of women with CHD. to interdisciplinary management team consultation. including the cardiac impact of the twinto-twin transfusion syndrome. The field of fetal ultrasonographic and echocardiographic imaging and fetal cardiac therapies and interventions is rapidly changing and developing. This is an important aspect of fully integrated comprehensive fetal echocardiography in every targeted organ scan. Simcha Yagel Norman H Silverman Ulrich Gembruch November 2008 . The burgeoning 3D/4D US–based literature shows that these modalities contribute to our understanding of the normal and anomalous fetal heart. and medical physicists. To this end we have included chapters on treatment options and pharmacological or surgical interventions available to affected fetuses. considerable strides have been made in all aspects of fetal cardiology. In two areas advances have been particularly noteworthy: the identification of additional genes implicated in CHD. This is a comprehensive guide intended for anyone interested in fetal organ scanning. Chapters have been added on many aspects of fetal cardiology. as well as future fetal treatments only now being developed. cerebral outcomes in congenital heart disease (CHD). The 3D/4D revolution has opened exciting avenues for clinical applications and research endeavors in fetal cardiology. the product of the combined efforts of many professionals: obstetricians. Most of these volumes devote only one chapter to fetal cardiac malformations. The subject of maternal status and how the mother’s treatments will affect the fetal heart is included.

Fetal Cardiology .

.

and the ventricular outlet segment is connected to the arterial trunk (aortic sac). because specific markers for each segment. which may be smooth or trabeculated. the atrium. the The venous sinus The venous sinus is the most caudal segment. Several molecular markers are known to distinguish cardiac segments.1). Much of this knowledge has been acquired from in vitro experiments using cell or tissue cultures. The field of embryology has been useful in the understanding of details of the anatomy of congenital malformations. which would remain recognizable throughout development. Currently. genetics. the primitive common pulmonary vein develops by sprouting from the dorsal wall of the left . the fields of fetal cardiology. Extrapolation of the segmental organization to the straight heart tube stage is disputable. have not yet become available. and primary interventricular transitions (Figure 1. and in slightly advanced stages by the accumulation of endocardial cushion masses on the inside and subepicardial tissue externally. and the ventricular outlet segment. been described before looping. therefore.4 The most reliable criteria to distinguish the segments in the looped heart are unfortunately still based on general morphology. atrioventricular. arranged in a caudocranial direction in the embryonic body. Although the tube starts off as a straight one.2 This approach is also useful for the description of consecutive developmental stages.1 Embryological–pathological correlations can be made only when both the development and the pathological anatomy have been adequately described. The venous sinus collects all the embryonic veins. Following the bloodstream from the venous to the arterial poles. After a sequential description of the development. Between these segments the intersegmental transitional zones are the sinoatrial. which leads to the four-chambered organ.5 In the looped heart. There has been considerable discussion about the relationship of the pulmonary veins to the venous sinus. histological and gross anatomical features of the embryonic heart are discussed in an attempt to link them with prenatal cardiac function. and molecular biology are developing rapidly. the segments are the venous sinus. Left and right sinus horns contribute to the segment and these collect all the veins of the left and right sides of the embryo. because it highlights the course of the bloodstream and. Integration of the information should lead to full understanding of the development within the embryo.1 Cardiovascular development Arnold CG Wenink Introduction The importance of the knowledge of cardiovascular embryology has changed over the years. which is anchored within the transverse septum. the ventricular inlet segment. cardiac segments can be distinguished by the presence of a series of constrictions in the myocardial tube and by the internal profile of the myocardium. however. has functional implications.3 Distinct atrial and ventricular cell lineages have. According to classical descriptions. some aspects will be separately highlighted. In the present chapter. Segments and intersegmental transitional zones in the embryonic heart Prior to septation. individual segments are best described when the looping process of the tube has taken place and the heart has adopted an S-shape. but in vivo studies of genetically manipulated mice have become gradually more important. because at that stage not all components have yet been recognized using labeling experiments. but their specificity is restricted to limited periods of embryonic development. although erroneous developmental explanations have hindered the understanding of congenital lesions. The description of the mature heart in normal individuals and in the case of congenital malformations has greatly benefited from the sequential segmental approach. the heart is a simple tube in which the blood is pumped through a number of serially connected segments.

During development the venous sinus is incorporated into the atrium. the blood continues into the atrioventricular canal (av) with superior (s) and inferior (i) endocardial cushions. which carries pectinate muscles. the inlet segment shows an opening. (a) The venous sinus. (e) The most downstream portion of the ventricular mass is the distal outlet segment (dout) which has started to be septated by two endocardial outlet ridges (arrows). In the mature heart. the sinoatrial transition gradually becomes more apparent. (c) The atrioventricular canal (av) completely drains into the ventricular inlet segment (in). which is encircled by the primary fold (pf).2 Fetal Cardiology ra l p la (a) r av s (b) i av Figure 1.7–9 Observations in the mouse embryo still claim the primary connection to be established beyond the venous sinus. opens into the common atrium with its left (la) and right (ra) atrial appendages. these ridges contain the extremities of the aortopulmonary septum (ap) which are attached to both sides of the myocardial wall of the distal outlet segment. In particular on the right side. which delimits the pulmonary venous orifice.1 pf (c) in pout (d) ap dout (e) Schematic dissection of the embryonic heart after looping. is never very prominent. Most of it contributes to the posterior wall of the mature right atrium. using HNK-1 (human natural killer antigen-1) expression as a marker for the sinoatrial boundary in the chicken and the rat. The sinoatrial transitional zone With the incorporation of the venous sinus into the atrium. those parts that are derived from the embryonic venous sinus are easily distinguished by their smooth internal wall. as mentioned above. the primary foramen. Note the close proximity of the primary fold and the atrioventricular canal. describe the pulmonary vein connecting originally to the venous sinus. A problem with nomenclature is apparent in this description. (b) From the atrium. The left venous valve is no longer discernible . Within their core.7 invagination into the right part of the atrium leads to the formation of large valves. atrium. but a small portion forms the posterior left atrial wall and will contain the pulmonary venous orifices. what is known is that the atrium in the mature heart is built from two distinct embryonic cardiac segments called the venous sinus and the atrium.6 More recent accounts.2). the venous valves. Although its left side. On its right side. (d) The proximal outlet segment (pout) receives the blood from the primary foramen (arrow). which receives the systemic veins in its right (r) and left (l) sinus horns and the common pulmonary vein (p) separately. which can be seen through the primary foramen (arrow). in contradistinction to the wall of the embryonic atrial compartment. the two embryonic compartments can be distinguished by the presence of structures derived from the sinoatrial transitional zone (Figure 1.10 This controversy is a clear example of the need for distinct cardiac segmental markers in addition to more classical methods of distinguishing between the segments.

2).3) does not explain the detailed final morphology of these valves. congenital valve pathology based on deviation of the general rp tc l OF ms ss cs ICV tv ev Figure 1. On the other hand. which becomes filled with subepicardial mesenchyme. It does explain the general architecture. It is not probable that this septation process forms the basis for the creation of separate left and right bloodstreams. from where it continues first into the terminal crest (tc) and then into the valve (ev) of the inferior caval vein (ICV) and the valve (tv) of the coronary sinus (cs). The described position at the entrance of the superior vena cava coincides with the sinus node. The basal portion of this transition may be responsible for the formation of atrial inputs to the atrioventricular node.13 Because they are compressed by atrioventricular contraction. the atrioventricular canal is represented by a groove. remains present as the valve of the inferior vena cava (Eustachian valve) and the valve of the coronary sinal orifice (Thebesian valve) (Figure 1. which courses along the posterior atrial wall to surround the entrance of the superior vena cava. it is represented by a muscular band (m) between the atrial appendage (MAW. Starting at the membranous septum (ms). many congenital malformations of the atrioventricular valves are not the result of defective valve formation but of preexisting abnormalities of the underlying ventricular myocardium. which is always absent from the atrium. but its growth is relatively slow.11 Internally.14 Externally. The Eustachian valve continues into the terminal crest. the same process also brings atrial and ventricular septal structures into continuity. The terminal crest has been implicated in descriptions of ‘internodal tracts’. The general process of atrioventricular valve formation14 (Figure 1. Distinction of the venous sinus and the sinoatrial transition from the embryonic atrial myocardium is possible because of the HNK-1 expression.15 and they are necessary for the creation of the inlet portion of the mature right ventricle and the splitting off of the ventricular inlet septum from the primary fold (see below). SCV m The atrioventricular canal The atrioventricular canal is the transitional zone that connects the atrium with the ventricular portion of the heart loop. Such bloodstreams already exist in early preseptation stages.Cardiovascular development 3 MAW by the left and right atrial appendages. medial atrial wall) and the entrance of the superior caval vein (SCV). the atrioventricular canal is invested with relatively large endocardial cushions which derive their cells from the covering endocardium by a process of endocardial–mesenchymal transformation. In preseptation stages this canal forms an important proportion of the total myocardium. in the mature heart because it is incorporated into the atrial septal structures. The relatively early development of these pectinate muscles in the embryonic atrial compartment may imply an early contractile function of the atria. Only the valve of the oval foramen (OF) and the limbus (l) of the oval fossa are of embryonic atrial origin (reproduced with permission from reference 8).2 Diagram of the opened right atrium of the mature heart to show embryonic derivation of the septal structures. with annular attachment. and they are indeed the precursors of the atrioventricular valves. Much of this intersegmental transitional myocardium is involved in the development of the conducting tissues. Straddling valves and the common valve in atrioventricular septal defects belong to this category. The right venous valve. a fibrous veil.12 Also.8 The atrium The atrium starts off as a relatively large unseptated cavity. Because the superior cushion is immediately related to the base of the atrial septum and the inferior cushion is attached to the developing ventricular inlet septum. just as has been described for the trabeculations in the embryonic ventricles. and in advanced fetal stages the canal has become no more than the boundary between atrial and ventricular structures. but the morphological differences between mitral and tricuspid valves are related to the morphology of the ventricles in which they develop. These remain recognizable by the pectinate muscles. Fusion of the endocardial cushions in the center of the canal forms the basis of this septation process. and chords attached to papillary muscles. Along the same lines. The right side of the sinoatrial transition is almost circular. neural crest cells have been demonstrated to migrate into the cushions. Most of its remnants in the mature heart are represented . The originally single atrioventricular canal is septated into two separate atrioventricular valve orifices. the cushion masses prevent regurgitation. however.

The communication has been described as an ‘interventricular foramen’. These trabeculations are thought to be the main structures that cause ventricular systole. (b) The continuity of atrial and ventricular myocardium is disrupted. the right atrioventricular The ventricular inlet segment Before septation. The cushion tissue has thinned out to form the valve leaflet (vl) and the tendinous chords (tc). (a) In the early stages. The future ascending aorta and pulmonary trunk are connected to the more distal outlet segment of the heart. that part of the ventricular septum which separates the bloodstreams coming from the two mature atrioventricular orifices. (e) The myocardial layer of the valve has completely disappeared.16 Their degree of myofibril assembly is considerably advanced when compared to that of the compact free wall. therefore.19 This term might cause confusion. The embryonic inlet segment has a sponge-like appearance because of its many trabeculations. The myocardium of the atrioventricular canal is involved in the formation of the atrioventricular node and the proximal His bundle. which is called the primary groove. all of the atrial blood is propelled into a single cavity. the primary foramen will never close. The transfer. It corresponds to an internal myocardial profile. the fold establishes the boundary between the arterial and atrioventricular regions of the ventricular mass (also known as the ventriculoinfundibular fold). making the inlet segment largely responsible for ventricular force. (d) The myocardial layer on the ventricular side of the valve gradually disappears and retracts towards the atrioventricular canal and towards the ventricular apex.e. as in the sinoatrial transitional myocardium. A special process is involved in the formation of the ventricular inlet septum. concerns the subaortic outflow tract in addition to the aortic orifice.11 The primary fold The subdivision of the ventricular mass in the looped heart is made by an external groove. Most of the primary fold will form the muscular ventricular septum. In the inner curvature of the heart loop. developmental mechanism is extremely rare. but the term ‘primitive left ventricle’ is not used here.17 In early stages. the ventricular inlet segment is considerably larger than that of the outlet segment. not yet been elucidated. Ebstein’s anomaly is an example of this category. the atrioventricular myocardium is covered by endocardial cushion tissue (c).3 Diagrams to show the mechanism of atrioventricular valve formation.4 Fetal Cardiology a s c m v p c tc vl (a) (b) (c) (d) (e) Figure 1. the myocardial volume of . This means that the process that brings the aorta above the mature left ventricle also makes the outlet segment contribute to the left ventricle. the ventricular inlet segment.20 Initially. which delimits the communication between inlet and outlet segments. Externally. Then. the primary fold. no other cavity than the outlet segment is present to the right of the primary fold. Internally. because any interventricular communication would logically be closed by the ventricular septation process.8 Any relationship between this immunohistochemical marker and the specific function of the conducting tissues has. i. the communication between the two segments within the ventricular mass is called the primary foramen. In the present account. the atrioventricular groove is filled with subepicardial sulcus tissue (s). These structures have been recognized by their HNK-1 content. which process creates a primitive flap valve consisting of cushion tissue (c) on its atrial side and of myocardium (m) on its ventricular side. To guarantee continuity of the bloodstream from the inlet segment into the outlet segment and the great arteries.18. Roughly. The embryonic heart may be described to have a univentricular atrioventricular connection. however. (c) The inner layer of ventricular myocardium is delaminated. the inlet segment may be compared with the mature left ventricle. because not all of the tissues of the mature left ventricle are derived from this segment. The remaining delaminated myocardium constitutes the papillary muscle (p). the myocardium of the atrium (a) and ventricle (v) is continuous at the atrioventricular canal. which is not the case for this foramen.

These processes involve the excavation of a right-sided inlet cavity and the The endocardium The splanchnic mesoderm also gives rise to a single endothelial plexus. the atrioventricular canal (AV) drains completely into the ventricular inlet segment (IN).12 Another source of cells in these cushions is the neural crest. The blood has to pass the primary fold (PF) to reach the ventricular outlet segment (OUT). its central portion giving rise to the ventricular structures. which is lined internally by the endocardium and externally by the epicardium.17.4). splitting of the primary fold into an inlet septum and a septomarginal trabeculation (see above).11. The cells within these ridges have originated from the covering endocardium by endocardial–mesenchymal transformation. which forms the internal endocardial . The enlargement of this inlet portion increasingly separates the right part of the primary fold from the inlet septum.21 As the atrioventricular canal.22 The tissues of the embryonic heart The heart tube. the myocardium of the distal outlet segment compresses the endocardial ridges to prevent regurgitation. because rather complex processes still have to lead to formation of the mature right ventricle.4 Diagram to show development of the inlet portion of the right ventricle. this segment is considerably smaller than the inlet segment. The proximal outlet segment looks very much like the ventricular inlet segment. The myocardium The myocardial constituent of the embryonic heart is derived from the splanchnic mesoderm. and is present in the mature heart as the inlet portion of the right ventricle. and its proportional volume and length diminish gradually. This myocardium is precociously differentiated and has particular functional properties. The ventricular outlet segment is composed of two subdivisions: the proximal outlet segment and the distal outlet segment. which forms a single horseshoe-shaped structure in front of the buccopharyngeal membrane. Initially.11 Hemodynamically. consists of three layers of different cellular composition. The main constituent is the myocardium. It may be that the ventricular inlet segment in these stages propels the blood directly into the next portion of the heart.Cardiovascular development 5 AV IN PF OUT IS OUT Figure 1. The term ‘primitive right ventricle’ should be avoided. in these early stages the heart might be compared to the congenital lesion called univentricular atrioventricular connection to the left ventricle.23 The limbs of this horseshoe contribute to the venous pole of the heart. The distal outlet segment has a smooth myocardial wall and is invested internally with thick endocardial ridges. The ventricular outlet segment The right side of the ventricular part of the heart loop is initially constituted by only the ventricular outlet segment. in which a comparatively small right ventricle is usually present. Before septation. the distal outlet segment shows only temporary growth. having initially a thin compact myocardial wall and showing many trabeculations.13 Before the presence of the arterial valves. the distal outlet segment. and the displaced portion is found in the mature heart as the septomarginal trabeculation between inlet and outlet parts of the right ventricle (Figure 1. which contribute to septal structures and to the arterial valves. The latter is pushed forward and has always to be passed by the bloodstream to reach the outlet segment (second arrow in bottom figure) (reproduced with permission from reference 20). The blood coming from the right part of the atrioventricular canal excavates a new cavity within the tissue of this fold which splits the fold into an inlet septum (IS) and the septomarginal trabeculation. which develops as a specialized portion of the vascular system. This part of the primary fold will continue to indicate the boundary between inlet and outlet segments. bloodstream seems to create an excavation in the posterior part of the fold and to split off a separate inlet septum to its left side. This excavation will enlarge.

although the processes and the tissues involved are variable. the coelomic epithelium forms the myocardium and the pericardium. At the site of the transverse septum the epicardial organ starts to protrude into the pericardial cavity and sends out vesicles.23. the myocardial wall of the heart develops as part of the coelomic wall. It appears that only a very small portion of the mature atrial septum. In addition to the endocardial–mesenchymal transformation. the space between the endocardium and myocardium is occupied by the cardiac jelly. neural crest cells have been demonstrated to invade the endocardial cushions in the distal outlet segment and to contribute to outflow tract septation. The epicardium As described above. In the text to follow. but when it descends it compresses the sinoatrial junction just to the right of the pulmonary vein. To the left of the left venous valve.25 This jelly becomes invaded by mesenchymal cells. all septal structures have to fuse to guarantee continuous left and right bloodstreams. only subheading the venous and arterial portions of the heart separately. It is this less well developed part of the sinoatrial transitional zone that is cut off from its right-sided counterpart by the atrial septum primum. which shows reflections into the pericardium at the venous and arterial poles. but forms a semicircular rim that eventually fuses with the left venous valve. the space between the myocardial and endocardial layers can be invaded by cells from extracardiac sources. it is vital to analyze the mature atrial septal structures to find out which of them belong to the atrium proper and which are of sinal or sinoatrial origin. The epicardium is of secondary origin. Septation and valve formation at the venous pole Because most of the venous sinus is incorporated into the atrium. receiving the systemic veins. other sources contribute to the cellular composition of the cushions. These flaps can be distinguished to consist of sinal and atrial layers. Evidently.24 Initially. the septation Septation and valve formation at the arterial pole Septation of the outflow tract takes place in immediate continuity with septation of the aortic sac35 (Figure 1.26 Indeed. which is also the site where the inferior parts of the venous valves become fixed.34 Septation and valve formation Complex processes transform the looped heart tube into the four-chambered organ with valves at the various transitions. It should be realized that the endocardial tube forms part of the continuous system of endothelial structures forming the veins as well as the arteries.29 processes will be dealt with in an integrated fashion. The sinus part. This structure never becomes a full septum.27 Neural crest cells also invade the inflow tract.6 Fetal Cardiology tube. At birth. and these tissues are continuous at the venous and arterial poles. As a result. not only forms the aortopulmonary septum between the ascending aorta and pulmonary trunk. which process creates prominent double structures called the venous valves.30–33 The epicardial vesicles have been described to carry mesenchymal cells from within the transverse septum. which has been demonstrated to send its derivatives into the atrioventricular cushions. Towards the pulmonary portion of the sinus. it is impractical to describe the septation processes individually for each separate segment. the systemic part of the venous sinus is kept to the right of the atrial septum primum. it forms the material to which the flap of the atrial septum primum is pressed to close the interatrial communication. The extracardiac mesenchyme. but .13 A second extracardiac source of cushion cells is the epicardium. including the early subepicardial vasculature. contained within the rims of the oval fossa.7 The posterior portion of its lower rim fuses with the inferior atrioventricular endocardial cushion. This muscular septum is completely contained within the embryonic atrium. the invaginated profiles gradually diminish and the pulmonary venous orifice is only temporarily bordered by distinct elevations. These gradually cover the entire myocardial mantle.12 The invasion leads to the formation of local masses of endocardial cushion tissue which are restricted to the atrioventricular canal and the distal outlet segment. which are derived from the endocardium. A main source of extracardiac contribution is the embryonic neural crest.1e).28 They also form the substrate for the smooth muscle cells and fibroblasts of the coronary vasculature.8 The invagination of the venous sinus into the embryonic atrium leads to the formation of several distinct profiles. between it and the atrial septum primum. which descends from the region between the fourth and sixth pharyngeal arch arteries. At the same time. where they give rise to the epicardial lining and the subepicardial mesenchyme. invaginates deeply into the right portion of the atrium. Therefore. is of embryonic atrial origin. the atrial wall still invaginates somewhat to form an internal profile which is called the atrial septum secundum. all other structures being derived from the sinoatrial transitional zone. For this reason. All segments and intersegmental transitional zones are septated into left and right portions.

588: 13–25. 7. Ikeda T et al. Gittenberger-de Groot AC. De La Cruz MV. Ann Anat 1993. 18. Wenink ACG. Diversification of cardiomyogenic cell lineages during early heart development. 19: 245–84. 26. Markwald RR. Circ Res 1995. Anat Rec 1995. 243: 93–100. Circ Res 1998. Intracardiac blood flow patterns related to the yolk sac circulation of the chick embryo. Yutzey KE. Brown NA. Pathogenesis of transposition complexes. 74: 365–7. 30. 201: 157–68. 239: 216–23. Anat Rec 1993. 250: 325–34. Development of the ventriculoarterial segment of the embryonic heart: a morphometrics study. Sinning AR. I. Poelmann RE. 207: 271–86. Thurkow EW. Gittenberger-de Groot AC. Davis CL. Embryology of the ventricles and great arteries. Alley RD. Wisse BJ. Verh Anat Ges 1980. Poelmann RE. Schulze M. Anat Rec 1981. 82: 1043–52. 175: 501–7. Paediatric Cardiology. 16. Vrolijk BCM et al. Circulation 1999. Franco D. Wilde AAM et al. 23. Gittenberger-de Groot AC. Wenink ACG. Wenink ACG. Knaapen MWM. Steding G. Growth of the myocardial volumes of the individual cardiac segments in the rat embryo. VanGroningen JP. Moorman AFM. Virágh S. 17. Early development of quail heart epicardium and . Lillehei CW. Anderson RH et al. Vrancken Peeters MPFM. Challice CE. Thoughts on concepts of development of the heart in relation to the morphology of congenital malformations. 38: 25–53. Virágh S. Anat Embryol 1996. Webb S. Anat Embryol 1992. Vrolijk BCM. 22. The downstream portions of the endocardial outlet ridges form the semilunar valves. Opthof T. Tissue Cell 1973. 13. 3. In normal development pulmonary veins are connected to the sinus venosus segment in the left atrium. Development of the cardiac conduction tissue in human embryos using HNK-1 antigen expression: possible relevance for understanding of abnormal atrial automaticity. Persisting zones of slow impulse conduction in developing chicken hearts. 199: 367–78. 28. Edwards JE. 10. Epicardium-derived cells contribute a novel population to the myocardial wall and the atrioventricular cushions. 76: 871–7. Carnegie Inst Contr Embryol 1927. Symersky P. Anderson RH. Cardiac Pathology. Wenink ACG. Smooth muscle cells and fibroblast of the coronary arteries derive from epithelial– mesenchymal transformation of the epicardium. 5. correlative morphogenetic study. Challice CE. Fukiishi Y. Goor DA. DeRuiter MC. Circ Res 1995. A subpopulation of apoptosis-prone cardiac neural crest cells targets to the venous pole. Inductive interactions in heart development: role of cardiac adherons in cushion tissue formation. and the attached myocytes will later invade the outflow tract septum which initially consisted of mesenchymal tissues only. 19. Wenink ACG. Diagnosis of congenital heart disease: the sequential segmental approach. 6: 447–62. Wenink ACG. 236: 664–70. Virágh S. Stranahan A. 11. These extensions are completely covered by the endocardial outlet ridges. Gittenberger-de Groot AC. Anat Embryol 1998. Blom NA. J Anat 1989. Poelmann RE. London: Lloyd-Luke. 24. Anat Embryol 1999. Lamers WH. 20. Anderson RH. Dev Dyn 1998. ed. 77: 216–19. Poelmann RE. Chest 1970. DeRuiter MC. 6: 800–6. 1968: 1–28. Wenink ACG.36 References 1. 198: 317–29. In: Becker AE. Hogers B. 25. Experientia 1988. Becker AE. Seidl W. Baasten AMJ et al. 12: 216–25. Am J Cardiol 1963. In: Watson H. Anat Rec 1994. 212: 373–84. 15. Development of the atrioventricular valve tension apparatus in the human heart. Wessels A. Anderson RH. 193: 559–67. Anat Rec 1995. Development of myocardial fiber organization in the rat heart. Oosthoek PW. Mammalian neural crest and neural crest derivatives. Ruberte E. Sánchez-Gómez C. Mjaatvedt CH. Patterns of expression in the developing myocardium: towards a morphologically integrated transcriptional model. Wenink ACG et al. DeJong F. Kluth D. HNK-1 expression patterns in the embryonic rat heart distinguish between sinuatrial tissues and atrial myocardium. 201: 39–50. Van Mierop LHS. 12. Morris-Kay G. VanderPlas-de Vries I et al. Multiple functions in heart development? Dev Biol 1999. DeRuiter MC et al. Mikawa T. Kálmán F. The development of the interventricular septum of the human heart. Development of the murine pulmonary vein and its relationship to the embryonic venous sinus. 185: 461–73. Bader D. 31. Groenendijk PM. 2. 165: 121–31. Gittenberger-de Groot AC. Development of the inlet portion of the right ventricle in the embryonic rat heart: the basis for tricuspid valve development. Development of the human heart from its first appearance to the stage found in embryos of twenty paired somites. Gittenberger-de Groot AC. Vrancken Peeters MPFM. The development of the myocardium and endocardium in mouse embryos. Gittenberger-de Groot AC. During expansion of the outflow tract the parts of its myocardial wall that are attached to the aortopulmonary septum seem to be kept together. 6. 21. Cardiovasc Res 1998. 29. 4. Anat Embryol 2000. Anat Rec 1998. 58: 453–67. The primitive cardiac regions in the straight tube heart (Stage 9) and their anatomical expression in the mature heart: an experimental study in the chick embryo. Mentink MMT et al. 44: 951–60. Die Enstehung des Endocards. 1983: 9/2–12. The architectural development of the early mammalian heart.Cardiovascular development 7 also sends two extensions toward the myocardial wall of the distal outlet segment. Palomino MA. 243: 84–92. Edinburgh: Churchill Livingstone. Knaapen MWM. 14. Mentink MMT. Untersuchungen an Hühnerembryonen. DeRuiter MC. Los JA. Neural crest cells in outflow tract septation of the embryonic chicken heart: differentiation and apoptosis. Krug EL. Vrolijk BCM. 71: 240–50. Circ Res 1992. Embryology. 9. Poelmann RE. 27. 8. eds. The origin of the epicardium and the embryonic myocardial circulation in the mouse. Ann NY Acad Sci 1990. Kausel HW.

34. Gittenberger-de Groot AC. 36. Development of the subepicardial mesenchyme and the early cardiac vessels in the dogfish (Scylorhinus canicula). VandenEijnde SM. 188: 381–93. Wenink ACG. Gittenberger-de Groot AC. Anat Embryol 1995. Bartelings MM. J Exp Zool 1996. Anat Rec 1995. Cytokeratins as a marker for epicardial formation in the quail embryo. Changes in the endothelial morphology of the developing semilunar heart valves. Int J Cardiol 1989. 167: 67–83. 35. 191: 503–8. The outflow tract of the heart. Hurle JM. Embryologic and morphologic correlations. Ramos C et al. . Muñoz-Chápuli R. Origin of subepicardial cells in rat embryos. Anat Embryol 1983. Anat Embryol 1993. Poelmann RE. Macías D. 32.8 Fetal Cardiology associated vascular and glandular structures. Vermeij-Keers C. 33. 242: 96–102. Mentink MMT. 275: 95–111. Vrancken Peeters MPFM. Colvee E. 22: 289–300.

There is most probably a dual origin of the endocardial cells inside the myocardial tube. One has to take into account. and. The crucial processes in human cardiac development take place within the first 6 weeks of embr yogenesis. therefore.13 .1b and 2. The increase in ultrasound technology to detect morphology and pathomorphology of the growing heart requires more insight into the morphogenetic and epigenetic pathways underlying normal and abnormal development. Knockout studies of genes that are essential for primary cardiogenesis lead to early embryo lethality in mice. as basic principles of heart formation can be compared between various animal models and human development. more recently. This chapter will describe in brief the major events in cardiac development.7 The formed primary heart tube is never completely symmetric (Figures 2. lined on the inside by atrioventricular cushions.4 Primary cardiogenesis The primary heart tube develops from the precardiogenic mesoderm (Figure 2. and as such cannot be followed using in vivo diagnostics.1c). It is.2a). These cardiogenic plates fuse rostrally in the midline and form a crescent-shaped structure. Furthermore. for example. however. It is more often the case that scientists searching for important genetic pathways in development unravel unexpectedly a role for a gene in question in cardiac development.5 mutations.2 There will be a focus specifically on the continuous recruitment of myocardium from the second heart field3 and on extracardiac cellular contributions to the heart and their modulatory role. important species differences such as. a right-sided aortic arch system in birds. consisting of differentiated precardiac mesoderm and splanchnic mesoderm-derived cells.10 Data on the primitive cardiac segments are somewhat confusing.2. still imminent that essential knowledge is incorporated from animal models such as mouse. an atrioventricular canal. and genetic determinants of sidedness8 and cardiac looping9 are present.2 Cardiac morphogenesis Adriana C Gittenberger–de Groot and Robert E Poelmann Introduction Cardiovascular development and the regulatory mechanisms underlying this major embryonic event have become essential knowledge for the fetal cardiologist. The ever-increasing information from experimental (transgenic) animal models allows insight into the role of many genes and proteins.5 The endoderm of the underlying developing primary gut plays an important inductive role6 through signaling molecules such as the bone morphogenetic protein (BMP) and fibroblastic growth factor (FGF) families.1c and 2. Already in the cardiogenic plate stage. but most recent data based on extensive and minute tracing studies are in favor of a small atrial compartment. however. which constitutes the primary myocardial heart tube (Figure 2. This research. those described for Nkx2. is mostly not directed by the need to understand and solve the basis of cardiac malformations.11 and a primitive left ventricle (Figures 2. Heterozygous mutations of some of these genes in the human population can lead to congenital malformations such as. as compared to a left-sided system in mammals. predetermination of future cardiac segments can be distinguished. which is not easily solved1 and which will undoubtedly continue with current new discoveries. the various converging fields of research have resulted in a confusing use of terminology.1b). The inner lining of this tube is formed by cardiac jelly and endocardial cells that are connected to the endothelium of the embryonic vascular plexus. chicken. It is essential to link genetic and epigenetic (environmental) clues from patient material to basic genetic knowledge to make progress in understanding the complicated interactive processes that direct heart development.1a) located bilaterally in the splanchnic layer of the lateral plate mesoderm of the embryo. for instance.12 In the human embryo this primary heart tube already starts to beat with peristaltic contractions at 3 weeks of development. zebrafish. and the lack of cardiac septation processes in fish with only a two-chambered heart tube as a final result.

and it is necessary to describe details at the outflow tract (arterial pole) and inflow tract (venous pole) separately. for example. (2) the complete arterial outflow tract with proximal and distal outflow tract cushions (the anterior heart field). pulmonary trunk.2a and 2.19 This gene is expressed in the second heart field mesenchyme at the arterial pole and .3 LacZ reporter gene experiments have shown that part of the outflow tract as well as the major part of the atria are derived from this heart field. ra. left ventricle. SCV. aorta. this myocardium can only reach the heart tube at the arterial pole and the venous pole. pulmonary stenosis and atresia.2a).3). and (3) the arterial outflow tract lined by the distal outflow tract cushion (secondary heart field). Ao. and atrioventricular node (*). This spectrum is ideally exemplified in the human 22q11 deletion syndrome that also shows other neural crest cell-influenced abnormalities in. based on neural crest ablation experiments18 this cell type was held responsible for many cardiac outflow tract malformations such as persistent truncus arteriosis. explaining that the secondary heart field is the distal part of the anterior heart field and that both are part of the second heart field.10 Fetal Cardiology brain paa pc pc (a) (b) as well as elegant tracing of cell clones in mouse embryos12 have proved that essential parts of the cardiac myocardium are newly recruited. The most essential gene in the 22q11 deletion syndrome is Tbx1. superior cardinal vein.3 provides a representation of this confusing nomenclature. la. showing the first pharyngeal arch arteries (paa). icv. Many textbooks have indicated such a schematic representation for simplicity. tetralogy of Fallot. and aortic arch malformations. for example. double outlet right ventricle. several temporospatial patterns for recruitment of myocardium of the second heart field have been distinguished in recent literature: (1) the complete right ventricle including the complete arterial outflow tract. This player is the neural crest cell derived from the crest of the neural tube. sinus node (sa). Pu la oft SCV sa ra a Ao lv Ao Recruitment of myocardium at the arterial pole lv (c) rv icv (d) Figure 2. inferior cardinal vein. right ventricle. The splanchnic mesoderm forming the primary heart tube is referred to as the first heart field. and migrates through the splanchnic mesoderm of the second heart field. the face and thymus.14 Earlier marker experiments in chicken embryos by De La Cruz et al15 and more recent work of Moreno-Rodriguez et al16 Dependent on the marker experiments and reporter gene constructs. The emergence of genes and gene patterns during development allows us to distinguish some of them as marker genes for specific cardiac progenitor lines. and the outflow tract (oft). rv. Until very recently. lv. left atrium. right atrium. Segmentation is complete. (b) The primary heart tube is fed by the sinus venosus with the left and right cardinal veins (blue). (c) The segmental organization of the heart tube allows one to distinguish the primitive atrium (a). While the dorsal mesocardium is interrupted in its mid-portion (Figure 2. Pu. The intersegmental transitional rings are depicted in various colors. (d) In an adult heart the transitional rings can still be traced. Other experimental studies show a refinement of this myocardial addition. Note development of the nodal tissue.1 (a) Primitive streak (*) stage embryo showing the right and left precardiac mesoderm (pc). the future aortic arch. and the newly recruited myocardium derives from the second heart field (Figures 2. For determination of the second heart field. the Isl-1 gene is an important marker. The latter also comprises the area of the developing pharyngeal arch arteries forming.17 Figure 2. To fully understand the possible clinical relevance of the above information it is necessary to introduce an essential further extracardiac cell population. The arterial pole is depicted in red. Secondary cardiogenesis and organogenesis That the primary heart tube is not a small homunculus in which all future segments of the heart are already present is becoming more and more clear. primitive left ventricle (lv).

contributing to both poles of the heart tube. SAN. paa. atrioventricular canal. pericardial cavity. left ventricle.2 (a) Mid-sagittal view of the cardiac region. Note that the posterior heart field has contributed to both the outflow tract (oft) and atrial myocardium. shf. PV. peo. right ventricle. SV. dAo. DOT. These are separated by a breakthrough of the pericardial cavity (*). OFT. LV. . pc.Cardiac morphogenesis 11 rv lv pc oft a paa paa shf dAo peo ls (a) g (b) Figure 2. PEO. dorsal aorta. atrium. rv. AVC. left ventricle inflow. CCS. left sinus horn. sinoatrial node. (b) Neural crest cells depicted in blue have been carried along by the shf.3 This illustration incorporates the various lineages as derived from the literature. Scientific World Journal 2007. inflow tract. ggL. gut. cardinal veins. 7: 1777–98. outflow tract. right ventricle outflow. a. central conduction system. proepicardial organ. g. distal outflow tract. POT. lv. pharyngeal arch arteries. proximal outflow. RV. pulmonary veins. the oblique sinus. Modified after H Lie-Venema et al. and shows the contributions from the first and second heart fields as well as the neural crest to the definitive heart. IFT. CV. proepicardial organ. pharyngeal arch arteries. sinus venosus. cardiac ganglia. second heart field. PAA. ls. Cardiac neural crest PAA Neural plate Secondary heart field DOT OFT Anterior heart field POT RV ggL Isl-1First heart field Splanchnic mesoderm Isl-1+ Second heart field LV AVC atria Posterior heart field Atria IFT SAN CCS SV PV&CV CCS PEO Figure 2.

Epigenetic factors such as hyperglycemia in diabetic pregnancies may likewise lead to similar malformations.2b). The specific sensitivity of the pulmonary outflow tract myocardium might relate to distinct genetic coding areas in the subpulmonary and subaortic outflow tract region28 (Figure 2. through abnormal interaction with the surrounding pharyngeal arch mesenchyme and endothelium. . the outflow tract cushions in blue.12 Fetal Cardiology not in the neural crest cells. and a persistent truncus arteriosus (PTA) or common arterial trunk will be the result. As a consequence. This area shows increased myocardial apoptosis and abnormal expression in the Notch and Jagged pathway.(b) The same view of a VEGF120/120 mouse.5-day-old vascular endothelial growth factor (VEGF)+/+ mouse embryo.20 Essential genes such as endothelin1. In a vascular endothelial growth factor (VEGF)120/120 model27 an aberrant upregulation of VEGF120 in the subpulmonary outflow tract is present. It explains that a great number of genes expressed in either cell population can lead to comparable cardiac malformations. including the possible role of the cytoskeleton through an endocardial/endothelial cilia system.23 Relevance for understanding outflow tract congenital malformations The above findings imply that many genetic as well as environmental pathways can lead to the spectrum of outflow tract malformations. The area with apoptotic cells is represented in pink.29. broadening immensely our scope of understanding of the pathomorphogenesis of outflow tract anomalies. In combination with factors already described for dextroposition of the aorta. Hampering myocardial recruitment at the outflow tract such as seen in Tbx1 mutations will lead to outflow tract shortening and abnormal remodeling of the inner curvature myocardium. the aortic orifice is not properly wedged in between the atria and not brought into correct position while connecting to the left ventricle.4 (a) Ventral view of a 3D reconstruction of a 12. However.25 A recent observation of our own group26 sheds some light on development of the tetralogy of Fallot with subaortic stenosis. The resulting abnormalities are dextroposition of the aorta with a double outlet right ventricle as its most severe form (the most common anomaly in animal models). The pulmonary trunk (Pu) is depicted in orange. as in the Pax3 knockout mouse. no outflow tract septation occurs at all.30 VEGF+/+ Pu Pu VEGF 120 120 lv rv (a) (b) Figure 2.24 If the neural crest cells are primarily abnormal. and shows hyperplastic outflow tract cushions leading to marked subpulmonary stenosis. The subpulmonary outflow tract is shortened. such as seen in the TGFβ2 knockout mouse. Note the decreased size of the pulmonary trunk.4 shows VEGF120/120 three-dimensional (3D) reconstruction). Kruppel-like factor 2 (KLF2). In these cases. A last epigenetic candidate that is emerging as an important factor in outflow tract remodeling is the role of shear stress on the wall and pharyngeal arch arteries. most probably affecting second heart field mesenchyme as well as neural crest cells. This leads to the conclusion that disturbed interaction of second heart field and neural crest cells is essential for the spectrum of anomalies. the development of tetralogy of Fallot can be clarified.21 and transforming growth factor (TGFβ1)22 might have their effect through mechanical sensors. neural crest cells can still migrate normally and reach the pharyngeal arch arteries and the endocardial outflow tract cushions (Figure 2. the outflow tract myocardium and its endocardial cushions do not induce normal aortic arch formation and outflow tract septation. Research is now focusing on the cell–cell and cell–matrix interactions of outflow tract populations.

2a and 2. pv.5-negative area (green) between the cardinal veins and wrapping around the sinus venosus.5-day-old mouse from dorsal.2b) that play a less conspicuous part compared to the outflow tract. migrating through the dorsal mesocardium into the atrioventricular cushion mesenchyme that takes up position surrounding the developing atrioventricular conduction system.5-negative area (green) is seen as a U-shaped part of the mesoderm connecting the left and right cardinal veins.36 has been contested in the literature. The first of these are the neural crest cells (Figure 2. they form a marked ring around the pulmonary venous anlage. Migration of the neural crest cells to the inflow tract. we and others have discovered that this sinus venosus myocardium is initially Nkx2.38 Epicardium A more prominent role at the inflow tract is seen for the development of the epicardium. The left and right sinus nodes are indicated (*).5). as determined by retroviral lacZ tracking studies in chicken embryos. left and right atria. Following a process of epithelial-to-mesenchymal transformation (EMT). There is still a large Nkx2. In our opinion. pulmonary vein (pink).34 cv oft la ra cv pv (a) (b) Figure 2. ventricular outflow tract (brown).36 These cells go into apoptosis.5 (a) A 3D reconstruction of an 11. and are postulated to have an inductive effect upon differentiation of the atrioventricular conduction system.33 (Figure 2. oft. cardinal veins (blue). Nkx2.41 After covering the heart.Cardiac morphogenesis 13 Recruitment of cardiac cells at the venous pole At the venous pole the growth of the atria including the incorporation of sinus venosus myocardium in the dorsal wall of the atria is an important mechanism (Figures 2. (b) A 13.5-negative31.39 The epicardial cells migrate over the naked myocardial heart tube.3). even in neural crest reporter mice such as the Wnt-1 Cre mouse. whereas the right node (*) has expanded considerably.2a and 2. The left sinus node has become inconspicuous.3 In contrast to the outflow tract this area and the derived sinus venosus myocardium have specific gene expression patterns including Tbx18. in parallel with the anterior heart field at the outflow tract.33 and BMPs. these EPDCs migrate Neural crest cells At the inflow tract there are two extracardiac cell populations that deserve some special attention. the epicardial cells undergo EMT in which a mesenchymal subepicardial layer is formed from epicardium-derived cells (EPDCs)42 (reviewed in references 43 and 44).5). we have reserved the term posterior heart field for this region.35 We have also described a population of neural crest cells with undetermined differentiation. Note the size difference.31 Shox2. the epicardial cells derive from the posterior heart field and its covering coelomic wall mesothelium (Figures 2.25 Our own studies have shown that neural crest cells reach the inflow tract. Subsequently. Furthermore.5-day-old mouse embryo from dorsal. Recently. cv.40 It is evident that retinaldehyde dehydrogenase (RALDH) and retinoic acid play an important role in guiding this process. .5-negative but podoplanin-positive areas form a U-shaped band that partly surrounds the cardinal veins on both left and right sides and includes the central area of the venous confluence (Figure 2. The Nkx2. Major congenital abnormalities have not been reported from avian ablation studies18 or neural crest cell-specific knockout mice. Therefore. the right one being larger.37 but was recently confirmed in two mouse reporter strains. la/ra.3).32 podoplanin. where they are important for development of the sympathetic and parasympathetic innervation.11 these mesothelial cells differentiate into the already described sinus venosus myocardium and form an epithelial structure referred to as the proepicardial organ (PEO) in the avian embryo. Tracing of Isl1-positive cells shows the extent of the incorporation.

47 formation of the main coronary arteries.33 The sinus venosus myocardium is transformed during embryonic development into what we refer to as the sinoatrial conduction system (Figure 2. and reveal interesting information on atrial isomerism and differentiation of the sinus venosus myocardium.49 and differentiation of the Purkinje network50 (Figure 2. This encompasses the anlage of a left sinoatrial nodal region. including the sinoatrial node.6 Part of the myocardial wall covered by the epicardium. The differentiation capacities of the EPDCs are still under investigation. cardiomyocytes. the septum spurium. cm.46 We have shown by PEO inhibition and rescue studies that EPDCs are essential for formation of the compact myocardium.1d). and the right (future terminal crest) and left (incorporated into the atrial system) venous valves and their confluence.14 Fetal Cardiology atrium av avc ventricle pc ec smc fb cm Cardiomyocytes Epicardium Endocardium Figure 2. and the development of parts of the conduction system might be disturbed when sinus venosus myocardium is not properly differentiated. In the inset: EPDCs have gathered around the coronary endothelial tubes (ec) to differentiate into the smooth muscle cells (smc) and adventitial fibroblasts (fb). in which expression of the matrix protein periostin by EPDCs plays a prominent role. into the atrial and ventricular myocardium to form the interstitial fibroblast and also take up a subendocardial position (Figure 2. with a main controversial issue being whether they are also the source of the coronary endothelial cells45 or whether the endothelial cells use the subepicardial space as a migration matrix but are otherwise derived from the liver endothelium. atrioventriucular cushion. av. part of the myocardial cells surrounding the pulmonary veins (Figure 2. A second wave of EMT is seen when the coronary capillary plexus is remodeled into an arterial and venous system in which the EPDCs are the source of smooth muscle cells and periarterial fibroblasts. The epicardium-derived cells (EPDCs) in gray have migrated between the cardiomyocytes and have differentiated into cardiac fibroblasts. Related genes such as Pitx2 are being studied.5).6). atrioventricular groove. avc. surrounding the coronary sinus.31 From our observations it can be deduced that both formation of the atrial septum. As EPDCs migrate through the atrioventricular sulcus into the endocardial cushions.51 Relevance for the understanding of congenital malformations Severe developmental inflow tract malformations are seen in atrial sidedness. The myocardial-derived Purkinje cells (pc) are depicted in green and are in close association with EPDCs.6). There is an indication that they are important for embryonic atrioventricular dissociation and formation of the atrioventricular fibrous annulus. in which the left venous valve is incorporated.48. we are currently studying their role in these areas. It is very important to emphasize that use of the term .

49 There is no major role for the persistence of embryonic ventricular communications. as once a cardiac abnormality develops it will automatically evoke flow disturbances. Our current feeling is that in the development of cardiac malformations we should more prominently include environmental (epigenetic) factors as seen in diabetes29. The inductive and formative role of EPDCs also sheds light on a number of cardiac malformations. Acknowledgments Nynke van den Akker. including pinpoint orifices and complete absence of main coronary stems. as suggested.58 Formation of the fibrous annulus and its electrical isolating capacities might be essential for the switch from base-to-apex conduction propagation. 3. because they do not exist in normal early life. Ron Slagter and Jan Lens are acknowledged for artwork. Wenink ACG. Fetal ultrasound diagnostics revealed the early presence of fetal VCAC in combination with pulmonary stenosis that later on developed into pulmonary atresia. and Nathan Hahurij are greatfully acknowledged for help with some of the figures. as the above described areas are clearly recognizable as arrhythmogenic foci in the adult heart and might reflect dedifferentiation of embryonic cells reinitiating conduction system capacities52 that become aberrant. DeRuiter MC et al. In cardiomyopathies the interaction of myocardium and the EPDC-derived interstitial fibroblast is essential. Anderson RH. Pediatr Res 2005. Shi Y et al. Recently. Clinically. as well as abnormally thin myocardium. 25: 255–64.33 The role of EPDCs in formation of the compact myocardium might be of relevance for the understanding of some cardiomyopathies. Most prominent are the anlage and differentiation of the coronary vascular system. and the presence of large VCAC.56 and also by our own observations in the SP3 knockout mouse. will develop to sustain myocardial perfusion. this was shown to be effective in the RXRα knockout mouse. Bartelings MM. there will be the necessity to merge the amount of data from molecular cardiac development programs with the increasing knowledge from large genomic data screens from human cardiac malformations. Concluding remarks In conclusion. Liang X. and Joke van Benten for secretarial assistance. Dev Cell 2003.48.or neural crest cell-derived. Basics of cardiac development for the understanding of congenital heart malformations. Pexieder T. Defects in either or both might lead to abnormal myocardial differentiation. 2. and the inductive role in Purkinje fiber differentiation. We have shown that partial inhibition of EPDC differentiation can lead to abnormalities of the main coronary arteries. A similar disturbance of EPDC function on the borderline of the atrioventricular cushion and the underlying myocardium might inhibit delamination of the tricuspid valve and subsequent development of the Ebstein malformation. References 1. In a clinical setting this background knowledge is important for differentiating between cases with pulmonary atresia without ventricular septal defect (VSD) that can present with or without VCAC. . a spectrum of rhythm disturbances can be explained from the disposition of the sinoatrial conduction system.53 This notion might be the erroneous origin of the so-called Thebesian vessels.57 The last aspect of EPDC involvement is in differentiation of the fibrous heart skeleton. Isl1 identifies a cardiac progenitor population that proliferates prior to differentiation and contributes a majority of cells to the heart. A suggested nomenclature for the developing heart. Int J Cardiol 1989. might play an important role in differentiation of the fibrous heart skeleton and formation of the atrioventricular fibrous annulus. We have shown that the gene periostin. Gittenberger-de Groot AC.30 and hyperhomocysteinemia60 as well as hemodynamic factors.54 Histopathology showed already severe coronary malformations with intimal thickening and even complete closure of vessels. Advancing refinement of prenatal ultrasound diagnostics and the tools to actually quantitate flow and shear stress during development will broaden our mechanistic insight.55 We hypothesize that these fetuses suffer from a primary coronary vessel (EPDC-derived) anomaly. including spongy myocardium or myocardial non-compaction. This annulus differentiates slowly.21 The greatest challenge is to sort out cause and effect.51 Relatively late progress of this isolation process during formation of the right atrioventricular orifice (Chapter 1 and reference 59) might explain transient late fetal rhythm disturbances and Wolff–Parkinson–White syndrome. the atrioventricular valves. and myocardial cells most probably transdifferentiate into a fibrous phenotype influenced by periostin-producing EPDCs.Cardiac morphogenesis 15 sinoatrial conduction system only links developmentally correlated structures. whereas in cases with primary pulmonary atresia and a normal coronary vasculature the original problem might have been second heart field. aberrant connections with the cardiac lumen. A functional capacity has only been reported during normal development for the right sinoatrial nodal region after an initial start on the left side. Cai CL. 5: 877–89. so-called fistulae or ventriculocoronary arterial communication (VCAC). Edris Mahtab. expressed by EPDCs. 57: 169–76.49 In the latter case.48.

Molecular regulation of atrioventricular valvuloseptal morphogenesis. 6. Unilateral vitelline vein ligation alters intracardiac blood flow patterns and morphogenesis in the chick embryo. 18. Molin DGM. Lough J. Wisse LJ et al. Tetralogy of Fallot and alterations in VEGF. 10. 14. Bader D. Gittenberger-de Groot AC. 7. Mommersteeg MT. Transforming growth factor beta-SMAD2 signaling and aortic arch development. Circ Res 2007. endocardial cushion differentiation. Colbert MC. 15. Origins and patterning of avian outflow tract endocardial tissues and cushion mesenchyme. Left/right patterning signals and the independent regulation of different aspects of situs in the chick embryo. Burch J et al. Hierck BP et al. Dev Dyn 2000. Endoderm and heart development. Christoffels VM. 38. Fate of the mammalian cardiac neural crest. 165: 121–31. 30. Meilhac SM. Expression of the atrial-specific myosin heavy chain AMHC1 and the establishment of anteroposterior polarity in the developing chicken heart. 235: 191–202. Zdanowicz M. 22. Lindsay EA. 9. Tbx1 haploinsufficieny in the DiGeorge syndrome region causes aortic arch defects in mice. A novel role for cardiac neural crest in heart development. Molin DG et al. Anat Rec 2007. The primitive cardiac regions in the straight tube heart (Stage 9) and their anatomical expression in the mature heart: an experimental study in the chick heart. 5. 35. Gittenberger-de Groot AC. Monocilia on chicken embryonic endocardium in low shear stress areas. 17. Kelly RG et al. Birth Defects Res A Clin Mol Teratol 2004. Stalmans I. Molin DGM. 8. Sanchez-Gomez C. 13. Development of a lethal congenital heart defect in the splotch (Pax3) mutant mouse. Dev Dyn 2006. Mutations in the cardiac transcription factor NKX2. Vitelli F. an effect that can be prevented by N-acetylcysteine. Krug EL. Markwald RR. 9: 173–82. Kirby ML et al. Dev Cell 2004. Circulation 2001. Gittenberger-de Groot AC. Circulation 2007. Bajolle F. 12. Rhee JT. Development 2000. Soriano P et al. Dev Dyn 2006. Groenendijk BCW. Yutzey C. Pagan S. The quintessence of the making of the heart. Formation of the venous pole of the heart from an Nkx2-5negative precursor population requires Tbx18. Distribution of different regions of cardiac neural crest in the extrinsic and the intrinsic cardiac nervous system. Exposure of neural crest cells to elevated glucose leads to congenital heart defects. Bidirectional fusion of the heart-forming fields in the developing chick embryo. 217: 191–204. 21. Targeted mutation reveals essential functions of the homeodomain transcription factor Shox2 in sinoatrial and pacemaking development. De La Cruz M. Science 1996. myocardialization. 27. Nature 2001. 28. Circ Res 1995. Vis L et al. Kelly RG. Blaschke RJ. Disturbed morphogenesis of cardiac outflow tract and increased rate of aortic arch anomalies in the offspring of diabetic rats. 16: 1–6. Hogers B. London: Mosby International. Peters PPWM et al. Moreno-Rodriguez RA. Waldo K. Gittenberger-de Groot AC. Neural crest and cardiovascular patterning.and Notchsignalling in mouse embryos solely expressing the VEGF120 isoform. Silberbach GM. Lambrechts D. eds. Nakamura T. Circ Res 1997. 235: 19–28. van den Akker NMS. Poelmann RE. 272: 671–6. Robbins J. 127: 1607–16. Trends Cardiovasc Med 2005. Mathab EAF. 15: 51–6. Molecular inroads into the anterior heart field. Nordstrand H et al. Cardiovasc Res 1997. Circ Res 2005. DeRuiter MC. DeRuiter MC. Changes in shear stress-related gene expression after experimentally altered venous return in the chicken embryo. Birth Defects Res A Clin Mol Teratol 2007. Circ Res 2006. Levin M. Poelmann RE. Development 1994. Kelly RG et al. A subpopulation of apoptosis-prone cardiac neural crest cells targets to the venous pole: multiple functions in heart development? Dev Biol 1999. 77: 211–15. Reyes L et al. Trowe MO et al. 207: 271–86. Roest PAM. Cardiology.5-negative myocardium of the posterior heart field and its correlation with podoplanin expression in cells from the developing cardiac pacemaking and conduction system. Su H et al. 96: 1291–8. 217: 327–42. . Bartelings MM et al. 80: 473–81. The clonal origin of myocardial cells in different regions of the embryonic mouse heart. Noden DM. 104: 1567–73. Roberts DJ et al. 6: 685–98. 103: 2745–52. 16. In: Crawford MH. J Anat 1989. Sugi Y.16 Fetal Cardiology 4. 100: 842–9. Azhar M. Circ Res 1995. Eisenberg LM. Hahurij ND et al. Verberne ME. Dev Biol 1997. Vrolijk J et al. 33. VanIperen L et al. The neural crest is contiguous with the cardiac conduction system in the mouse embryo: a role in induction? Anat Embryol (Berl) 2004. Double-outlet right ventricle and overriding tricuspid valve reflect disturbances of looping. 70: 927–38. Zaffran S. Molecular pathways controlling heart development. Bartram U. Circ Res 2006. Rotation of the myocardial wall of the outflow tract is implicated in the normal positioning of the great arteries.5 affect diverse cardiac developmental pathways. Palomino MA. Conway SJ. 32. 98: 1547–54. Kirby ML. DiMarco JP. Rowitch DH. 189: 57–67. Cardiol Young 2003. and apoptosis in TGFβ2-knockout mice. Hierck BP. 24. Benson DW. Srivastava D. J Clin Invest 1999. 98: 1555–63. Molin DG. Esner M. 25. Kuijper S et al. Gittenberger-de Groot AC. Circ Res 2006. Roest PAM. Hahurij ND. 29. 77: 1–6. 34. Jongbloed MR. Olson EN. Gittenberger-de Groot AC. 26. Kavanaugh-McHugh A et al. 23. 290: 115–22. Groenendijk BCW. 79: 231–5. 98: 421–8. 13: 175–83. 31. 115: 1830–8. Embryology of congenital heart disease. 36. Desmet F et al. 120: 871–83. Nkx2. 226: 72A–3A. van Iperen L. Trends Cardiovasc Med 2006. VEGF: a modifier of the del22q11 (DiGeorge) syndrome? Nat Med 2003. Jiang X. 36: 163–73. Gittenberger-de Groot AC et al. J Clin Invest 1999. 103: 1499–507. 410: 97–101. Dev Dyn 2000. Molin DG. 11. 2001: 1217–27. 20. Henderson DJ. Neural crest cells retain multipotential characteristics in the developing valves and label the cardiac conduction system. 208: 389–93. 37. Mannheimer Lecture. Waldo KL. Anat Rec 1990. 19. Van der Heiden K.

Phelps A. Blom NA et al. Perez-Pomares JM. 64: 365–73. 15: 349–55. 219: 129–41. 27: 8571–82. 191: 503–8. Perez-Pomares JM. 64: 692–703. Eralp I. Cardiac outflow tract malformations in chick embryos exposed to homocysteine. 13: 157–64. Göldner B et al. Kolditz DP. Origin of coronary endothelial cells from epicardial mesothelium in avian embryos. Ultrasound Obstet Gynecol 1997. 46. Circ Res 2005. . Lie-Venema H. 55. 45. Early development of quail heart epicardium and associated vascular and glandular structures. Circ Res 1993. Int J Dev Biol 2002. Lie-Venema H. Epicardial outgrowth inhibition leads to compensatory mesothelial outflow tract collar and abnormal cardiac septation and coronary formation. 87: 969–71. 46: 1005–13. Boot MJ. Cell Mol Life Sci 2007. DeRuiter MC et al. Poelmann RE et al. 47. Chaoui R et al. 96: 776–83. Hutson DR. 41. Mahtab EA. Bergwerff M et al. Poelmann RE et al. Gonzalez-Iriarte M et al. Cardiovasc Res 2004. Am Heart J 1933. Gittenberger-de Groot AC. Epicardium-derived cells contribute a novel population to the myocardial wall and the atrioventricular cushions. Ets-1 and Ets-2 transcription factors are essential for normal coronary and myocardial development in chicken embryos. Mentink MMT et al. Shapiro MD. 9: 194–7. Mettier SR. 53. The contribution of the proepicardium to avian cardiovascular development. in chickenquail chimeras. 96: 526–34. Circulation 2006. 44. 52. Klumpp TG et al. Persistence of functional atrioventricular accessory pathways in postseptated embryonic avian hearts: implications for morphogenesis and functional maturation of the cardiac conduction system. 73: 559–68. Cytokeratins as a marker for epicardial formation in the quail embryo. Tennstedt C. 288A: 1272–80. Mentink MMT. Development of the right ventricular inflow tract and moderator band: a possible morphological and functional explanation for Mahaim tachycardia. 54. Lie-Venema H. Circ Res 1998. 49. Jongbloed MR. Wijffels MCEF. Circ Res 2000. Circ Res 2005. 9: 143–64. 56. Prog Pediatr Cardiol 2001. 43. Embryonic conduction tissue: a spatial correlation with adult arrhythmogenic areas? Transgenic CCS/lacZ expression in the cardiac conduction system of murine embryos. Analysis of the proepicardium-epicardium transition during the malformation of the RXRalpha-/. 233: 1091–101. The nature of the vascular communications between the coronary arteries and the chambers of the heart. 115: 17–26. Bax NA et al. Van Loo PF. Purkinje fibers in the avian heart. Wisse LJ et al. Gittenberger-de Groot AC. Tennstedt C. Anat Embryol (Berl) 1993. Schalij MJ. Viragh S. 82: 1043–52. Schalij MJ et al. Jenkins SJ. 76: 809–19. Wearn JT. Jongbloed MRM.Cardiac morphogenesis 17 39. Munoz-Chapuli R et al. Mentink MMT et al. Gittenberger-de Groot AC. Kubalak SW. Sequential programs of retinoic acid synthesis in the myocardial and epicardial layers of the developing avian heart. 42. Gittenberger-de Groot AC. 40. 57. Houghton L et al. Lie-Venema H. Anat Rec 2006. Gittenberger-de Groot AC. Epicardiumderived cells are important for correct development of the 51. Int J Dev Biol 2001. Poelmann RE et al. Winter EM. Epicardium-derived cells (EPDCs) in cardiogenesis and cardiac regeneration. Poelmann RE et al. Coronary artery and orifice development is associated with proper timing of epicardial outgrowth and correlated Fas ligand associated apoptosis patterns. Differentiation 2008. Periostin expression by epicardium-derived cells is involved in the development of the atrioventricular valves and fibrous heart skeleton. van Empel LJP et al. Gittenberger-de Groot AC. Steegers-Theunissen RP. 59. Mol Cell Biol 2007. 48. Circ Res 2003. Ventriculo coronary arterial communications (VCAC) and myocardial sinusoids in hearts with pulmonary atresia with intact ventricular septum: two different diseases. J Cardiovasc Electrophysiol 2004. Vrancken Peeters M-PFM. Transcription factor Sp3 knockout mice display serious cardiac malformations.epicardium. 92: 749–56. Vrancken Peeters M-PFM. 50. Eralp I. Prenatal diagnosis of ventriculo-coronary communications in a secondtrimester fetus using transvaginal and transabdominal color Doppler sonography. Anat Embryol (Berl) 1995. Gittenberger-de Groot AC. Wijffels MC. 45: S155–6. Markwald RR et al. Eralp I. studied with antiendothelial antibodies. 60. Carmona R. Xavier-Neto J. Development of the cardiac coronary vascular endothelium. Poelmann RE. 188: 381–93. Vrancken Peeters M-PFM. Chaoui R. Dev Dyn 2005. 58. Dev Biol 2000.

.

the septum. forming the connection between the venous component and the right appendage. it is connected with the smooth-walled venous component of the atrium. it is necessary that the system for describing and diagnosing such conditions be both simple and comprehensive. The morphologically right appendage has a broad. The external wall of the atrium consists of two parts: the venous component (sinus venarum cavarum. the vestibule. The tendon of Todaro is the continuation of the venous valves. This The morphologically left atrium The morphologically left atrium is the most posterior chamber of the heart. triangular shape. The triangle between the tendon and the tricuspid valve marks the location of the atrioventricular conduction tissue. Internally. becoming the venous valves of the inferior caval vein (Eustachian valve) and the coronary sinus (Thebesian valve). The ostium is smooth-walled and the leaflets of the tricuspid valve are attached to its edge (Figure 3. which is called the Chiari network. open. and applicable to all cases. The diagnosis of congenital heart malformations is considered to be difficult.4 The septal surface consists of the floor of the oval fossa and the atrioventricular septum. The morphologically right atrium consists of the following components: the venous component. The Eustachian valve can be prominent. This need can be adequately met only by autopsy. The Eustachian valve is usually small. it is possible to observe congenital heart malformations in increasingly greater detail and at an ever earlier stage of gestation. The border of the crest is reinforced by fibrous structures. The morphologically right atrium In the normal heart. connection is marked in the morphologically right appendage by a prominent crest with pectinate muscles. there is a critical need for monitoring and confirmation of the prenatal diagnosis. which extends forward and surrounds the right wall of the aorta. on the one hand. and the appendage.2). The so-called ‘septum secundum’ does not belong to the septum proper. It emerges from the anterior part of the septal surface and curves in front of the orifice of the superior caval vein. a septal surface. or posterior part). The goal of a descriptive system of the heart is to allow the heart atria and ventricles. into which the superior and inferior caval veins. on the other. most cases of congenital heart malformation are relatively easy to diagnose.1 Since it is on the basis of ultrasound findings that decisions to terminate pregnancies are made. the characteristic features of the various components of the normal fetal heart are described. The superior caval vein is located in the roof of the atrium and enters the right atrium between the terminal crest and the superior rim of the oval fossa. it protects the mouth of the inferior caval vein and may extend to the inlet of the superior caval vein. The ostium of the tricuspid valve points diagonally to the right and can be considered as the base of the vestibule.3 Cardiac anatomy and examination of specimens Cornelia Tennstedt Cardiac anatomy Owing to the quality of prenatal ultrasound and the expanded experience of prenatal diagnosticians.2. In this chapter. in individual cases it can form a network with pronounced fenestration.3 For this reason. The coronary sinus opens above the posterior interventricular groove into the right atrium. although there are cases that are extremely difficult. the appendage. the morphologically right atrium forms the right. and an anterior part.1). with the proper systematic approach. and an appendage. In fact. the vestibule. The terminal crest runs laterally. but simply forms the surrounding wall of the atrial chambers. as well as the coronary sinus. These structures separate the orifices of the inferior caval vein and the coronary sinus from the atrial appendage. It consists of the following components: the venous component. and the pulmonary trunk to be identified and distinguished from each other. . front part of the cardiac mass (Figure 3. the aorta. but it can also be completely lacking.

coronary sinus. All the leaflets arise at the atrioventricular junction. the apical trabecular. The most characteristic feature of the tricuspid valve is the presence of tendinous cords. tricuspid valve. RA. showing the various components of the normal fetal heart. ascending aorta. which provide the best criterion for identification of the morphologically right ventricle in cases in which there is no inlet component. It consists of the following: the inlet. The connection between the smooth-walled venous component and the left appendage is narrower than in the right atrium. At the top. internally the left appendage is trabeculated. PT. Characteristic of the apical trabecular component of the right ventricle are coarse trabeculations (Figure 3. right atrium. The valve overlaps the surrounding atrial wall (the septum secundum) from above. The trabecular zone reaches the apex of the heart. which fix the septal leaflet to the ventricular septum. The morphologically left appendage has a tubular.2 Dissection showing the typical atrioventricular junction of the right heart. It has characteristic grooves at the edge and partly surrounds the pulmonary artery. In contrast with the right atrium. which generally has smoother walls than the trabecular component. AAO. pulmonary trunk. RV.20 Fetal Cardiology Figure 3. descending aorta. The leaflets of the tricuspid valve are in the septal. the venous component admits the four pulmonary veins. The three leaflets of the pulmonary valve are attached to a completely muscular infundibulum. and inferior (or mural) position. two on each side. left ventricle.3). LA. and the outlet components. The posterior wall of the infundibulum is separated from the aorta by an extracardiac space. and it is not limited by a crest. DA. and do not project into the body of the atrium. The septal surface of the left atrium is at an angle. As with the right appendage. TV. right ventricle.1 Frontal view of the heart. the crista supraventricularis. . which go from the lower right to the upper left in the ventricular mass. anterior–superior. The border between the inlet and the trabecular component is formed by the attachment of the papillary muscles. fossa ovalis. the pectinate muscles do not extend into the area of the atrioventricular junction. FO. LV. The inlet component surrounds and provides support for the leaflets and the tension apparatus of the tricuspid valve. It is considerably larger than the appendage. The surface of the vestibule of the left atrium is smooth. The morphologically right ventricle Figure 3. the leaflets of the mitral valve are attached to it. The inlet and outlet components are separated from each other in the roof of the ventricle by a prominent muscular crest. left atrium. which extends dorsally to the crux cordis. hooked shape. The outlet component (infundibulum) of the right ventricle is a muscular tube. it is in part uneven and consists of the left atrial surface of the oval fossa. ductus arteriosus. but the trabeculations are finer. DAO. CS. The pectinate muscles are less pronounced here than in the right atrium. The morphologically right ventricle occupies the greatest part of the ventral mass of the heart.

Figure 3. MV. These emerge immediately behind the aortic valve in the area of the bulbus aortae. mitral valve. with the membranous component being very narrow. the membranous septum (Figure 3. with the inlet and outlet components being separated by the anterior of the mitral valve. and then runs intrapericardially somewhat to the right. which are attached to the root of the aorta.4 Specimen in which the left ventricle is displayed to show the typical fine criss-crossed trabeculations. the inlet area septum is bordered by the cordal attachments of the septal tricuspid leaflet. and then runs caudally from in front of the thoracic spine (this section is called the descending aorta). The inlet and outlet components of the left ventricle form an acute angle.3 Section showing the typical coarse trabeculations of the right ventricle. The membranous septum lies at the point where the inlet. The interventricular septum The interventricular septum consists mainly of muscular tissue. The component consists partially of muscular. On the rightventricular side.4). The apical trabecular part of the left ventricle has fine criss-crossed trabeculations (Figure 3. The ascending aorta emerges from behind the pulmonary trunk out of the conus arteriosus of the left ventricle. The two mitral valve leaflets are connected to two groups of papillary muscles. which occupy a posteromedial and anterolateral position below the commissural areas. an apical trabecular component. The ascending aorta has only two branches: the right and left coronary arteries. and non-coronary leaflets. and outlet component. The outlet component of the septum lies below the distal part of the infundibulum. Dorsally it is incomplete. and only to a small extent of fibrous tissue. whereas they are fine on the left-ventricular side. apical trabecular component. The leaflets are separated from each other by the anterolateral and posteromedial commissures. The aortic valve has three semilunar leaflets called the right coronary. The morphologically left ventricle As with the right ventricle. the left ventricle consists of an inlet. The most characteristic feature of the mitral valve is that it has no cordal attachments to the ventricular septum. so that the mitral and aortic valves are connected to each other by fibrous tissue. The septum of the normal heart has a muscular and a membranous component. the trabeculations of the apical trabecular component of the septum are rough. The aorta The aorta leaves the heart in the cranial direction (this section is called the ascending aorta). On the right-ventricular side. The trabecula septomarginalis is a powerful ridge that abuts the right-ventricular surface of the septum. AO. The outlet component of the left ventricle contains the aortic valve. and an outlet component. this line is connected with the dorsal end of the smooth-walled septum. the left . On the left-ventricular side.Cardiac anatomy and examination of specimens 21 Figure 3. separating the outlet paths of the right and left ventricles. partially of fibrous tissue. which has an aortic (anterior or septal) leaflet and a posterior (or mural) leaflet. curves dorsally to the left (this section is called the aortic arch). aorta. called the muscular septum. left coronary. and outlet component of the muscular septum come together. The trabecular component of the left ventricle extends from the origin of the papillary muscles to the apex of the heart. The muscular septum is divided into the inlet. apical trabecular component. The inlet component contains and surrounds the mitral valve.5).

muscular septum. the atrioventricular junction. Below the aortic arch it divides into right and left pulmonary arteries.6). muS. The pulmonary trunk The pulmonary trunk emerges from the conus arteriosus of the right ventricle. and the left subclavian artery. membranous septum.7 This is followed by a description of the position of the heart in the thorax. The ascending aorta passes over into the aortic arch. The descending aorta designates that part of the vessel which extends from the ostium of the arterial duct to the aortic bifurcation. and the right coronary artery in the sinus aortae below the right coronary leaflet (Figure 3. the duct closes and becomes the arterial ligament. The shorter left pulmonary artery reaches the left bronchus over a direct path in front of the ascending aorta. S. terminating there at the pulmonary valve. a relatively simple and reliable diagnosis of complex congenital heart malformations can be achieved. the ventricular mass. Congenital heart anomalies can affect one or more of these segments or the great veins. and any associated malformations. in this order.5 Specimen sectioned along the cardiac long axis. the orientation of the apex of the heart. Each of these three sections of the normal heart has a right and a left side. The arterial duct is structurally different from the aorta and the pulmonary artery. To determine which of the cardiac chambers are normal. The normal heart consists of three segments: the atria. The arterial duct demarcates the isthmus of the aorta between the site of emergence of the left subclavian artery and the aortic insertion of the duct. The wall of the arterial duct is thicker Atrial arrangement The most constant component of the atrium is the atrial appendage. the left common carotid artery. The right and left atrial appendages are . and the great arteries. the two atrioventricular valves. and has irregular ridges running lengthwise along it.22 Fetal Cardiology Figure 3.5. which connects the pulmonary trunk with the descending aorta. than that of both the aorta and the pulmonary trunk.6 Longitudinal section through the aortic valve (AV) showing the ostium of the left coronary artery (LCA) below the left coronary leaflet and the ostium of the right coronary artery (RCA) below the right coronary leaflet. showing the four cardiac chambers. After birth. and the ventriculoarterial junction. coronary artery in the sinus aortae below the left coronary leaflet.6 This requires a description of the atrial arrangement. and the interventricular septum. In the fetal circulation the arterial duct emerges from the bifurcation of the pulmonary trunk. Examination of specimens By applying the method of sequential segmental analysis of the heart. The luminal surface of the duct in the newborn is less smooth than that of the great arteries. The great branches of the aorta emerge from the convex side of the aortic arch: the brachiocephalic artery (which divides into the right subclavian artery and the right common carotid artery). The longer right pulmonary artery passes under the aortic arch and behind the inferior vena cava to the right bronchus. meS. septum. the structure of their constant components has to be evaluated morphologically. Figure 3.

is rare. The mirrored position. and no pectinate muscles. In the usual arrangement. the so-called situs solitus. a long hyparterial bronchus on the left means situs solitus. left. and marked by an extensive crest with pectinate muscles extending all round the atrioventricular junction. There are four different possibilities here (Figure 3. in the morphologically left appendage this junction is narrow.8). There are two groups of atrioventricular junction: biventricular and univentricular. Bronchi of equal length are an indication of atrial isomerism. There are four different possibilities here (Figure 3. The first step in the sequential analysis is the determination of the arrangement of the atrial chambers. the situs inversus. The most reliable feature differentiating between the right and left appendages. the morphologically left bronchus is crossed by the left pulmonary artery before the bronchus divides. The atrial arrangement almost always corresponds to the bronchial morphology.or right-atrial isomerism occurs in cases of visceral heterotaxy.8 The shape of the morphologically right appendage is that of a broad triangle.7 Diagram showing the four possible arrangements of the atria (reproduced with permission from reference 9). In addition. The identification of the bronchi is based on the fact that the morphologically left bronchus is almost twice as long as the morphologically right bronchus. whereas a long hyparterial bronchus on the right is a sign of situs inversus. which is Figure 3. different. An atrioventricular connection is called discordant when the morphologically right atrium . In biventricular connections each atrium is connected to one ventricle. The relationship of the bronchi to the pulmonary arteries is the basis for distinguishing between right and left isomerism. not the case for the morphologically right bronchus. This is especially important in cases in which the arrangement of the atrial chambers is not clear. If the morphologically right atrium is connected to the morphologically right ventricle and the morphologically left atrium is connected to the morphologically left ventricle. both atrioventricular connections are called concordant. In some cases the lateralization of the atria is lacking. In the morphologically right appendage this junction is wide. Since the judgment as to whether isomerism exists is normally made on the basis of the atrial appendages. and the interrelationship of the ventricles. and has no crest.Cardiac anatomy and examination of specimens 23 Figure 3. is the nature of the junction between the appendage and the smooth-walled venous component of the atrium.7). Generally. the morphology of the atrioventricular valves and of the ventricles.8 Diagram showing the four variants of bronchial morphology usually correlating with atrial arrangement (reproduced with permission from reference 9). the morphologically right atrium is located to the right of the morphologically left atrium. Therefore. Variation of the atrioventricular junction This section describes the junction (also called ‘connection’) of the atria with the ventricles. so that there are two morphologically right appendages (right isomerism) or two morphologically left appendages (left isomerism). however. the expressions ‘isomerism of the right atrial appendage’ and ‘isomerism of the left atrial appendage’ are also used. whereas the left appendage is tubular and hooked.

Here one has to distinguish between a right-hand and a left-hand topology (Figure 3. rudimentary left ventricles are normally in the posteroinferior position within the ventricular mass. regurgitant. If less than 50% of one valve overrides the septum the connection is described as biventricular. or to a solitary ventricle of indeterminate morphology. Atrioventricular concordance and discordance can each occur together with either situs solitus or situs inversus of the heart. right. There are no defined intermediate categories.10 Figure 3. to a dominant right ventricle (the left ventricle is rudimentary or incomplete). is connected to the morphologically left ventricle or the morphologically left atrium is connected to the morphologically right ventricle (Figure 3. The atrioventricular junctions are concordant. but they are ambiguous (reproduced with permission from reference 9).10 Frequently the valves can be stenotic. posterior. The connection is called univentricular only when more than 50% of both valves open into one ventricle. the biventricular connections cannot be classified in this way and are called ambiguous. incompetent. Diagram showing the isomeric arrangement of the atrial appendages.9 Diagram showing concordant and discordant atrioventricular connections in the usual and mirror image arrangements (reproduced with permission from reference 9). If the atrial appendages are isomeric.24 Fetal Cardiology Figure 3. inferior. In cases of an ambiguous atrioventricular connection the ventricular topology should be described. straddling. one of them can be stenotic. Variation of the ventriculoarterial junction The connection of the arterial trunks to the ventricular mass is here described. In such cases there are either two atrioventricular valves or a single common valve. and left. Rudimentary right ventricles are normally located anterosuperior. both atria open into a single ventricle) or the absence of an atrioventricular junction on the right or the left side.10). The second group of atrioventricular junctions is called ‘univentricular. In describing the spatial relationship of the ventricles to each other the following terms are used for the position of the right ventricle with regard to the left ventricle: anterior. or sit astride the septum. The relationship between the ventricles does not depend on the atrioventricular connection. imperforate or straddling.e. The possible relationships of the . If there are two atrioventricular valves. The atria can be connected to a dominant left ventricle (the right ventricle is rudimentary or incomplete). If one atrioventricular connection is missing. or imperforate. or overriding.9).’ This connection is characterized by a double ventricle inlet (i. If there is one atrioventricular valve sitting astride the septum. superior. there is only a solitary valve that can be incompetent. the heart is intermediate between having a completely biventricular and a completely univentricular connection.

The term ‘ventriculoarterial discordance’ is used to describe the reverse connection. One or both of the arterial valves may override the ventricular septum. The structure separating the two arterial valves and the two ventricle outlet paths is the infundibular septum. to the right. . or one of the arterial valves can be imperforate. i. each great arterial trunk originates in a separate ventricle. In a concordant or discordant ventriculoarterial connection. In normal hearts the aorta gives rise to the systemic and the coronary branches and the pulmonary trunk divides into the right and left pulmonary arteries. or to the middle. or to a pulmonary trunk with an atretic aorta lacking a connection to the ventricle (Figure 3. When the aorta and pulmonary trunk arise from the appropriate ventricles. Figure 3. it is assigned to the ventricle associated with its greater part. or in the midline. simply. muscular infundibulum lies below the arterial valves or there is a fibrous continuity on both sides between the arterial valves and the atrioventricular valve. discordant. If the ventricular mass is connected to only one arterial trunk. The ventriculoarterial connection is concordant when the morphologically right ventricle is connected to the pulmonary trunk and the morphologically left ventricle is connected to the aorta.11). are described. Position of the heart On describing the position of the heart in the chest it is necessary to give the basic position of the heart and the orientation of the cardiac apex. The cardiac apex can point to the left. If there is a common arterial valve. as well as the morphology of the infundibular structures. double outlet. these are concordant. simply.Cardiac anatomy and examination of specimens 25 Figure 3. They are identified by their branching pattern. it is connected to a single common arterial trunk or to an aorta with an atretic pulmonary trunk lacking a connection to the ventricle. ‘double outlet’). If one of the great arterial valves overrides the ventricular septum. primarily within the right chest (dextrocardia). the arterial valves can be perforate.12 Diagram showing the four variants of a single arterial trunk from the heart (reproduced with permission from reference 9). or indeterminate morphology. and single outlet (Figure 3. Terms used for describing the ventriculoarterial junctions of the heart are: concordant.9 The ventricle may be of right.11 Diagram showing concordant and discordant ventriculoarterial junctions. The most frequent arrangement with regard to the infundibular morphology is that the right ventricle has a complete.12). muscular infundibulum which is lower than its arterial valve. With regard to the arrangement of the ventriculoarterial junction. ‘single outlet’). Origins from inappropriate ventricles produce discordant junctions (reproduced with permission from reference 9). Another possibility is that on both sides a complete. The arterial trunks are the third segment of the heart. either it overrides the ventricular septum or it is committed to one ventricle.e. the arterioventricular connection is called a single-outlet ventriculoarterial connection (or. The heart can lie primarily within the left chest (levocardia). left. arterial valves to one another and the relationships of the arterial trunks. When more than half of both arterial valves are connected to the same ventricle it is called a double-outlet ventriculoarterial connection (or. A fibrous continuity exists between the arterial valve issuing from the left ventricle and the left atrioventricular valve.

Anatomie des Herzens. Anderson RH. Stuttgart: Georg Thieme Verlag. Anderson RH. Pathologie des Herzens. 2. Freedom RM et al. World J Surg 1985. Sequential segmental analysis of congenital heart disease. Terminology of congenital heart disease: glossary and commentary. Prenatal diagnosis of ventriculo-coronary communication in a second trimester fetus using transvaginal and transabdominal color-Doppler-sonography. Ho SY. The determination of atrial arrangement by examination of appendage morphology in 1842 heart specimens. 60: 227–31. Pacifico AD. Surgical treatment of double outlet left ventricle with an intact ventricular septum. 10. Anderson RH. Clinical and autopsy diagnosis and developmental implications. The Heart. 1992. 5: 281–8. Large RV plus small LV is not single RV. Sharma S. Van Praagh R. 9: 194–7. Br Heart J 1988. Kirklin JM. Circulation 1980. Devine W. 4. Van Praagh R. Becker AE. 5. Bollmann R. A morphologically normal heart can be abnormally located. Soto B. Anderson RH. Becker AE. Anderson RH. The basic position of the heart should be described separately from the cardiac morphology. In very rare cases the heart can have an extrathoracic position (ectopia cordis). Wright GB. Anderson RH.26 Fetal Cardiology The orientation of the cardiac apex is independent of the position of the heart. Pediatr Cardiol 1984. 7. 48 (Suppl III): 19–23. eds. 8. 56: 139–43. Ultrasound Obstet Gynecol 1997. 1985. 3. . Becker AE. References 1. Van Praagh S. Stuttgart: Georg Thieme Verlag. 1982. Ein Farbatlas. 61: 1057–8. Tennstedt C. 9: 550–67. Bargeron LM. Zuberbuhler JR. London: Gower Medical. 6. 9. Göldner B. An unusual cardiac position does not represent an abnormality of the cardiac morphology. Conduction tissue in congenital heart surgery. Chaoui R. Ein Farbatlas. Circulation 1973. Circulation 1977. David I. Becker AE.

and has far-reaching effects on healthy cardiovascular function and development. 36. Two days later. The terminal villi are highly capillarized. Therefore. The size and function of the placenta have a profound effect on normal fetal growth and development. The fetal heart pumps blood not only to the fetal brain and other organs. Villous development Twelve to 18 days post-conception (pc). 47. Any aberration affecting the placenta has serious consequences for the developing fetus. proliferation.1–9 The mesenchymal villi can differentiate into several types of specialized villi. as the organ responsible for the supply of oxygen to the fetus. coordinate various biological processes: angiogenesis. Some mesenchymal villi and immature intermediate villi are retained in the centers of the villous trees. The first generation of tertiary villi make up the mesenchymal villi. The placenta functions as the fetal lung in gas exchange. which are the first structures to provide surface area for maternal–fetal exchange of nutrients. but also – in similar quantity to that within the fetal venous system – to the placenta. especially the extravillous trophoblasts (EVTs). and their effects on fetal cardiac development and function (Chapters 10. i. into the maternal blood surrounding them. This makes the understanding of normal and pathological placental implantation and function indispensable in a textbook dealing with fetal cardiology. Rather. the first fetal capillaries appear. placental pathology is most relevant to fetal well-being. the tertiary villi begin to develop. From days 18 to 20.18 Fibroblast growth factor (FGF) signals. 11. rather. and 49). and are efficient vessels for maternal–fetal diffusional exchange.1. This invasiveness may be a general characteristic of branching morphogenesis or its driving force. We found that FGFR 1–4 are expressed in placenta but not in the decidua. and branching morphogenesis. At their appearance. forming a sort of growth reserve (Figure 4. while here we focus on the cellular and molecular processes involved in placental development and implantation. Other chapters in this book deal with the physiology of placentation and placental function. Mature intermediate villi begin to differentiate from the mesenchymal villi from about pregnancy week 23. Between days 20 and 42 pc.e. We examined human placental branching morphogenesis through the expression of FGF receptors (FGFR) 1–4 and FGF10 in placentas and decidua. new capillaries are formed from mesenchymal precursor cells. among them intrauterine growth restriction (IUGR) and preeclampsia (PE).1). as well as the fetal kidney in waste removal. differentiation. the primary villi. In this chapter we in no way attempt to provide an exhaustive review of placental form and function: whole textbooks have been devoted to this topic. gases. the first generations of mesenchymal villi begin vasculogenesis. these primary villi are invaded by embryonic connective tissue and thus transformed into secondary villi. Increased diameter and numerous stromal channels characterize the immature intermediate villi. These eventually become stem villi through stromal fibrosis. it must always be remembered.4 Placental implantation and development Simcha Yagel and Debra S Goldman-Wohl Introduction The placenta. is fetal in origin. we focus our discussion on selected aspects of placental development having particular influence on the fetal cardiovascular system.19 We also showed that Spry2 is expressed through the three trimesters of pregnancy by the placental villous macrophages (Hofbauer cells)20 and that FGF10 induces . through their receptor tyrosine kinases (RTKs). as branches enter surrounding tissues. while FGF10 is expressed by decidual cells and by the placenta. 44. These differ from immature intermediate villi in that they do not mature into stem villi.10–17 The process of branching morphogenesis begins with selection of a cell subgroup induced towards motility and invasivity. and wastes. trophoblastic trabeculae begin to proliferate and form finger-like protrusions. The mesenchymal villi form the only pool for subsequent villous sprouting and development of the villous trees. they produce many terminal villi along their surfaces.

Failure of the trophoblasts to invade the uterus appropriately. is often observed in placentas of preeclampsia. the syncytiotrophoblast and the invasive trophoblast.20 Invasion and implantation Trophoblasts are the first cells to differentiate in the embryo. The syncytiotrophoblast of the chorionic villi is responsible for placental nutrient and gas exchange. not penetrating to one-third of the myometrial layer proximal to the decidua. fewer invasive trophoblasts. suggesting that mesenchymal–epithelial interaction and cross-talk among the different placenta cell types may play a role in the regulation of placental development. This aids the process of forming a vessel of low resistance and high capacitance. We speculate that trophoblast outgrowth and invasion at the fetal–maternal interface are in part positively regulated by FGF10 and by Sprouty 2 negatively.17. FGF10 is in turn a strong inducer of cell migration and collagenolytic activity. while the latter migrate to the maternal uterine spiral arteries. Silencing of Spry2 expression by siRNA also enhanced the outgrowth of trophoblasts. The former migrate through and invade the uterine tissue and anchor the placenta to the uterus. Early in pregnancy. . Stem villus Spry2 expression in Hofbauer cell culture. It is these cells that adhere to the uterus and initiate the implantation process. the cytotrophoblasts begin to proliferate and form cell columns.19.28 Fetal Cardiology 1st and 2nd Trimester 3rd Trimester Trophoblastic sprout Villous sprout Mesenchymal villus Hot spot Immature intermediate villus Mature intermediate villus with terminal villi Figure 4. This trophoblast-mediated conversion of the spiral arteries must occur by the end of the first trimester for a healthy pregnancy to continue. and upregulates metalloprotease activity in trophoblast cell culture. Hot spots correspondingly mark the sites of future villous branching (modified from reference 1). as well as increasing the migration of single cell trophoblasts through MatrigelTM. Spiral artery mean diameter remains low: the persistence of unmodified narrow spiral arteries results in reduced placental perfusion. also continues to the end of pregnancy. by displacing and replacing the endothelial cell lining of the vessels.22–25 Conversion of the maternal spiral arteries is mediated by trophoblast invasion. lineage derived from the cytotrophoblasts may be defined as the endovascular or interstitial lineage. Trophoblasts retain a stem cell population of villous cytotrophoblasts throughout pregnancy. and differentiation of trophoblasts into two major cell lineages. equipped to meet the ever-increasing demand for blood flow required to maintain the growing pregnancy.1 Schematic representation of the formation and differentiation of placental villi during early and late pregnancy. The second.21 Spry2 was shown to be a potential regulator of FGF10 activity. invasive. It is in the spiral arteries that the endovascular invasive trophoblasts begin the work of conversion of the spiral arteries. or failure to convert the spiral arteries. whether through shallow trophoblast invasion. It is from these cell columns that extravillous trophoblasts emanate and from which the interstitial and endovascular invasive trophoblasts are derived. It is within the chorionic villi that the fetal placental arteries and veins develop. Histological characteristics of the various villus types and their typical topographical relationships.26 Interstitial endovascular invasion may be superficial in preeclampsia. Note the immature intermidiate villus (left) showing a ‘hot spot’ which subsequently (top of the left villus) develops via trophblastic and villous sprouts into a new mesenchymal villus. as well as most placental hormone and growth factor production. This is coincident with loss of the muscle and elastic tissue surrounding the arteries and replacement of the endothelial cells with trophoblasts.20 Exogenous FGF10 promotes invasion and outgrowth of villi in organ culture.

27 Other molecular mechanisms necessary for trophoblast invasion include transcription factor expression and differentiation to the invasive phenotype. αVβ6. Defects in the molecules responsible for trophoblast invasion and degradation of the extracellular matrix. digestion of the extracellular matrix (ECM). trophoblast immune cell cross-talk and induction of chemokines and cytokines that govern trophoblast migration and angiogenesis.35 was shown to be downregulated as trophoblasts enter the more invasive oxygen-enriched environment.Placental implantation and development 29 Some light has recently been shed on the mechanisms of cytotrophoblast induction of vessel remodeling in a transplantation model of placental villi into severe combined immunodeficiency disease (SCID) mice. ultimately. Certain adhesion molecules characterizing the stem cell population of villous cytotrophoblasts inhibit invasion: these molecules are downregulated as trophoblasts enter the invasive pathway of differentiation (α6β4. getatinase B) is closely associated with the invasive phenotype of trophoblasts. The enzymatic activities of urokinase plasminogen activator and plasminogen inhibitor are also altered in preeclampsia.33. Investigators showed that trophoblasts mediated both maternal endothelial cell. which is critical for HIF1-α and HIF1-β regulation. including those involved in placental vascularization processes. apoptosis. gene expression and the activity of several molecules involved in degradation of the extracellular matrix are abnormal. Conversion of the spiral arteries depends on these modeling steps.39.2). On the other hand. and platelet endothelial cell adhesion molecule 1 (PECAM-1)). thus enhancing the invasive phenotype. vascular endothelial (VE)-cadherin. VEGF and FGF are involved in most of the heparinbinding angiogenic activity produced by both ovarian59 and placental60–64 tissues.31 In their pioneering work.54 Major angiogenesis factors have been identified. and the angiopoietin (ANG) protein families.41.28. VEGF s specific stimulation of . Furthermore. inadequate spiral arteries result. expression of the tumor suppressor protein von Hippel–Lindau (pVHL). through which trophoblasts must migrate. trophoblasts must induce the repertoire of genes involved in digestion of the extracellular matrix. Extravillous trophoblasts express the stage-specific repertoire of adhesion molecules (aVβ4.36 Invasive trophoblasts undergo an epithelial to endothelial cell transformation and express endothelial cell-specific adhesion molecules. and their respective receptors. and is critical to normal tissue growth and development. α1β1. Genbacev and colleagues32 showed that the 5–8% O2 conditions of the late first trimester fetomaternal interface favor the invasive trophoblast phenotype.50 One of the earliest stages in embryonic development is the establishment of functional circulation. is supported by the extensive increase in uterine and umbilical blood flow.25.38–40 Placentation includes angiogenesis in both fetal and maternal tissues to ensure sufficiently increased uterine and umbilical blood flow41–47 to afford the most favorable surroundings for meeting the demands of the developing fetus. which makes possible the rapid growth of the fetus in the second half of pregnancy. suggesting a role for these molecules in invasion and migration of trophoblasts.42.55–58 Indeed. It has also been shown that under hypoxic conditions trophoblasts retain adhesion molecule properties mimicking those of trophoblasts of preeclampsia.33 It has also been demonstrated34 that hypoxia-inducible factor (HIF1-α) and its downregulating response to higher oxygen tension can adjust trophoblast invasion through inhibition of transforming growth factor β3 (TGFβ3). To achieve successful invasion. Later in pregnancy. For example. which fails to be upregulated in preeclampsia. vascular cell adhesion molecule 1 (VCAM-1). trophoblast proliferation is preferentially observed.29 When trophoblast invasion is shallow (for example in preeclampsia). mechanisms of migration and invasion. Among them are vascular endothelial growth factor (VEGF).41. and vascular smooth muscle cell. expression of angiogenic factors involved in spiral artery remodeling and.48 indeed. MMP (92-kDa matrix metalloproteinase. and failure to acquire endovascular integrin markers are indicative of a problem in a fetal pathway of trophoblast migration (Figure 4. Placental vasculogenesis and angiogenesis Angiogenesis is the formation of new vascular beds. angiogenesis factors influence physiological exchange.51–53 Increased transplacental exchange. Expression of adhesion molecules by trophoblasts Adhesion molecules and integrins play an important role in cell migration. and E-cadherin). leading to deficient placentation and its sequelae. If the endovascular invasive trophoblasts do not undertake the function of the endothelial cells which normally line the spiral artery. under 2% O2 conditions.30 including expression of MMP.49. decreased vascular development and increased vascular resistance have been shown to be associated with early embryo loss.25. fibroblast growth factor (FGF).37 Extracellular matrix degradation Trophoblast invasion and digestion of the extracellular matrix is governed by intrinsic trophoblast cell programming as well as interaction with the maternal cellular environment.

71 Defects included abnormal vascular formation. vascular permeability and vascular endothelial cell protease production and migration all influence the angiogenic process. major vessels.65–67 and has been implicated in the primary regulation of angiogenesis in normal and pathological processes such as luteal growth.2 Blastocyst Implantation. including angiogenesis.and inhibitors molecule expression Endometrial glands Regulation of cytotrophoblast invasion and vascular mimicry Endometrial capillaries Figure 4. In late pregnancy. organization. During early pregnancy.69 Gene knockout studies show a central role for VEGF in fetal and placental angiogenesis.73 . coronary ischemia. more than in maternal (endometrial) placental tissues. The diagram shows an invading blastocyst (about 9 to 10 days after conception) and the processes necessary for trophoblast invasion (reproduced with permission from reference 37). wound healing. aorta. which led to fetal demise before mid-gestation. VEGF mRNA is increasingly expressed in the placentome and intercotyledonary fetal membranes.55.73 Heterozygous knockout embryos with decreased VEGF expression showed similar defects and also died by day 11–12.70. homozygous knockouts of the genes for VEGFRs led to defects in the initial formation of placental vasculature and angiogenesis. while basic FGF (bFGF) mRNA is expressed more in endometrial than in fetal placental tissues. Homozygous knockouts of VEGF were lethal by day 11 and led to dramatic cardiovascular defects: delayed and abnormal development of the heart. VEGF mRNA is expressed in fetal placental. and tumor growth.59 VEGF has also been shown to increase angiogenesis in both in vivo and in vitro models.68 Mouse models have been used to study various aspects of placentation.72. patterning.30 Fetal Cardiology Cytotrophoblast Syncytiotrophoblast Blastocyst cavity Bilaminar embryonic disk Amniotic cavity Decidualized stroma Oxygen tension Transcription factors Growth factors and cytokines Regulation of trophoblast proliferation and differentiation Regulation of placental growth Blood vessel Regulation of Proteinases trophoblast adhesion. and extraembryonic vasculature. including the yolk sac and the placenta.55. In mice.72. while bFGF is greatest in the intercotyledonary fetal membranes. and endothelial morphology.38–40.

Specifically. and ANG1 die later than VEGFR knockouts. Upregulation of endometrial expression of VEGF and bFGF mRNA. which has been shown to mediate an estrogeninduced increase in uterine blood flow. and partners VEGF in the angiogenic process. because the receptor Tie2 is present primarily on endothelial cells.90 ANG2 is a natural Tie2 antagonist.85.87–89 Like VEGF. ANG1 and ANG2 appear to be vascular-specific growth factors.89. ANG1 is a Tie2 agonist. as well as increased uterine vascularization and blood flow. In (b). critical in embryonic vascular development.39.91 Ang2 is expressed in the syncytiotrophoblast and Tie2 in both fetal and maternal endothelial cells (Figure 4. The endothelial cell-specific receptors Tiel and Tie2 and the angiopoietin ligands of Tie2 play fundamental roles in angiogenesis and in remodeling of vessel structure.92–95 In addition. Ang2. ANG1 is produced by periendothelial cells. The four factors that are discussed in detail in this review are highlighted in red. The ephrin subfamily of endothelial cell ligands and receptors plays a role in endothelial cell and nerve cell targeting and migration but not proliferation.91 ANG does not affect endothelial cell proliferation. and demise by mid-gestation. In (c) only those members of the large ephrin ligand family (and only their counterpart Eph receptors) that have been implicated in vascular growth are shown (reproduced with permission from reference 96). .80–84 NO also regulates the expression of VEGF and bFGF. Tie2.96 The downregulation of EPHB4 and upregulation of ephrin-B1 in Ephrin-B1 PIGF VEGF-A VEGF-B VEGF-C VEGF-D Ang 1 Ang 2 Ang 3 Ang 4 Ephrin-B2 Ephrin-A1 + – + – + ? ? (a) VEGFR-1 VEGFR-2 VEGFR-3 (KDR/Flk-1) (Flt-4) (Flt-1) Tie1 (b) Tie2 (c) EphB2 EphB3 EphB4 EphA2 Figure 4.87 Null mutations of Tie1.74.3 Schematic representation of three families of vascular growth factors and their receptor interactions.76 VEGF and FGF regulate placental blood flow. is involved in the widening of vessels and disruption of vessel integrity. VEGFR knockouts lack primary vascular growth and organization.91. ‘+’ or ‘–’ indicates whether the particular angiopoietin activates or blocks the Tie2 receptor.3). (a) Vascular endothelial growth factors (VEGFs). an antagonist ligand for Tie2. while ANG and Tie knockouts are affected at the stage of vascular remodeling.38. (b) angiopoietins.56. (c) ephrins. but increases microvascular organization and endothelial cell survival. and influence angiogenesis and other developmental and differentiated functions.Placental implantation and development 31 which may point to a threshold level of VEGF expression for normal vascular development to occur. Lack of ANG1 leads to significant cardiovascular defects.74.96 This receptor–ligand interaction and fetal–maternal cross-talk has the potential to mediate widening of both fetal and maternal vessels.86 ANGs have also been shown to be major angiogenic factors that regulate (both increase and decrease) vascular growth and development.90 We have described the expression patterns of Tie2 and its antagonist ligand ANG2 at the fetal–maternal interface. were demonstrated following estrogen treatment in ovariectomized ewes and mice.47. leading to vascular regression and modulation of vascular growth. but not in cell proliferation. FGFs are also potent angiogenesis factors. a local vasodilator.77–79 Both VEGF and bFGF stimulate endothelial production of nitric oxide (NO).75 FGFs are unique among the angiogenic growth factors in that they are pleiotropic.56.56. whereas ‘?’ indicates that a potential interaction has not yet been confirmed experimentally. and have been shown to stimulate in vivo and in vitro uterine arterial and fetal placental arterial endothelial cell proliferation. pointing to a later role for Tie and ANG in vascular development.

In addition. Furthermore. The mechanisms involved in placental trophoblast. which. IL-8 and IP-10. maternal immune rejection of the developing fetus has been implicated in pathologies of pregnancy such as recurrent miscarriage and preeclampsia. establishing a placental insufficiency ripe for the development of preeclampsia. and maternal and fetal tissues are in intimate contact at the placenta and maternal decidua across the maternal–fetal interface. increased the risk for preeclampsia.106 Several mechanisms work to maintain the delicate balance between immune tolerance and activation. Both extravillous trophoblast (EVT) and decidual lymphocytes . Trophoblasts invade maternal tissue while eluding maternal immune surveillance. Conversely. in turn. the presence of the maternal KIR-B haplotype was protective for preeclampsia. avoidance of maternal immune surveillance remain inadequately understood. cytokines. This maternal NK cell and fetal trophoblast cross-talk would exemplify how immune interactions between the mother and fetus at the maternal–fetal interface could contribute to the development of preeclampsia by both failure of trophoblasts to invade and lack of the angiogenenic signals necessary to convert the spiral arteries leading to uteroplacental blood flow insufficiency. specifically human leukocyte antigen (HLA)-C2 genotypes in the fetus and specific genotypes of maternal NK KIR receptors (killer immunoglobulin-like inhibitory receptors).4).103–105 Since trophoblasts are the placental cell population most in contact with the maternal immune system. since in a normal pregnancy trophoblasts closely associate with NK cells at the implantation site. we demonstrated that the genetic interactions described by Hiby et al100 cause overinhibition of NK cells. These express their respective ligands. IP-10) as well as angiogenic factors vascular endothelial growth factor and placental growth factor (VEGF and PLGF)102 (Figure 4. but in most cases it does not. Protection of trophoblasts from attack by NK cells is of critical importance.97 We investigated whether the ligand. KIR-AA. We did not observe ephrin-Al in syncytiotrophoblast nor in cytotrophoblasts. This overinhibitory model necessitated an intellectual shift among researchers. and thereby fetal. It yet remains to be demonstrated that the maternal– fetal immune genetic interactions that predispose a pregnancy to preeclampsia actually result in narrow. The molecular signals partly responsible for trophoblast attraction are being revealed through chemokine receptor expression on invading trophoblasts and chemokine ligand expression by the NK cells populating the maternal–fetal interface. This NK cell regulation of spiral artery conversion can extend to the effect that they have on trophoblast attraction and migration throughout the decidua. Indeed. research focuses on maternal immune recognition of these cells. unconverted spiral arteries. that genotypes that encode greater inhibition of NK cells are associated with the development of preeclampsia.99 Groundbreaking population-based genetic studies100 showed evidence that certain combinations of fetal and maternal genotypes. which could lead to rejection of the embryo by the decidual lymphocytes. The role of natural killer cells: when killers become builders Natural killer cell (NK cell) cytotoxicity is regulated by both inhibitory and activating receptors. This intimate contact could be expected to generate a maternal immune rejection response. ephrin-Al.101 Research into animal models over the past decade laid the foundation for a theory of uterine NK cells’ pivotal role in angiogenesis and artery remodeling at the fetal–maternal interface.32 Fetal Cardiology cytotrophoblasts is associated with endovascular migration.26 Maternal immune tolerance to the fetus Mother and fetus are distinct genetically. is involved in trophoblast targeting as well as angiogenesis in the placenta. Using RNA in situ hybridization analysis. Specifically. increases the risk for preeclampsia. These observations100 can be reconciled through the pioneering work in the mouse model. EphrinAl’s role in the defective cell invasion observed in preeclampsia remains to be investigated. The cell-specific distribution of ephrin-Al suggests that it may play a role in migration of trophoblasts and in the vascular remodeling induced by the invading extravillous trophoblasts.101 The role of decidual NK cell interactions with extravillous trophoblasts. growth factor secretion. we found ephrin-Al expressed in cell columns of extravillous trophoblast in first trimester placenta. where they represent the major lymphocyte population.98 In late first trimester and in second and third trimesters. activated decidual NK cells express higher levels of IP-10 and IL-8. and angiogenic stimulators that have the potential to enhance spiral artery widening and trophoblast invasion102 demonstrate these observations at the human fetomaternal interface. and leads to decreased secretion of specific cytokines interleukin 8 and interferon γ inducible protein-10 (IL-8. ephrin-Al is found in the extravillous trophoblast cells that have invaded the decidua. Hanna et al102 demonstrated that trophoblasts expressing the chemokine receptors CXCR1 and CXCR3 are attracted to decidual NK cells. whereas previous theories of preeclampsia had focused on the need to inhibit the NK cell response as a prerequisite for allowing trophoblast invasion.

Values are mean ± SD for each sample. (a–c): Concentrations of IL-8. Spiral arteries are not remodeled and remain narrow. HLA-G is normally expressed only in anchoring extravillous trophoblasts.221 (c) Figure 4.221 –HLA-G 721.Placental implantation and development 33 VEGF PLGF IL-8 ** IP-10 Concentration (pg/ml) 1000 750 500 250 0 721. Student t test.221 –Cw6 MHC class I-recognizing receptors regulate growth factor production by dNK cells. When HLA-G expression (dark knobs on cell) is lacking.221 (b) Concentration (pg/ml) 1000 750 500 250 0 721.221 –Cw4 721. not in syncyciotrophoblast nor in cytotrophoblasts.05.221 –Cw4 721. HLA-G protects these invasive cells from lysis by natural killer (NK) cells and is the laissez-passer necessary for successful trophoblast endovascular and interstitial invasion. One representative data set is shown out of two experiments performed (reproduced with permission from reference 102). resulting in a poorly perfused placenta (reproduced with permission from reference 130).5 Model of HLA-G expression in the placenta. IP-10.221 721. The extravillous trophoblasts normally invade the uterus and the maternal spiral arteries. Fetal blood vessel Intervillous space Syncytiotrophoblast HLA-G negative trophoblast NK cell Trophoblast lysis Decidua Nitabuch’s layer HLA-G positive trophoblast Spiral artery Normal Preeclampsia Figure 4. The average concentrations of cytokines secreted by a representative clone out of four to eight dNK clones from each receptor subgroup are shown.221 cells mock-transfected or transfected with the indicated MHC class I molecule. trophoblasts invading the decidua are lysed by NK cells. as it is in preeclampsia. and PLGF in supernatants from dNK clones having the indicated phenotypes that were coincubated for 72 hours with irradiated 721.221 –Cw6 LIR– KIR2DL1– KIR2DS4+ dNK clone (a) Concentration (pg/ml) 1000 750 500 250 0 721. **P < 0. VEGF.4 ** LIR– KIR2DL1+ KIR2DS4– dNK clone 721.221 –Cw4 721.221 –HLA-G 721.221 –Cw6 ** LIR+ KIR2DL1– KIR2DS4– dNK clone 721. and remodel the arteries for increased blood flow.221 –HLA-G 721. .

etc. HLA-G expression. or risk factors for cardiovascular disease. etc. in turn. etc. the spiral artery may have undergone conversion but is narrowed because of thrombosis or atherosis of the vessel. This may be caused by defects in trophoblast invasion or angiogenesis. In the case of a defective maternal pathway in placental development. HLA-G expression. risk factors for cardiovascular disease Intervillous space Shallow trophoblast invasion Unmodified spiral artery Arterial thrombosis and atherosis Shallow trophoblast invasion unmodified spiral artery Arterial thrombosis and atherosis Placental insufficiency Endothelial cell dysfunction Maternal syndrome Figure 4. are involved in these mechanisms. acquired (aPLS). HLA-E108 and HLA-G. results in uteroplacental blood flow insufficiency and ultimately the maternal cascade of events of preeclampsia (reproduced with permission from reference 26). There are several possibilities indicative of a defect in fetal trophoblast development. In normal pregnancy.109 along with the classical HLA-C protein. This may be indicative of defects in trophoblast differentiation. angiogenesis. widening the artery and increasing blood flow to the intrauterine space.). the cytokine release of decidual lymphocytes is closely regulated. as most of the cytotoxic T lymphocytes (CTLs) are directed against HLA-A and -B proteins. angiogenesis or cross-talk between trophoblasts and decidual NK cells. thrombophilic disorders. etc. HIF1-α. adhesion molecule switching.). Therefore.34 Fetal Cardiology Normal Blood Cytotrophoblast Fetal blood vessel Syncytiotrophoblast Anchoring cell column Endovascular invasive trophoblast Interstitial invasive trophoblast Endothelial cell Spiral artery Decidua Intervillous space Fetal pathway Defects in: trophoblast differentiation. Both the fetal and maternal pathways lead to narrow spiral arteries which. molecule switching. Modulation of the local cytokine profile107 is thought to control EVT invasion. The muscle and elastic tissue surrounding the vessels is destroyed. EVT also expresses two non-classical class I major histocompatibility complex (MHC) proteins. invasion (ECM degradation. immune factors. genetic (factor V. The cytokine cross-talk occurring across this interface has been the subject of extensive research. MTHFR. invasion of trophoblasts is insufficient and the spiral artery never converted to a widened vessel. MTHFR. NK cell–trophoblast cross-talk Maternal pathway Thrombophilic disorders: genetic (factor V. In this case. HIF-α.6 Schematic representation of normal and defective fetal and maternal pathways in the development of preeclampsia. One of the chief activities of leukocytic and non-leukocytic cells at the maternal–fetal interface is to produce and release cytokines. The striations beside the spiral artery in the fetal pathway diagram depict a vessel that has maintained muscle and elastic tissue. migration.). but does not express HLA-A and HLA-B proteins. adhesion.110 This unique pattern of expression of class I MHC may prevent rejection of the semi-allogeneic fetus by the mother’s immune system. migration. .) or acquired (aPLS) immune factors. invasive trophoblasts enter both the decidua and maternal spiral arteries. invasion (ECM degradation. cytokines help to facilitate communication between the host and guest cells. They replace the endothelial cells of maternal uterine spiral arteries.

Th2 cells help antibody production by B cells. which are followed by vasculogenesis and angiogenesis of fetal and maternal blood vessels at the fetal–maternal interface. 6.5). The Human Placenta. They are known to be involved in immune response regulation. 8. and finally resulting in the maternal syndrome. Reduced HLA-G expression in preeclampsia and vulnerability to attack by the maternal NK cells supports the idea that the disruption of trophoblast invasion would lead to failure of spiral artery conversion. many key questions are still open for further investigation. Since large granulated lymphocyte cytotoxicity against trophoblasts is induced in vitro only following activation by IL-2. New York: Springer. Many of the mechanisms discussed above. Castellucci M. 1970. 136: 190–203. However.26 Conclusions In the last decade. The question of local suppression of Th1-type cytokines in human pregnancy remains open.6). Fetal vasculogenesis and angiogenesis in human placental villi. Castellucci M et al. that HLA-Gpositive trophoblasts will be unable to infiltrate the decidua. These findings may also help to elucidate a mechanism for maternal immune tolerance of the developing fetus. narrow spiral arteries and poor placental perfusion may be the consequence of a fetal pathway of shallow trophoblast invasion and unconverted spiral arteries or a maternal pathway of normally converted spiral arteries that become blocked. leading to a hypoxic uterine environment and endothelial dysfunction. 2. The role of cytokines at the maternal–fetal interface is not yet fully understood.130 The Fas receptor and its ligand FasL belong to the tumor necrosis factor and nerve growth factor receptor family. 4th edn. 6: 485–94.30. References 1. 2000. Cambridge: Heffner. 178: 30–44. King BF.128–130 It has been suggested that HLA-G.107 It is possible. Kaufmann P. and pregnancy without prior cohabitation increase the risk of preeclampsia. Interferon γ is expressed in the placenta and found in amniotic fluid at term.131 When invasive trophoblasts lacking HLA-G encounter decidual NK cells they are destroyed (Figure 4. eds. Scheffen I et al. and that these cells are able to induce apoptosis in activated lymphocytes. Rubanyi G. Demir R. insemination with donor sperm. with its initiating factor being narrow spiral arteries. Castellucci M. Enders AC. These two types of response may be mutually exclusive. Th1 cytokines. 181: 327–40. 181: 117–28. IL-4 and IL-6 have been found in maternal and fetal tissues at the fetomaternal interface. Scheper A. Formation and differentiation of extraembryonic mesoderm in the rhesus monkey. A three-dimensional study of the normal human placental villous core. The decidua is heavily infiltrated by NK cells. Kingdom JC. however. help macrophages and cytotoxic T lymphocyte-mediated responses. 3: 269–86. and in addition maternal factors such as risk for atherosclerosis. which will better elucidate the complex processes involved in placental implantation and development. protects invasive trophoblasts from attack by NK cells. if the maternal natural killer cell recognition of HLA-G is defective. Villous sprouting: fundamental mechanisms of human placental development. Acta Anal 1989. Kauma et al134 showed that FasL is expressed on trophoblasts at the maternal–fetal interface.133 This preeclampsia model in turn exemplifies the balancing act of escaping maternal immune surveillance. Castellucci M.114. Kaufmann P. Kosanke G. Verdenelli F et al. these pathways converge in the maternal disease of preeclampsia (Figure 4. Am J Anat 1988.Placental implantation and development 35 Activated T cells (Th0 cells) produce a variety of cytokines. Boyd JD. and cytotrophoblasts and maternal lymphocytes produce IL-10.134–136 Preeclampsia may be considered as a two-step disease. interleukin 2 (IL-2) and interferon γ. Ultrastructural differentiation of stromal and vascular components in early macaque placental villi. its absence may be a method of large granulated lymphocyte cytolytic behavior regulation.125–127 HLA-G expression is absent or reduced in preeclampsia as compared to normal placenta. activated CD4+ T cells (Th1 and Th2 cells) are grouped according to the cytokines they produce. which suggests an immune component for development of the disease.132. would lead to narrow spiral arteries. Am J Anat 1987.117 IL-2. Placenta 1982. .111–113 Cytokine secretion by fetal and maternal tissues at the interface seems to be high. 4. is absent from uteroplacental tissues. King BF. 5. II Stromal architecture. In: Risau W. 3. The development of the human placental villous tree.114–121 Of the Th2 cytokines. Receptors for a number of cytokines have been found on fetal trophoblasts. perhaps in conjunction with HLA-E. Kaufmann P. 9. and thereby initiate the cascade of events observed in preeclampsia.115 Cytokines or growth factors produced by the decidua may potentially affect trophoblast growth and invasion. Kaufmann P. 7.103–105. The Pathology of the Human Placenta. tremendous progress has been made in our knowledge of the cellular and molecular mechanisms precipitating these events. Benirschke K.116.122–124 There is epidemiological evidence to indicate that primipaternity. Hamilton WJ. and attention has focused on maternal NK cell response to HLA-G. Development of the vascular system in the placenta. Anat Embryol 1990. which result in insufficient placental blood flow. That is. Hum Reprod Update 2000.

Schanz A et al. Genbacev O. Angiogenesis in the female reproductive system. eds. particularly the development of that organ in sheep. . 175: 555–62. 1995: 1437–76. 102: 807–12. Natanson-Yaron S. Dushnik M et al. 18: 613–26. 26: 476–83. 16. Moodley J. Brosens I. Utero-placental vascular development and placental function. Khong TY. Todros T. 28. Brosens IA. Cytotrophoblast induction of arterial apoptosis and lymphangiogenesis in an in vivo model of human placentation. D’Amore PA. Redmer DA. 26. Piccoli E et al. Am J Obstet Gynecol 1968. Development and anatomy of the placenta. Dev Biol 2001. Morphological changes of the spiral arteries in the placental bed in relation to preeclampsia and fetal growth retardation. Kohnen G et al. Structural analysis of placental terminal villi from growth-restricted pregnancies with abnormal umbilical artery Doppler waveforms. 105: 577–87. 88: 876–81. J Clin Invest 2006. 17. Hurwitz A. in vivo. Am J Obstet Gynecol 1996. 53: 217–39. Histology. Graham CH. Human placental Hofbauer cells express sprouty proteins: a possible modulating mechanism of villous branching. Annu Rev Physiol 1991. Anteby EY. 39. Fisher SJ. Fisher SJ. 34. Anteby EY. Kingdom JC. Assheton R. 233: 526–36. Eur J Obstet Gynecol Reprod Biol 2005. 18. The role of the spiral arteries in the pathogenesis of preeclampsia. and notes upon the placenta of the elephant and hyrax. 93: 499–503. 92-kD type IV collagenase mediates invasion of human cytotrophoblasts. 24. Angiogenic factors. Krtolica A. 19. 37. De Wolf F. Morphogenesis of Endothelium. 171: 832–8. Caniggia I. Natanson-Yaron S. Damsky CH. Natanson-Yaron S. Yagel S. Kaufmann P. Robertson WB. Amsterdam: Harwood. Genbacev O. 1: 177–91. 233: 442–7. Edinburgh: Churchill Livingstone. Huisjes HJ. Umbilical Doppler waveforms and placental villous angiogenesis in pregnancies complicated by fetal growth restriction. Reynolds LP. Werb Z. Dixon HG. 277: 1669–72. 29. Am J Pathol 1997. Folkman J. Kaufmann P. Br J Obstet Gynaecol 1995. Gerretsen G. Oxygen and placental villous development: origins of fetal hypoxia. 15. 6: 886–92. Regulation of human placental development by oxygen tension. Robertson WB. 73: 1839–51. Kingdom JC. Am J Obstet Gynecol 1994. Hamani Y et al. One cause of defective endovascular invasion in this syndrome? J Clin Invest 1997. Shimonovitz S. Elaboration of stem villous vessels in growth restricted pregnancies with abnormal umbilical artery Doppler waveforms. Examination of distinct fetal and maternal molecular pathways suggests a mechanism for the development of preeclampsia. Salvatore CA. Zhou Y. Kingdom JC. Inadequate maternal vascular response to placentation in pregnancies complicated by preeclampsia and by small-for-gestational age infants. Distribution patterns of extracellular matrix components and adhesion receptors are intricately modulated during first trimester cytotrophoblast differentiation along the invasive pathway. 41. Fisher SJ. Damsky CH. Human cytotrophoblasts adopt a vascular phenotype as they differentiate. Preeclampsia is associated with failure of human cytotrophoblasts to mimic a vascular adhesion phenotype. 40. Br J Obstet Gynaecol 1981. Macara L. Placental bed morphology in black women with eclampsia. Winte RJ et al. Altered expression of gelatinase and surface-associated plasminogen activator activity by trophoblast cells isolated from placentas of preeclamptic patients. Science 1997. 30. Klagsbrun M. Implantation and the survival of early pregnancy. Dev Cell 2003. Fitzgerald ML. Placenta 1997. 20. Human cytotrophoblast differentiation/invasion is abnormal in pre-eclampsia. FGF 10 and Sprouty 2 modulate trophoblast invasion and branching morphogenesis. 13: 511–19. Obstet Gynecol Annu 1972. McCrae KR. 36. Mostachfi H. 102: 347–53. 200: 409–23. 13. 12. Itoh N et al. 32. 22. Kaufmann P et al. 4: 11–18. 119: 27–35. 99: 2139–51. Developmental regulation of the expression of 72 and 92 kd type IV collagenases in human trophoblasts: a possible mechanism for control of trophoblast invasion. 38. Librach CL. Fisher SJ. Zhou Y. I. Castellucci M. N Engl J Med 2001. Redmer DA. Obstet Gynecol 1999. Ludlow JW. 17: 37–48. Schweikhart G. 89: 210–22. Zhou Y. Schust DJ. Janatpour M et al. Janatpour M et al. Red-Horse K. 31.36 Fetal Cardiology 10. 25. Greenfield C et al. J Anim Sci 1995. Ramsaroop R. Greenfield C et al. Fisher SJ. Fibroblast growth factor-10 and fibroblast growth factor receptors 1–4: expression and peptide localization in human decidua and placenta. Goldman-Wohl DS. Elema JD et al. Placenta 1996. Anteby EY. 35. 151: 1809–18. Macara L. 42. Sen DK. Tube or not tube: remodeling epithelial tissues by branching morphogenesis. Killilea SD. 113: 437–49. 75: 376–8. Br J Obstet Gynaecol 1986. Lim K-H. 345: 1400–8. Klagsbrun M. Haines and Taylor’s Obstetrical and Gynaecological Pathology. Fitzgerald ML et al. 11. S Afr Med J 1989. J Cell Biol 1991. 14. Sciarrone A. J Clin Invest 1992. 93: 1049–59. Reynolds LP. 33. 23. Human cytotrophoblast expression of the von Hippel-Lindau protein is downregulated during uterine invasion in situ and upregulated by hypoxia in vitro. J Clin Invest 2000. 198: 143–200. Placenta 2005. The morphology of the ungulate placenta. Phil Trans R Soc Lond Ser B 1906. Zhou Y. Cell Tissue Res 1979. In: Fox H. Kaelin W. Norwitz ER. Classification of human placental villi. 9: 2152–64. Regulators of angiogenesis. Wells M. Affolter M. J Reprod Immunol 2007. Hypoxiainducible factor-1 mediates the biological effects of oxygen on human trophoblast differentiation through TGFβ(3). A comparative study. 76: 54–60. 2000: 255–75. Fisher SJ. 27. 116: 2643–52. 21. Mol Hum Reprod 2007. Kaufmann P. FASEB J 1992. Rivera J. Science 1987. A strategy for successful endovascular invasion? J Clin Invest 1997. Bellusci S. The placenta in acute toxemia.

63. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. 376: 66–70. 8: 961–5. and insulinlike growth factor-1 promote rat aortic angiogenesis in vitro. Fraser HM. 2: 289. Chen H et al. 64. 54. 55. 638: 1–8. Kirsch JD et al. Cachianes G. 140: 1399–407. Zheng J. Reynolds LP. Grazul-Bilska AT. Uterine vascularity during stimulation and its correlation with implantation in in-vitro fertilization. Plate KH. Nature 1995. 86: 689–96. 83: 497–502. Reynolds LP. 76: 89–96. Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. J Reprod Fertil 1987. Fong GH. section 2. eds. Breier G et al. Activation of the nitrogen-activated protein kinase cascade is necessary but not sufficient for basic fibroblast growth factor. Nature 1996. 133: 829–37. Matemal cardiovascular and other physiological responses to the endocrinology of pregnancy. . 49. Kuang WJ et al. Ford SP. Characterization of heparin-binding endothelial mitogens produced by the ovine endometrium during early pregnancy. Breier G. Reynolds LP. Kennaugh TM. Metabolic and circulatory studies of fetal lamb at midgestation. J Reprod Fertil 1988. 141: 995–1000. Jones BD et al. 79. 277: 48–50. Biological activities of fibroblast growth factors. Zheng J. Witschi E. In: Shepherd JT. 2nd edn. 45. Reynolds LP. 50.and epidermal growth factor-stimulated expression of endothelial nitric oxide synthase in ovine fetoplacental artery endothelial cells. Nature 1996. Endocrinology 1993. Vascular endothelial growth factor (rhVEGF165) stimulates direct angiogenesis in the rabbit cornea. Reynolds LP. 380: 435–9. Dickson SE. Gertsenstein M. Ferrara N. 57. Cullinan-Bove K. 59. Suppression of luteal angiogenesis in the primate after neutralization of vascular endothelial growth factor. 4: 284–92. Growth and microvascular development of the uterus during early pregnancy in ewes. Reynolds LP. Hanahan D. Battaglia FC et al. 73. Kirsch JD. Reynolds LP. Vascular endothelial growth factor. 62. Circ Res 1998. 48. Phillips GD. Carver-Moore K. Poltorak Z. Gospodarowicz D. Nicosia RF. 106: 30–8. 380: 439–42. Reynolds LP. 61. Kirsch JD. 145: 1023–9. J Anim Sci 1984. Cale JM. 59: 613–20. New York: McGraw-Hill. In Vivo 1995. Magness RR. 70. Endocrinology 1999. Damert A. Carmeliet P. 47: 698–708. Am J Physiol 1986. Itoh H et al. Reynolds LP. Neufeld G. 46.Placental implantation and development 37 43. Babaei S. Pregnancy induces an increase in the expression of glyceraldehyde-3-phosphate dehydrogenase in uterine artery endothelial cells. Teichert-Kuliszewska K. 68. 250: E538. Vascular endothelial growth factor (VEGF) and its receptors. Hum Reprod 1995. Nicosia SV. Risau W. Bethesda. Killilea SD. 65. Development of Vertebrates. Time-course of the uterine response to estradiol-17β in ovariectomized ewes: uterine growth and microvascular development. Smith M. Meegdes HLM. 66. Bird IM. Rossant J. Angiogenic activity of placental tissues of cows. 44. 10: 1497–501. Biochem Cell Biol 1998. Nature 1995. Foundations of Embryology. Timecourse of the uterine response to estradiol-17β in ovariectomized ewes: expression of angiogenic factors. Koos RD. ed. Millican DS. The biology of vascular endothelial growth factor. Uterine blood flow during early pregnancy in ewes: interaction between the conceptus and the ovary bearing the corpus luteum. Stone AM. J Anim Sci 1986. J Soc Gynecol Investig 1995. Angiogenesis in the corpus luteum. The Placenta. J Soc Gynecol Investig 1997. The Endocrinology of Pregnancy. Monge JC et al. Early pregnancy wastage: relationship between chorionic vascularization and embryonic development. Ann NY Acad Sci 1991. Redmer DA. Vascular endothelial growth factor/vascular permeability factor expression in the rat uterus: rapid stimulation by estrogen correlates with estrogen-induced increases in uterine capillary permeability and growth. Am J Pathol 1994. 13: 9–22. 47. Endocr Rev 1997. Redmer DA. Lunn SF et al. Role of the flt-1 receptor tyrosine kinase in regulating the assembly of vascular endothelium. Role of nitric oxide in the angiogenic response in vitro to basic fibroblast growth factor. Gengrinovitch S. 1998: 507–39. Peelers LLH. 1956. Redmer DA. J Reprod Fertil 1989. Patten BM. Fertil Steril 1988. Bassil S. NJ: Humana Press. Biol Reprod 2001. 75. Angiogenic activity of ovine placental tissues: immunoneutralization with FGF-2 and VEGF antisera. 72. 246: 1306–9. MD: American Physiologocal Society. Angiogenic activity of maternal and fetal placental tissues of ewes throughout gestation. Vascular endothelial growth factor is a secreted angiogenic mitogen. 49: 216–20. 56. Circulation to female reproductive organs. Endocrinology 2000. Meschia G. 74. Redmer DA. platelet-derived growth factor. Rossant J. 80. Abboud FM. Redmer DA. 67. Handbook of Physiology. 52. Am J Obstet Gynecol 1970. Yamaguchi TP et al. 78: 678–83. Leung DW. Bell AW. Redmer DA. 53. Ferrara N. Anderson SG. 59: 606–12. Science 1997. 18: 4–25. part 1. Reynolds LP. 51. Redmer DA. Failure of blood island formation and vasculogenesis in flk-1 deficient mice. Utero-ovarian interactions during early pregnancy: role of conceptus-induced vasodilation. Magness RR. 64: 1033–40. Biol Reprod 1992. Millaway DS. 77. 82: 1007–15. 76. Totowa. Shalaby F. Roth J et al. Reynolds LP. Ingenhoes R. Magritte JP. Ramsey EM. Melsaether AN. 78. Ferreira V. Kirsch JD et al. 1964. 81: 233–40. Magness RR. 1983: 241–69. 58. 58: 423–9. 62 (Suppl 2): 47–61. 69. Zheng J. Redrner DA. Magness RR. Greiss FC. Kraft KC et al. In: Bazer FW. Human and Animal. 60. Angiogenesis in embryos and ischemic diseases. Cohen T. Exalto N. Angiogenesis in the placenta. vol III. Thromb Haemost 1997. Endocrine 2000. Signaling vascular morphogenesis and maintenance. Philadelphia: WB Saunders. Science 1989. Breitman ML. Secretion of angiogenic activity by placental tissues of cows at several stages of gestation. FASEB J 1999. 1982. Kraft KC. Biol Reprod 1998. 376: 62–6. Uterine blood flow during early ovine pregnancy. Millaway DS. Biol Reprod 1998. Davis-Smyth T. New York: Praeger. 12: 1–9. 71.

Hattori T. Nature 2000. Placenta 2000. Biochem J 1999. HLA-E is expressed on trophoblast and interacts with CD94/NKG2 receptors on decidual NK cells. Hellstrom M. King A. Placenta 2004. Shaw CE. Am J Physiol 1998. 56: 1–21. So JN. III. J Reprod Immunol 2003. cell survival. 3: 153–9. VEGF-A induces expression of eNOS and iNOS in endothelial cells via VEGF receptor-2 (KDR). Lab Invest 1999. EPHB4 regulates chemokine-evoked trophoblast responses: a mechanism for incorporating the human placenta into the maternal circulation. and the life of the fetus. and NO production in human endothelial cells. O Shaughnessy KM et al. 99. Goldman-Wohl DS. 97. 104. Esadeg S. Allan DS. Gardner L et al. Development 2005. Decidual NK cells regulate key developmental processes at the human fetal-maternal interface. their interactions with decidual spiral arteries and homing of their precursors to the uterus. Ovarian and estrogen effects on NO synthase. Red-Horse K. interferongamma and granulocyte-macrophage colony-stimulating . 84. Wegmann TG. Nichols-Johnson VR. Faulk WP. Angiopoietin-2. 28: 201–6. Proc Natl Acad Sci USA 1997. 85. 98: 2158–66. FEBS Lett 1999. Suri C. Manier C et al. 87: 1171–80. Cell 1996. Combinations of maternal KIR and fetal HLA-C genes influence the risk of preeclampsia and reproductive success. Obstet Gynecol 1986. Immunology of human placental implantation: clinical implications of our current understanding. Immune recognition. Am J Physiol 1998. Frank S. 59: 175–91. Requisite role of angiopoietin-1. 98. Goldman-Wohl D. Direct actions of angiopoietin-1 on human endothelium: evidence for network stabilization. Loke YW. 95. Garcia-Cardena G. Lindahl P. Growth factors and cytokines at the maternal/fetal interface. 274: HI054–8. Lin H. 91. Granger HJ. Eph and ephrin expression in normal placental development and preeclampsia. Immunoregulation and Fetal Survival. 286: 2511–14. King A. Ziche M. Kwak HJ. Goldman-Wohl D. a ligand for the Tie2 receptor. Robertson SA. Immunologic testing and immunotherapy in recurrent spontaneous abortion. during embryonic angiogenesis. Jones PF et al. 58: 224–37. Science 1997. McLaughlin PJ. genetic regulation. Guilbert L. 67: 169–75. Bidirectional cytokine interactions in the maternal-fetal relationship: is successful pregnancy a TH2 phenomenon? Immunol Today 1993. Recognition of trophoblast HLA class I molecules by decidual cell receptors. Hanna J. Waltengberger J. VEGF upregulates ecNOS message. protein. Hanna J. 83. Rosenfeld CR. Chantakru S et al. Endothelial vasodilator production by uterine and systemic arteries. Eur J Immunol 2000. Bowen M et al. Leakage-resistant blood vessels in mice trans-genically overexpressing angiopoiectin-1. 407: 242–8. 79: 213–23. Yancopoulos GD. Angiopoietin-1 and its receptor Tie-2 participate in the regulation of capillarylike tubule formation and survival of endothelial cells. Biochem Soc Trans 2000. 111. Johnson PM. Fisher S. Huang WQ. 86. Lee SJ et al. Hiby SE. Mosmann TR. Marchal-Bras Goncalie R. The role of cytokines in gestation. 9: 407–11. Direct evidence to support the role of HLA-G in protecting the fetus from maternal uterine natural killer cytolysis. Thurston G. 30: 1623–31.38 Fetal Cardiology 81. Nitric oxide mediates interleukin-1-induced gene expression of matrix metalloproteinases and basic fibroblast growth factor in cultured rabbit articular chondrocytes. 87. 114. Nitric oxide contributes to estrogen-induced vasodilation of the ovine uterine circulation. McIntyre JA. Crit Rev Immunol 1994. 21: 376–87. 110. 102. Endothelial-perivascular cell signaling in vascular development: lessons from knockout mice. Patan S. Jones PF. Greenfield C. Science 1999. J Clin Invest 1996. 123: 431–9. Implantation and the placenta: key pieces of the development puzzle. Wegmann T. Surface expression of HLA-C antigen by human extravillous trophoblasts. Papapetropoulos A. Wegmann TG. Greenfield C et al. eds. Nat Med 2006. Betsholtz C. Hood JD. Mallah J et al. Stallmeyer B. Vince GS. 94. Differential regulation of vascular endothelial growth factor and its receptor fms-like-tyrosine kinase is mediated by nitric oxide in rat renal mesangial cells. and interaction with other angiogenic growth factors. Vagnoni KE. 132: 4097–106. Dengler TJ et al. Cross J. 113. Hampson I. 103. 105. Smith K et al. Update on pathways regulating the activation of uterine natural killer cells. Sasaki K. Mol Med Today 1998. Kampfer H et al. Kalen M. J Exp Med 2004. 266: 1508–18. 277: 55–60. King A. 101. Tie-2 and angiopoietin-2 expression at the fetal-maternal interface: a receptor ligand model for vascular remodelling. Hayes AJ. 112. J Biochem 1998. 107. 25: 623–30. Patan S et al. Fujisawa T et al. Johnson PM. 275: HI845–56. Gale NW et al. 93. Suri C. When killers become helpers. Mandelboim O. Microvasc Res 1998. 1987: 3–11. 12: 1065–74. Guilbert LJ. Nisbet-Brown E. Hiby SE. Angiopoietin-1 is an apoptosis survival factor for endothelial cells. King A. Davis S. 82. 106. Werb Z. a natural antagonist for Tie2 that disrupts in vivo angiogenesis. Maisonpierre PC. 200: 957–65. Kapidzic M. Croy BA. Kroll J. Vascular specific growth factors and blood vessel formation. New York: Oxford University Press. In: Gill T. Ariel I. Trends Immunol 2007. Hiby SE et al. Tie 1 and Tie 2 receptor tyrosine kinases inversely regulate embryonic angiogenesis by the mechanism of intussusceptive micro-vascular growth. 109. Burrows TD. Curr Opin Lipidol 1998. 338: 367–74. Taylor CG. 14: 353–6. Rouas-Friess N. 28: 191–5. 100. 21: s81–5. Placenta 2000. Microvasc Res 1999. Mol Hum Reprod 2000. 89. 90. 94: 11520–5. 108. 14: 239–92. Walker JJ. Haimov Kochman R et al. Cox BE. Meininger CJ. Phernetton TM et al. Zhou Y et al. Magness RR. 448: 249–53. Science 1994. Low-affinity receptors for tumour necrosis factor-alpha. Biochem Biophys Res Commun 1998. Seamark RF. Gill T. 92. Hamani Y et al. 96. Wegmann T. 6: 81–7. 88. 252: 743–6. Roy T. Suri C.

Roth I. Robillard PY. Olah KS. Zorzi W et al. 179: 1359–75. Masuhiro K et al. Am J Pathol 1993. 79: 485–90. Immnogestosis: a new etiologic concept of ‘essential’ EPH gestosis. J Reprod Immunol 1996. 127. 90: 87–94. 125. Eur J Immunol 1999. 1: 548–9. 129. 126. 22: 109–6.Placental implantation and development 39 115. Cytokine 1995. Association of pregnancy-induced hypertension with duration of sexual cohabitation before conception. Science 1994. Eisenman J. Vince GS. 32: 89–98. Mol Hum Reprod 2000. 6: 88–95. Altered expression of human leukocyte antigen G (HLA-G) on extravillous trophoblasts in preeclampsia: immunohistological demonstration with anti-HLA-G specific antibody ‘87G’ and anti-cytokeratin antibody ‘CAM5. Kauma SW. Johnson PM. Huff TF. 130. Hulsey TC.2’. 84: 2188–94. Corry DB. 120. Nishino E. 71: 436–41. 116. Am J Obstet Gynecol 1977. Coumans B. de Morales-Pinto MI. Etiology and pathogenesis of preeclampsia. 119. Perianin J et al. Nagata S. Screening for cytokine mRNA in human villous and extravillous trophoblasts using the reverse-transcriptase polymerase chain reaction (RT-PCR). Morrison L. Bulmer JN. J Clin Endocrinol Metab 1990. Ferry BL. Starkey PM. Am J Obstet Gynecol 1998. Human placental cytotrophoblasts produce the immunosuppressive cytokine interleukin 10. interferon-gamma. 29: 277–83. IL-15 has stimulatory activity for the induction of B cell proliferation and differentiation. Llano M. J Exp Med 1996. Tessier JL et al. 124. Flanagan BF et al. 128. 122. Fas and Fas ligand: a death factor. 135. Immunology 1993. Nilkaeo A. Cell Immunol 1991. 154: 483–90. Goldman-Wohl DS. interleukin-8. Sargent IL. Localization of IL-4 and IL-4 receptors in the human term placenta. Neilson JP et al. Am J Obstet Gynecol 1997. 118. Meager A. Armitage RJ. Trophoblastderived interleukin-6 (IL-6) regulates human chorionic gonadotropin release through IL-6 receptor on human trophoblasts. and its receptor. Macduff BM. Hermann U. with special consideration of the primigravid patient. current concepts. Dekker GA. 136. . Analysis of p60 and p80 tumor necrosis factor-alpha receptor messenger RNA and protein in human placentas. Shanebeck K et al. Preeclampsia in pregnancies by different fathers: immunological studies. J Clin Endocrinol Metab 1999. Am J Obstet Gynecol 1994. Br Med J 1975. 12: 731–7. Jokhi PP. King A. Hara N. Smith GN. Walker M. Yamashita T et al. Need JA. decidua and amniochorionic membranes. Hayes N. 7: 364–71. Immunology 1997. Increased incidence of preeclampsia in women conceiving by intrauterine insemination with donor versus partner sperm for treatment of primary infertility. Adv Immunol 1994. 170: 1244–50. Am J Reprod Immunol 1990. Smith SK et al. Colburn GT. Hunt JS. Cloning of a T cell growth factor that interacts with the beta chain of the interleukin-2 receptor. ectopic. J Immunol 1995. Mellon T et al. 264: 965–8. Lack of HLA-G expression in extravillous trophoblasts is associated with preeclampsia. Marti JI. and granulocytemacrophage colony stimulating factor levels in human amniotic fluid at term. and molar pregnancy. Thellin O. 121. Main EK. 57: 129–44. 143: 1131–41. factor are expressed on human placental syncytiotrophoblast. 134. Navarro F. 131. Greenfield C et al. 132. Fujii T. 132: 140–9. Am J Reprod Immunol 1996. 133. Yelavarthi KK. 184: 539–48. Eisenman J et al. Grabstein KH. 128: 489–93. Redman CW. 177: 455–8. Placental Fas ligand expression is a mechanism for maternal immune tolerance to the fetus. Locksley RM et al. Expression of the nonclassic histocompatibility antigen HLA-G by preeclamptic placenta. Ariel I. Matsuzaki N. ILT2 (LIRI) and CD94/NKG2A NK receptors respectively recognize HLA-G1 and HLA-E molecules co-expressed on taget cells. Lancet 1994. Cytotoxic activity against trophoblast and choriocarcinoma cells of large granular lymphocytes from human early pregnancy decidua. Vince GS. 36: 349–58. Chiang MH. Immunohistochemical localization of interferons in human placental tissues in normal. Sibai BM. 123. 117. 344: 973–5. Interleukin-6. Tolerance to the foeto-placental ‘graft’: ten ways to support a child for nine months. Curr Opin Immunol 2000.

.

1. the basic structures of the early placenta.1). Human implantation is almost unique amongst mammals in that it is highly invasive. The uterus loses its innervations during pregnancy3 and the placenta and cord are not innervated at all. providing an extensive and intimate interface for maternofetal exchange. In this chapter we have reviewed the basic hemodynamic concepts of the placental circulations and their relationships to maternal cardiac adaptation to pregnancy and embryonic and subsequently fetal cardiac development.1 Ample dilatation of the uteroplacental circulation together with rapid villous angiogenesis are the key factors necessary for adequate placental development and function and subsequent fetal growth. Figure 5. whereas the villi on the opposite pole degenerate to form the placental membranes (Figure 5. The correlation of Doppler ultrasound findings with anatomical and physiological features suggests that the establishment of high flow–low resistance circulation in both placental circulations is primarily the consequence of the considerable increase of the diameter of the corresponding vascular bed. and the conceptus embeds itself completely within the maternal uterine endometrium and superficial myometrium. The chorionic villi.2 and surround the entire gestational sac until 8–9 weeks of gestation. the length of the vascular network and the blood viscosity having a much smaller influence. the villi at the implantation site become elaborately branched and form the definitive placenta. form during the 4th and 5th weeks postmenstruation.4 These findings imply that the development of a low resistance to blood flow in the placental circulation is essentially the result of anatomical transformations and/or biochemically induced vasomotor mechanisms.5 The tortuous ascending uterine artery gives off approximately 8–10 arcuate branches which extend inward for about one-third of the thickness of the myometrium and envelop the anterior and posterior walls of the uterus. Normal development of the uteroplacental circulation The human uterine vasculature is made of a complex vessel network which anastomoses with branches of the ovarian and vaginal arteries to establish a vascular arcade perfusing the internal genital organs. From this network arise .5 Placental circulations Eric Jauniaux and Graham J Burton Introduction The placenta in mammals is the essential interface between the maternal circulation carrying O2-rich blood and nutrients and the fetal circulation. whereas in the periphery an intervillous circulation is established (arrows). Toward the end of pregnancy the villi present a surface area of 12–14 m2.1 Diagram of a complete gestational sac at 8 weeks of gestation showing the presence of the trophoblastic shell and intravascular spiral artery plugs in the center of the placental bed. Between the 3rd and 4th months.

Modern anatomic and in vivo studies have shown that human placentation is in fact not truly hemochorial in early pregnancy. The architecture of their decidual and myometrial parts is disrupted with the loss of myocytes from the media and the internal elastic lamina.1). but it may play a pivotal role in affecting the quality of that flow in terms of the perfusion pressure. and these essential arterial components are progressively replaced by fibrinoid material. Anatomical and radiographic studies including uterine perfusion experiments have demonstrated that the uterine vascular network elongates and dilates steadily throughout pregnancy. the spiral arteries which. which are discharged into the intervillous space until at least 10 weeks.14 This transformation.11 They gradually extend laterally.17 Development of the intervillous circulation The intervillous circulation in the hemochorial placenta has been referred to as an open system compared to other circulatory beds where the blood is retained within arteries. rather than decreases. a combination of sonographic in vivo investigation.42 Fetal Cardiology the radial arteries followed by the spiral arteries which enter the endometrium. and to a large extent that is still the case in the second half of pregnancy when fetal weight gain is greatest.13 Human placentation is also characterized by a remodeling of the spiral arteries. and have led to the hypothesis that the placenta limits. transport just a few milliliters of blood per minute need to carry around 600 ml per minute. and equal to that of the internal iliac arteries. termed ‘physiological changes’. the human placenta has been considered to act as a large arteriovenous shunt. reaching the periphery of the placenta around midgestation. However. in the non-pregnant state. and that by mid-pregnancy the diameter of the arcuate arteries exceeded that of the uterine arteries. This is supplemented by secretions from the uterine glands. Physiological transformation of the uterine circulation At term. as they approach their target organ. It is a gradual process which starts at implantation and which is then linked to the trophoblastic infiltration of the endometrium and superficial myometrium. but they are not observed in the opposite wall of the uterus. by term some of the arcuate arteries were twice the diameter of the uterine arteries. In the past. However. Our recent combined in vivo–in vitro investigations have resulted in a new understanding of the maternofetal relationship during the first trimester of pregnancy. From radiologic measurements. From the time of implantation.8.12 These trophoblastic cells penetrate the myometrium via the intercellular ground substance. vascular casting. can be first found both within and around the spiral arteries in the central area of the placenta. but the extent of invasion is progressively shallower toward the periphery. Depth-wise the changes normally extend as far as the inner third of the uterine myometrium within the central region of the placental bed. oxygen supply to the fetus during the period of organogenesis. affecting its mechanical and electrophysiological properties.6 The basis of the adjustment of the maternal placental flow rates is the transformation of the uterine vasculature. through capillary beds to veins.10 Consequently these vessels lose their responsiveness to circulating vasoactive compounds.18. trophoblastic cells infiltrate the decidua from the proliferating tip of the anchoring villi and from the trophoblastic shell. results in the metamorphosis of small-caliber spiral vessels into flaccid distended uteroplacental arteries. In normal pregnancies. the pulsatility and rate of blood flow. Indeed.9.7.20–23 The shell and the plugs act like a labyrinthine interface that filters maternal blood.24 Recently. and even in normal pregnancy.8.16.7–10 These extravillous trophoblastic cells. which is associated with the peripheral widening of the supplying arteries by tissue growth and remodeling of the arterial wall. there is a gradient in the infiltration of the trophoblast along the spiral artery. rather than facilitates. it has been assumed that the principal function of the placenta is to supply the fetus with as much oxygen as possible. Burchell15 observed that the diameter of the uterine artery doubled by 6. which are non-proliferative. not all spiral arteries are completely transformed. The cells of this shell anchor the placenta to the maternal tissue but also form plugs in the tips of the uteroplacental arteries.5 weeks of pregnancy.19 Because the spiral arteries open into essentially a large lake of blood and the intervillous space does not impose any impedance to flow. the transformation of spiral arteries into uteroplacental arteries is described as completed around mid-gestation. permitting a slow seepage of plasma but no true blood flow into the intervillous space. Within this context the uteroplacental circulation differs from other vascular beds in that the diameter of the vessels increases. The main aim of these vascular changes is to optimize the distribution of maternal blood into a low-resistance uterine vascular network and ultimately inside the placental intervillous chamber. the physiological conversion may not be so important in terms of volume of intervillous blood flow. and the consistency of the flow.6 When the blastocyst attaches to the uterine wall.25 Whether the formation of these shunts is related to trophoblast invasion is not clear. the extravillous trophoblast not only invades the uterine tissues but also forms a continuous shell at the level of the decidua (Figure 5. and oxygen measurements has shown conclusively that extensive shunting occurs within the myometrium under the placental bed. .

16. and the conceptus is fully embedded in the uterine wall before the primitive streak has formed.2). placental development is precocious.Placental circulations 43 In the human.2. Consequently.26–28 In normal pregnancies this is a very well controlled phenomenon that has to provide a delicate balance between the metabolic needs of the fetus and its placenta and the potential danger of oxygen free radicals (OFR).29 These features support the concept that in normal pregnancies. reflecting to some extent the evolutionary path. the placental villi display only a few capillaries and fetal erythrocytes are nucleated. Placental trophoblastic cells are extremely sensitive to oxidative stress because of their extensive cell divisions and the concomitant exposure of their DNA. and the early placenta and fetus are separated by the exocoelomic cavity (Figure 5. Furthermore. other strategies need to be employed to restrict exposure of the fetus to oxygen. Spectral analysis of blood velocity waveforms obtained from the main uterine artery at 14 weeks (c) and 20 weeks (d) of gestation.1. and as a result the human placenta is exposed to major fluctuations in O2 concentration from conception to delivery. the earliest stages of development take place in a low oxygen environment. which occupies most of the space inside the gestational sac. it is well established that maternal metabolic disorders such as (a) (b) (c) (d) Figure 5. during the first trimester the villous membrane is twice the thickness it will be in the second. Note the protodiastolic notch at 14 weeks (c).28 Thus.17 suggesting that the fetal blood is extremely viscous and consequently the fetoplacental blood flow limited. During that period.2 Two-dimensional (2D) (a) and 3D (b) color flow mapping of the main uterine artery crossing the iliac vessels. .

and cows.44 Fetal Cardiology diabetes.37 It has been suggested that the fall in peripheral systemic vascular tone is the main factor triggering the rise in cardiac output in early pregnancy. allowing maternal blood to flow progressively freely and continuously within the intervillous space. The second example involves an ischemia–reperfusion (I/R) phenomenon. are known to be associated with a higher incidence of miscarriages. this well-controlled oxidative stress may play a role in continuous placental remodeling and essential placental function such as transport and hormonal synthesis.1). at the same time plasma volume increases by about 40%. catalase. arising from spontaneous vasosonstriction.32 These events bring the maternal blood closer to the fetal tissues. and glutathione peroxidase. direct intra-arterial uterine infusion of estradiol in ewes induces a dramatic increase of the uterine blood flow in both the pregnant and non-pregnant uterine vascular bed. A lower maternal blood viscosity potentiates the fall in peripheral vascular resistance. and there are two examples when this occurs physiologically during human pregnancy. Their role is emphasized by the fact that in the human. resulting in hemodilution and. which are associated with an increased generation of OFR. At the end of the first trimester the trophoblastic plugs are progressively dislocated. allowing limited maternal blood flow to enter the placenta from 8–9 weeks of gestation (Figure 5. and fetal structural defects. Most of the 30–60% increase in cardiac output and half of the 10% fall in arterial pressure is accomplished in the first trimester.28.34 This chronic stimulus could lead to upregulation of the anti-OFR defense in the placenta. the exocoelomic cavity is obliterated by the growth of the amniotic sac.33 Although equivalent studies have not been performed in the human.39 Furthermore. at the end of the first trimester. Focal trophoblastic oxidative damage and progressive villous degeneration trigger the formation of the fetal membranes.31 for at physiological levels free radicals regulate a wide variety of cell functions. about 40% above the non-pregnant level. This leads to higher local oxygen concentrations at a stage of pregnancy when the trophoblast possesses low concentrations and activities of the main antioxidant enzymes superoxide dismutase. and even by postural changes. in particular transcription factors. The physiological hypoxia of the first trimester gestational sac may protect the developing fetus against the deleterious and teratogenic effects of OFR. First. The oxidative stress may also stimulate the synthesis of various trophoblastic proteins. and the gradual pattern of plasma volume expansion is out of phase with the rapid pattern of increase in maternal cardiac output and drop in arterial pressure. such as horses. However. It is only from the end of the first trimester onward that this increased cardiac output is due to an increase in stroke volume.26 The underlying uteroplacental circulation in this area is never plugged by the trophoblastic shell. in some animal species. Maternal total red cell volume increases steadily throughout gestation. Placental inflow may also be compromised by external compression of the arteries during uterine contractions in both the rhesus and human. Angiographic studies of the uterine vasculature of the rhesus monkey have demonstrated that during normal pregnancy flow from spiral arteries into the intervillous space is often intermittent. maternal tissue is not infiltrated by trophoblastic cells and the conceptus is adequately perfused by maternal blood. thus reducing oxidant stress. Recent evidence also indicates that it is necessary to maintain stem cells in a fully pluripotent state. two-thirds of the primitive placenta disappears. while progesterone partially inhibits the vascular effect of estradiol. A variety of endocrine substances are known to influence the uterine artery resistance. Maternal hemodynamic changes and regulation of the uteroplacental circulation The maternal plasma volume increases gradually from the first month of gestation to a plateau in the third trimester. therefore. For example. and its extent is considerably influenced by oral iron intake.33 Some degree of I/R stimulus may therefore be a feature of normal human pregnancy.30 indicating that the mammalian conceptus can be irreversibly damaged by oxidative stress. the decrease in uterine vascular resistance starts during the luteal phase of the menstrual cycle.35 The increment varies from 18 to 30% of non-pregnant values.38 The resulting rapid fall in preload and afterload leads to a compensatory increase in heart rate and activation of the volume restoring mechanisms. in a decrease of maternal blood hematocrit and blood viscosity. It is well established that estrogen produces important changes in uterine blood flow in both human and various animal species. facilitating nutrient and gaseous exchange between the maternal and fetal circulations. and maternal blood flows progressively throughout the entire placenta. Blood viscosity is directly related to the hematocrit.32 which is an essential developmental step enabling vaginal delivery. As in early pregnancy. pigs. especially toward term when the fetus and placenta are extracting large quantities of O2 from the intervillous space.40 A similar response is observed in postmenopausal .28 Excessive production of OFR results in oxidative stress. vasculopathy. During the transitional phase of 10–14 weeks’ gestation.35 The mechanisms regulating these changes in human pregnancy are still unknown. the general similarity of the uteroplacental vasculature and delivery of blood into the intervillous space to those of the rhesus monkey has led to the assumption that intermittent perfusion of the intervillous space also occurs in our species.36 and cardiac output continues to increase even in late pregnancy. a burst of oxidative stress is evidenced in the periphery of the early placenta.

61 They also demonstrate that. therefore. The evidence available suggests.Placental circulations 45 women receiving hormone replacement therapy. Blood flows in the spiral (Figure 5. there is a rapid increase from 50 to 120 cm/s in the mean peak systolic velocity (PSV) of the main uterine artery. showed negative correlation of estradiol and a positive correlation of progesterone with uterine artery impedance but no correlation with spiral artery flow characteristics.32.49 When comparing Doppler features of the placental circulation at different gestational ages.45 Although there is no obvious role for the hCG molecule in the regulation of uteroplacental hemodynamics.44 Discrepancy between different studies suggests that the relationship of sex steroids to uterine Doppler velocimetry during pregnancy is more complex than previously thought.2). Our recent data support the concept that in normal pregnancies the intervillous circulation is gradually established between the beginning of the third month and the end of the fourth month of gestation.32 Abnormal development of the uteroplacental circulation Placental-related disorders of pregnancy are almost unique to the human species.8 The resistance index (RI) or the pulsatility index (PI) measured from the FVWs recorded at the level of the main uterine arteries reflects the downstream flow impedance in the whole uterine circulation. this finding suggests that hCG secretion may be influenced by changes in intraplacental hemodynamics and possibly by changes in placental oxygen tension.46 Doppler studies have demonstrated a progressive decrease in the downstream resistance to blood flow in the uterine circulation from implantation to term.60 In view of the potential danger from oxidative stress. Maternal serum 17β-estradiol levels may also have a significant influence on uterine resistance to blood flow during pregnancy. in normal pregnancies.2 Both left and right main arteries must be investigated at the same time.3).51 These Doppler data are in agreement with histologic data. Ultrasound/Doppler features of the uteroplacental circulation The various branches of the uterine circulation can be differentiated by means of color Doppler imaging. A decrease of the resistance and pulsatility indices from the main uterine artery towards the spiral arteries can also be demonstrated at different stages of pregnancy.56 and anatomists57.48–54 This decrease can be observed in all segments of the uterine circulation. with communication between the uteroplacental arteries and the intervillous space being established in a small number of vessels at a time from the end of the second month of pregnancy onward. independent of gestational age. the debate has become obsolete. and this is partially reversed by progestogen.3) arteries are characterized in pregnancy by a low-impedance irregular flow pattern which shows no significant changes in shape throughout pregnancy. In 85% of pregnancies.6. we found that a nonpulsatile signal corresponding to maternal intraplacental blood flow could not be identified inside the intervillous space before 10 weeks of gestation.59. starting in the periphery and expanding progressively to the rest of the placenta thereafter. and this may be related to regional differences in the extent of plugging of the maternal arteries. published around the same time. Impedance to blood flow through spiral arteries in the second trimester is lower in the central area of the placental bed than in peripheral areas. and of limited value in understanding the pathophysiology of placental-related pregnancy disorders.47 and may reflect the end of the implantation process and its associated physiological changes. Thus.2 This finding has been confirmed by other authors.46 However.2 and 5.42 In addition.58 about the status of the maternal circulation during the first trimester. and requires further investigation.43 Another study. End-diastolic flow (EDFs) increase in the main uterine arteries and their branches during the second half of the menstrual cycle. maternal serum levels of relaxin may also be a significant contributor to uterine resistance to blood flow as assessed by Doppler indices. which affect around . as unilateral measurement may provide erroneous results concerning true uterine perfusion. that the normal establishment of a continuous intervillous circulation is an incremental phenomenon.41 At physiological levels estradiol decreases the resistance to flow in the uterine circulation. Between 12 and 14 weeks of gestation. These disorders. This concept is supported by the immunohistochemical and morphological evidence of temporospatial differences in the degree of trophoblastic oxidative stress. the uteroplacental circulation is a dynamic model in which the magnitude of blood flow through a single vessel may vary importantly (Figures 5. and it is likely that onset of maternal blood flow to the placenta is normally a progressive phenomenon. A close temporal relationship has been found between the Doppler detection of a continuous intervillous flow and the human chorionic gonadotropin (hCG) peak. In non-pregnant women and during the first half of normal pregnancy.52. flow velocity waveforms (FVWs) from the main uterine arteries are characterized by a well-defined protodiastolic ‘notch’ (Figure 5.55 provoking a vigorous debate involving both ultrasonographers53. the evaluation of blood flow in single uteroplacental vessels is often difficult to interpret. the protodiastolic notch disappears before 20 weeks of gestation. onset of maternal blood flow is most often initiated in the peripheral regions of the placenta. and the overall Doppler mapping features correlate well with the classical and modern anatomic findings. and this increase continues as pregnancy advances.

46 Fetal Cardiology (a) (b) (c) (d) Figure 5.64 There is mounting evidence that oxidative stress or an imbalance in the oxidant/antioxidant activity in uteroplacental tissues plays a pivotal role in the development of placental-related diseases.32 This is associated with the absence of physiological changes in most spiral arteries. such as monkeys.21.4). the consequences are .62 and spontaneous preeclampsia has only been exceptionally described in ‘patas’ monkeys63 and guinea pigs. a third of human pregnancies.3 2D (a) and 3D (b) color flow mapping of the placental bed. reduced cytotrophoblast invasion of the endometrium. and indirect O2mediated widespread trophoblastic oxidative damage and increased apoptosis. Although epidemiological data on animals living in the wild.32. the excessive entry of maternal blood into the intervillous space has two effects: a direct mechanical effect on the villous tissue which becomes progressively enmeshed inside large intervillous blood thrombi. and incomplete plugging of the lumen at the tips of the spiral arteries.12.65 Overall. In about two-thirds of early pregnancy failures there is anatomical evidence of defective placentation. laboratory rodents are known to have postimplantation pregnancy loss rates of less than 10%. Independent of the etiology of the miscarriage. primarily include miscarriage and preeclampsia. In other mammalian species. intervillous flow) in (d). Spectral analysis of blood velocity waveforms obtained from spiral artery at 10 weeks of gestation (c) and from the intervillous space at 14 weeks of gestation (d). which is mainly characterized by a thinner and fragmented trophoblast shell. the incidence of both disorders is extremely low.20. and leads to a premature onset of the maternal circulation throughout the entire placenta (Figure 5.12. are limited. Note the discontinuous blood flow with a venous pattern (i.e.

These ultrasound anomalies are commonly found in association with raised maternal serum α-fetoprotein levels and with early fetal growth restriction (FGR).4 Diagrams showing placentation in a normal ongoing pregnancy (a). there is evidence of an abnormal intervillous circulation from the beginning of the second trimester. In the most severe cases. maternal blood flow will enter the intervillous space at greater velocity and pressure than normal. in an early pregnancy failure (b). also called maternal lakes.69 Most studies have focused on the fact that an increased impedance to flow in the uteroplacental circulation and persistent uterine artery notching at mid-gestation predict the development of placental-related pregnancy disorders. The differences in Doppler technique employed in the different studies might partly explain their discrepant results. the plugs in the lumens of the spiral arteries. Failure of conversion by itself should not influence the volume of maternal blood to the placenta. Note in (a) the continuous trophoblastic shell.Placental circulations 47 (a) endothelial cell dysfunction. and the interstitial migration of the extravillous trophoblast through the decidua down to the superficial layer of the myometrium.10 Incomplete conversion of the spiral arteries results in retention of smooth muscle cells within their walls. which is characterized by major placental anatomical changes that have been described on ultrasound as ‘jelly like’. and in a complete hydatidiform mole (c). and the examination was repeated in those with increased impedance to flow at around 24 weeks. with an overall reduction in the placental mass echogenicity. in a manner similar to the formation of the central cavity of a lobule at the start of the second trimester. have demonstrated an association between increased impedance to flow in the uterine arteries and subsequent development of preeclampsia. a phenomenon that we described as the ‘hose effect’. FGR. Other factors such as the definition of abnormal flow. Rather.68 These features refer to the overall appearance of the placenta.34 and that consequently the placenta suffers a chronic low-grade ischaemia–reperfusion type injury. the populations. the gestational age . the absence of plugs. for while invasion is sufficient to anchor the conceptus it is insufficient to fully convert the spiral arteries into low-resistance channels. and the reduced migration of extravillous trophoblastic cells. though lesser. with its chorionic plate being pushed up by jet-like bloodstreams. but more importantly to intermittent perfusion. This may lead not only to diminished perfusion of the intervillous space. This mechanism is common to all miscarriages. placental degeneration with complete loss of syncytiotrophoblast function and detachment of the placenta from the uterine wall. suggesting damage to the villous membranes and in particular to the trophoblastic layer. Initial screening by continuous-wave Doppler was carried out at 18–20 weeks of gestation. Since the placenta and fetus continually extract oxygen. In vivo features of abnormal placentation (b) (c) Figure 5. and the force is sufficient to drive apart the villous branches and form intervillous lakes. and perinatal death. defect in early trophoblast invasion. both in the second and more recently in the first trimester of pregnancy. which strongly suggests that preeclampsia is at the extreme end of a continuum of changes seen in every pregnancy. it is expected that transient hypoxia will result. The nature of the inflammatory response and oxidative stress seen in preeclampsia is not intrinsically different from that seen in normal pregnancies. as this is mainly determined by changes in the radial and arcuate arteries. leading to chronic oxidative stress in the placenta and finally to diffuse maternal The ability of transvaginal color Doppler to detect flow velocity waveforms from small vessels such as from the terminal part of the uteroplacental circulation has given rise to much enthusiasm from clinicians interested in predicting early and late pregnancy complications related to an abnormal placentation. and so some vasoreactivity persists in 30–50% of the placental vascular bed. The difference is one of degree.17 Preeclampsia stems from a similar.66–68 On ultrasound the inflow appears as jet-like streams surrounded by turbulence. with the primary pathology being an excessive or atypical maternal immune response. in (b) the discontinuous trophoblastic shell. in (c) the absence of trophoblastic plugs and interstitial migration. the time at which it occurs in the first trimester depending on the etiology.68 Several Doppler screening studies.17 This would impair the placentation process. These findings have led us to suggest that preeclampsia is a threestage disorder.

the premature entry of a larger than physiological quantity of maternal blood inside the placenta could be associated with a significant increase in second trimester complications such as preeclampsia. and this ultimately unites with the allantoic vessels developing in the connecting stalk to establish the fetal circulation to the placenta. As a consequence. oxidative damage to the trophoblast is significantly increased.5.74 From 6 to 9 weeks. and differentiate to form pericytes.5). in particular in the central area.69 In complicated early pregnancies. indeed it is thought that the differential elongation of the capillary network to that of the containing villus causes vascular loops to obtrude from the surface. and frequently on the apex of a tight bend the capillaries become greatly . By the beginning of the 4th week the cords have developed lumens and the endothelial cells become flattened. Recent data suggest that increased impedance at 23–24 weeks identifies about 40% of those that subsequently develop preeclampsia and about 20% of those that develop FGR.71 These findings suggest that in early pregnancy failure the initial central trophoblastic migration and vascular plugging is insufficient. this observation has not been disputed.69 Overall. and this will prevent the normal villous trees from developing and hence compromise the anchoring of the placenta. even at an early stage. and this pivotal phenomenon has been documented in utero as early as 36 days’ menstrual age.69 In women with increased impedance to flow.48 Fetal Cardiology at which women were examined. and the sensitivity of abnormal Doppler features for preeclampsia requiring delivery prior to 34 weeks of gestation is about 80% and for FGR it is about 60%. The premature entry of maternal blood into the intervillous space at this stage of pregnancy will disrupt the placental shell. Angiogenesis continues until term through a series of different phases which most probably reflect different concentrations and combinations of growth factors induced by the changing intrauterine environment. and is probably the mechanical cause of abortion. This mechanism is common to all miscarriages.1 Around the end of the 5th week of gestation. In a study correlating ultrasonographic with pathological features. so creating a new terminal villus. connections form between neighboring tubes to create a plexus.32.50 Normal development of the umbilicoplacental circulation The development of the fetal vasculature begins during the 3rd week post-conception (5th week of pregnancy) with the de novo formation of hemangioblastic cell cords within the villous stromal core. the umbilicoplacental circulation is considered to be a passive circulation in which the flow rate is determined by the mean effective perfusion pressure. and the criteria for the diagnosis of preeclampsia and FGR may also have contributed to the wide variation in detection rates. Surrounding mesenchymal cells become closely apposed to the tubes.32. the time at which it occurs in the first trimester depending on the etiology. we found extensive dislocation of the trophoblastic shell and a massive infiltration of the intervillous space by maternal blood in cases presenting with a continuous intervillous blood flow on color mapping before 12 weeks of gestation. the primitive heart begins to beat. and most of these have shown that. Because of the absence of innervation beyond the proximal 1–2 cm of the umbilical cord.70 We have recently shown that at 8–9 weeks and 10–11 weeks an intervillous flow is significantly more commonly detected inside the placenta of abnormal pregnancies. The fetoplacental circulation is established from around 8 weeks of gestation. the placenta appears on color flow mapping to be hypervascularized well before the end of the first trimester.78 The caliber of the fetal capillaries is not constant within intermediate and terminal villi. There are only a few first trimester screening studies.61.71 The predictive value of Doppler measurements in early pregnancy is limited.70 Unlike the findings on the development of the intervillous circulation in normal pregnancies. The pooled LRs for FGR were about 2 and 0. allowing the entry of larger than normal quantities of maternal blood into the placenta. respectively.77 From 25 weeks onwards. Around 28 days post-ovulation (6 completed menstrual weeks). and for those with normal Doppler results the LR was about 0. there is a rapid increase of the mean heart rate to a plateau in the second and third trimesters. the villous vasculature is connected with the primitive heart and the vascular plexus of the yolk sac via the vessels of the connecting stalk.12.66 The development of the fetoplacental circulation is characterized by a progressive rise in blood flow and decrease in vascular resistance to flow. abnormal Doppler is better at predicting severe disease. the pooled likelihood ratio (LR) for the development of preeclampsia was about 5. All Doppler studies in the first trimester have failed to demonstrate abnormal blood flow indices in the uteroplacental circulation of pregnancies that subsequently ended in spontaneous abortion or in anembryonic gestation.9. impedance to flow in the uterine arteries can be increased in pregnancies that subsequently develop preeclampsia and FGR.75 Development of the villous circulation Early in pregnancy the capillary network is labile and undergoes considerable remodeling. terminal capillary loops are produced.76.61.17. and the differences between the Doppler signals and the histopathological findings in early pregnancy failure compared to normal pregnancies are so striking that they constitute one of the strongest arguments in favor of the Hustin and Schaaps hypothesis (Figure 5.12.72. However.73 During the next few days.

78.80 The villous circulation of the definitive placenta is composed of muscularized stem arteries (750 μm). (c) Color flow mapping of a gestational sac in a missed miscarriage at 9 weeks (last menstrual period) showing diffuse intervillous signals (arrow). and perivascular capillary networks. dilated. forming sinusoids. 15–20 μm in diameter. Hemodynamics of the umbilicoplacental circulation During the second half of pregnancy.5 (a) Color flow mapping of a gestational sac containing a 5-mm embryo with a heart activity and showing the placental bed circulation.Placental circulations 49 (a) (b) (c) (d) Figure 5. branching more than 10 times and ending in long capillary loops.79 Placental capillary formation is only completed by mid-gestation. De novo formation of capillaries from the transformation of mesenchymal cells is rare in mature placental tissue. These regions may help to reduce vascular resistance and so facilitate distribution of fetal blood flow through the villous trees. and only occurs in persisting mesenchymal villi. (b) Histological view of a decidual biopsy collected from the area under the definitive placenta showing trophoblastic plugs inside the lumen of the tips of the spiral arteries at 9 weeks of gestation (arrows). the vascular network with the lowest resistance in the entire fetal circulatory . (d) Histological view of a decidual biopsy in a similar case of missed miscarriage at 9 weeks showing the poor transformation of the tip of the spiral arteries and the absence of trophoblastic plugs (arrows).

and more recently spectral Doppler index mapping92–95 have facilitated the visualization of smaller intraplacental vessels. the resistance depending on the radius to the fourth power. Shear stresses may be highest in the first trimester umbilicoplacental circulation. The number of fetal capillaries per villous profile and the proportion of the villi (volume fraction) occupied by the fetal capillaries remains low (1–2/villi) during the first 2 months of pregnancy. indicating that another vascular-endothelial mechanism. the end-diastolic flow (EDF) develops rapidly but is incomplete and/or inconsistently present.7).90 In early pregnancy. the extraembryonic circulation is mainly vitelline (as opposed to placental). or through angiogenesis. a decrease in blood viscosity and/or in the vascular system length and an increase in the mean radius of the vascular system collectively decrease the resistance to flow. Changes in the pressure gradient due to the expansion of the intervillous space and/or a modification of the local concentration of vasodilators could also influence relaxation of the small placental arteries. Doppler signals can be obtained from the different segments of the umbilicoplacental circulation (Figure 5. Vasoconstrictor substances include thromboxane A2 produced by the platelets and the umbilical vessels. there is no relationship between umbilical artery Doppler impedance indices. or new villous vessel formation. the reduction in impedance occurs despite a decrease in NO production. as the mean radius of the villous vascular system is low and the viscosity high. The addition of color flow mapping2. the resistance to flow in the early umbilicoplacental circulation must be high. all embryonic erythrocytes are nucleated. three-dimensional (3D) power Doppler.88 Between 12 and 14 weeks. and endothelin-1 synthesized locally within the placenta. The mechanism by which a reduction in vascular impedance through the fetoplacental circulation occurs during the second trimester is not known. As gestation advances. fetal heart rate. Since umbilical vessels are not innervated. PI values from the umbilical artery remain high. vasodilatation must occur through one of two mechanisms: through a direct vasorelaxant paracrine effect of an agent acting upon vascular smooth muscle. Subsequently. angiotensin II produced by the fetal kidney. the rapid appearance of the EDF in the umbilical circulation and the drop in umbilical artery PI value coincide with the time of the establishment of the intervillous circulation. these cells account for less than 10% of the total red cell population in the fetal circulation.85 Maternal serum hCG and relaxin levels have an independent correlation with the fetal heart rate. This changing pattern of blood flow after 12 weeks could lead to reduced endothelial stimulation.2 There is also no correlation between umbilical artery resistance to flow and fetal blood hematocrit or umbilical cord length later in pregnancy. Thus.6).89 or villous angiogenesis.50 Fetal Cardiology system is at the level of the placental vascular bed. the villous capillary bed has the greatest potential to influence total umbilicoplacental vascular impedance.84 Possible vasodilators produced by the fetus are the atrial natriuretic peptide and prostacyclin.2 Several humoral factors are known to modulate the tone of the cord and villous vessels of the term placenta.87 These data suggest that NO and cGMP may play an important part in maintaining flow through the early first trimester fetoplacental circulation. The passage of fetal blood through an anatomically high-resistance circuit could lead to endothelial stimulation of nitric oxide synthase (NOS) activity. and does not appear to be directly related to the rapid drop in umbilical artery impedance observed between 12 and 14 weeks’ gestation. the number of nucleated erythrocytes decreases rapidly. umbilical arteries show a high degree of vascular resistance to blood flow expressed by narrow systolic waveforms. and spectral analysis also shows a decline in resistance indices with advancing gestation and toward the .16.87 In vivo features of the umbilicoplacental circulation In the first 3 months of gestation. multigate spectral velocimetry. absence of EDF.81 According to Poiseuille’s law.83 Furthermore.91 to Doppler equipment. which suggests minor changes in umbilical placental vascular resistance until the end of the first trimester. and consequently to a rapid reduction of both NOS and GMP levels from the villous trophoblast. the formation of villous capillaries is progressive. the trend in the RI or PI shows a gradual decrease until the end of the third trimester.2. As for the uteroplacental circulation. and high PI values (Figure 5.1. End-diastolic velocities are present at the level of intracerebral arteries 2 weeks earlier than in the umbilical artery. thus maintaining vasodilatation within the umbilicoplacental circulation until anatomical changes occur. After that period. Diastolic frequencies throughout the entire cardiac cycle are recorded in the umbilical artery of the normally developing fetus from 14 weeks onward (Figure 5.2 indicating that the fetal blood viscosity is high in early pregnancy.82 The latter effect is particularly strong. is responsible for reducing umbilical artery resistance at this critical stage in the development of the fetoplacental circulation. prior to the development of new villous blood vessels and the appearance of a low-impedance.7).64 In early pregnancy. up to the middle of the second month of gestation. From an anatomical perspective. However.86 but the influence of other maternal or placental factors on the hemodynamics of the developing umbilicoplacental circulation remains uncertain. and clearly visualize umbilical cord coiling (Figure 5.17 This suggests that for at least the first 2 months of gestation. NO and cyclic guanosine monophosphate (cGMP) concentrations are positively correlated with umbilical artery impedance between 9 and 15 weeks’ gestation. positive end-diastolic flow umbilical artery velocity waveform.6). and by 12 weeks of gestation.

Placental circulations 51 (a) (b) (c) (d) Figure 5. Doppler investigation of the branches of the umbilicoplacental circulation is limited to chorionic and main stem villous (approximately 750 μm in internal diameter) vessels.97–99 Various attempts have been made to correlate the Doppler abnormalities with placental structural changes in order to provide a mechanistic explanation for their origin. From analysis of the umbilical waveform it is possible to assess the impedance to placental blood flow. It has been proposed that the placenta is hyperoxic. and are found within the context of FGR associated with congenital infection or chromosomal abnormalities and disorders of uteroplacental circulation leading to chronic fetal hypoxia. and include mainly chorioangioma and molar villi. 14 weeks (c).96 This theory may explain the basis for many of the morphological changes observed. and the villous capillary bed cannot be directly explored with these techniques. but does not account for how the hyperoxia is initiated. in cases of severe intrauterine growth restriction. The results have been . rather than hypoxic as is commonly assumed. Abnormal villous vascular development Primary anomalies of villous angiogenesis are rare. Note the progressive appearance of the end-diastolic flow.6 Color Doppler mapping and spectral analysis of the umbilical cord blood flow at 7 weeks (a). and 16 weeks (d). intraplacental arterioles. 11 weeks (b). Secondary anomalies are more common. and to accurately predict fetal hypoxia.49 However.

varied. Note the difference in coiling of the cord between the second and the third trimester views. 2D and 3D color Doppler mapping of a free loop of umbilical cord at 16 weeks (c) and 32 weeks (d). it is most likely that they are a secondary phenomenon. Histomorphometric features are consistent with previous findings that increasing severity of abnormal Doppler waveforms in the uterine and umbilical circulations is associated with fetal distress and hypoxia.95 As Doppler abnormalities of the umbilical circulation are rarely seen in the absence of uterine arterial abnormalities. The absence and a reverse of EDF in the umbilical arteries are common findings in pregnancies complicated by severe fetal growth restriction.52 Fetal Cardiology (a) (b) (c) (d) Figure 5. although the fact that Doppler changes in the umbilical circulation are invariably seen subsequent to similar changes in the uterine arteries strongly suggests that they are a secondary phenomenon.7 Color Doppler mapping of the umbilical cord placental insertion at 12 weeks (a) and 20 weeks of gestation (b).100 Abnormal intraplacental blood flow at 28–34 weeks of gestation is also strongly associated with FGR. The underlying cause of the placental lesions is not known. Deficient trophoblast invasion during early pregnancy . ranging from claims of a reduction in the number of arteries within the supporting stem villi to a reduction in the capillary vascular bed within the terminal villi. the principal site of gaseous exchange. Note the entry of one of the umbilical arteries inside the placenta (a) and the branching of the chorionic vessels on the surface of the placenta (b).

102 The umbilical artery PI is also abnormally high in some fetuses investigated with trisomy 18 and triploidy. Pijnenborg R. Placenta 1980.103 Elevated intraplacental resistance to blood flow indicates increased maternal intraplacental impedance as early as week 8 of gestation. The intermediate and terminal villi are the principal sites of gaseous exchange. 2nd edn. . Hum Reprod Update 2006. Russell PT. 1992: 469–85. 1994: 441–84. Placenta 2006. 34: 1265–75. 5. but also reduced oxygen extraction from the intervillous space and so hyperoxia on the venous side of the placenta as a tertiary event. Poston L. Hanssens M. Donner NW.104 Overall. Ultrasound Obstet Gynecol 1991. Itskovitz J. 9. The Human Placenta. 16. Jauniaux E. Kaufmann P. The pattern of interstitial invasion of the myometrium in early human pregnancy. appears to improve a number of obstetric care outcomes. 14. and appears promising in helping to reduce perinatal deaths. Campbell S.100 Such regression would increase vascular impedance in a reverse of the pattern seen during normal pregnancy. Brosens I. Garfield RE. Hempstock J. Burchell C. 55: 67–70. Robertson WB. 17. The Physiology of Reproduction. A functional and structural study of the innervation of the human uterus. Heidelberg: Springer-Verlag. Anatomy and genesis of the placenta. 7. suggesting an abnormal development of the villous circulation. Morizaki N. Brosens I. van Asshe A. 160: 218–28. Hanssens M. Moll W. Burton GJ. Hustin J. Eur J Obstet Gynecol Reprod Biol 2003. Hamilton WJ. 5: 67–74. eds. 12: 747–55. 15. Bland JM. Vercruysse L. Am J Obstet Gynecol 1969. 6. 21.49. Anat Rec 1979. Ramsey EM.Placental circulations 53 leads to incomplete conversion of the spiral arteries. Placenta 1990. Lindenbaum ES. McFayden IR. The human first trimester gestational sac limits rather than facilitates oxygen transfer to the foetus: a review. 1: 435–45. 1: 3–19. The uterine spiral arteries in human pregnancy: facts and controversies. Obstet Gynecol 1980. Meekins JW. These findings indicate that an abnormal EDF in early pregnancy can also be an ominous sign of adverse fetal outcome before mid-gestation as it is during the second half of pregnancy. Placenta 1981. Burton GJ. Brosens I. Placental Vasculature and Circulation. Arterial blood flow in the human intervillous space. eds. and so account for the changes in umbilical waveform observed. Br J Obstet Gynaecol 1994. Histological study of the materno-embryonic interface in spontaneous abortion. Bland JM. Trophoblastic invasion of human decidua from 8 to 18 weeks of pregnancy.100 There have been several reports of a fatal outcome of chromosomally normal fetuses between 11 and 18 weeks of gestation following the discovery of absent or reversed EDF in the umbilical artery earlier in pregnancy. These vessels remain of higher resistance than normal. Placenta 1983. Pijnenborg R. 71: 615–20. Burton GJ. Dixon G.105 Conversely. This in turn promotes a mild ischemia– reperfusion injury in the placental tissues. Kunzel W. 188: 277–86. Neill JD. 98: 303–11. 1970. 101: 669–74. The expansile behaviour of the human uterus. Placental-related diseases of pregnancy. Brandes JM. 10. 8. Jauniaux E. 4: 397–414. Structure adaptation and blood flow control in the uterine arterial system after hemochorial placentation. Burton GJ. 18. Hemodynamic implications of hemochorial placentation. Hustin J. Jauniaux E.67. Uteroplacental arterial changes related to interstitial trophoblast migration in early human pregnancy. and repeated insults during mid-pregnancy may lead to regression of the capillaries. particularly as a high percentage are not stabilized by a pericyte covering. Pijnenborg R. 20. 110: S19–27. Boyd JD. Cambridge: Heffer and Sons. Jauniaux E. 19. and decreased vascularization will inevitably impair placental exchange. Hayashi RH. and the retention of smooth muscle within the spiral arteries exacerbates their normal contractility. 13. New York: Raven Press. 2. Jurkovic D. Pijnenborg R. Robertson WB. The contribution of placental oxidative stress to early pregnancy failure. 27: 939–58. Neurohistochemical evidence supporting an absence of adrenergic and cholinergic innervation in the human placenta and umbilical cord. Morizaki J.34 leading to oxidative stress in the fetal vasculature. J Obstet Gynaecol Br Commonw 1964. routine Doppler ultrasound in low risk or unselected populations does not confer benefit on mother or baby. Jauniaux E. Eur J Obstet Gynecol Reprod Biol 1975. Herberger J. 2: 303–16. resulting in longer periods of vasoconstriction and hence greater fluctuations in oxygen tension. Hustin J. Stuttgart: Georg Thieme. the EDF at 13 weeks of gestation was either incomplete or absent. there is no doubt that the Doppler ultrasound study of umbilical artery waveforms helps to identify the compromised fetus in ‘high risk’ pregnancies. In: Barnea E.66 Absent or opposite gradient between the umbilical artery and the placental vessels was associated with adverse pregnancy outcome. Hum Pathol 2003. Dixon G. 11: 477–86. Am J Obstet Gynecol 1989. Greenwold N. The First Twelve Weeks of Gestation: A New Frontier for Investigation and Intervention. 24 (Suppl A): S86–93. Pijnenborg R. 1980. References 1. Placenta 2003. Moll W. A study of placental bed spiral arteries and trophoblast invasion in normal and severe pre-eclamptic pregnancies. In vivo investigations of anatomy and physiology of early human placental circulations.106 3. Arterial anastomosis in the pregnant human uterus. 4. This will lead to fetal hypoxia and growth retardation. 11. Reilly FD.102 In these cases.101. Gulbis B. Blood flow resistance indices are lower in the superficial and deep placenta compared with the cord insertion area. In: Knobil E. 12. Wood C. Jauniaux E. Jauniaux E. Robertson WB. Oxidative stress is a powerful inducer of endothelial cell apoptosis. Morphology and mechanisms of abortion. human evolution and oxygen. Schaaps JP.

6: 84–96. 35. Giffin F et al. 10: 947–50. early embryonic metabolism and free radical-mediated embryopathies. The First Twelve Weeks of Gestation: A New Frontier for Investigation and Intervention. 10: 2875–9. Doppler ultrasound features of the developing placental circulations: correlation with anatomic findings. Obstet Gynecol 1995. 23. Jauniaux E. Burton GJ. de Ziegler D. Blackburn ST. 24. Watson AL. Sibai BM. Heidelberg: Springer-Verlag. Jurkovic D. 1992: 78–96. Gonzales E. Jurkovic D. Kollen M. 164: 486–8. Walkinshaw SA. Mabie WC. eds. Jauniaux E. Das P. 90: 819–23. Johnson MR. Matijevic R. 55: 775–9. Jauniaux E. 50. DiSessa TG. Bower S. Jauniaux E. Relationship between the establishment of intervillous blood flow and the human chorionic gonadotropin early pregnancy peak. 77: 365–9. Jauniaux E. Early pregnancy changes in hemodynamic and volume homeostasis are consecutive adjustments triggered by a primary fall in systemic vascular tone. Marsal K. A longitudinal study of cardiac output in normal human pregnancy. 42. Am J Obstet Gynecol 1977. 44. Shunting the intervillous space: new concepts in human uteroplacental vascularization. oestradiol and progesterone on uteroplacental and corpus luteum blood flow in normal early pregnancy. Jauniaux E. 38. 87: 2954–9. Clapp J. In: Barnea E. Hempstock J. Am J Obstet Gynecol 1992. Hustin J. 41. Hum Reprod Update 2000. 29. Kurjak A. Campbell S. Brink GW. Buytaert P. Am J Obstet Gynecol 1987. 84: 338–42. 37. Am J Obstet Gynecol 1988. Dickey RP. Crocker LG. Fetal and Neonatal Physiology: A Clinical Perspective. 157: 2111–22. Hum Reprod 1992. 159: 1031–43. Jurkovic D et al. 36. 26. 86: 289–92. . Proc Natl Acad Sci USA 2005. Hempstock J. 31. Campbell S. Martin CB. Hum Reprod 1995. Relationship of uterine blood flow to chorionic sac and embryo growth rate. Meuris S. Campbell S. Jauniaux E. Investigation of placental circulations by color Doppler ultrasound. Vascular resistance of uterine arteries: physiological effects of estradiol and progesterone. Soderberg H. Resnik R. Meuris S. Jurkovic D. Kurjak A et al. 28. Jawerbaum A. Burton GJ. 1992. Am J Obstet Gynecol 1996. 102: 4783–8. Jauniaux E. 58: 49–54. Seaward BL. Bewley S. 45. 39. McGaughey HS. Jauniaux E. Skepper JN. 51. Menheere PA. Fluid compartments of the embryonic environment. 7: 114–21. Watson AL. Loquet P. 52. Am J Obstet Gynecol 1993. 174: 768–75. 43. Meekins JW. 162: 115–25. 170: 849–56. Ezashi T. Uteroplacental and luteal circulation in normal first trimester pregnancies: Doppler ultrasonographic and morphologic study. 34. Jauniaux E. Jurkovic D. Tsatsaris V. Burton GJ. Burton G. Hustin J. The role of alterations in arachidonic acid metabolism and nitric oxide homeostasis in rat models of diabetes during early pregnancy. Am J Obstet Gynecol 1994. Transvaginal color Doppler assessment of utero-placental circulation in early pregnancy. Pieters FA. 54. Reprod Biomed Online 2003. Campbell S. Am J Obstet Gynecol 1999. 53. Gulbis B. Oxygen.54 Fetal Cardiology 22. Am J Pathol 2003. 11: 1327–42. Maternal physiologic adaptation to early human pregnancy. Obstet Gynecol 1994. Intermittent functioning of the uteroplacental arteries. Jauniaux E. Delogne-Desnoek J. The effect of progesterone on estrogen-induced uterine blood flow. Kaiser IH et al. 32. Delogne-Desnoeck J. Doppler ultrasound investigations of pelvic circulation during the menstrual cycle and early pregnancy. Curr Pharm Des 2005. Hustin J. Hower JF et al. Schaaps JP. 169: 1382–92. Burton GJ. Skepper JN. Schaaps JP. Uterine glands provide histiotrophic nutrition for the human fetus during the first trimester of pregnancy. In vitro ischemia-reperfusion injury in term human placenta as a model for oxidative stress in pathological pregnancies. Maternal. Makikallio K. Plumer MH. Obstet Gynecol 1993. Jauniaux E. Hung TH. Onset of maternal arterial blood flow and placental oxidative stress. Philadelphia: Saunders. Echographic and anatomic studies of the maternotrophoblastic border during the first trimester of pregnancy. Burton GJ. Peeters LL. 47. Laurini R. 30. Jurkovic D. Valentin L. Obstet Gynecol 1991. Uteroplacental hemodynamics during early human pregnancy: a longitudinal study. Sladkevicius P. Am J Pathol 2001. 192: 323–32. 159: 1456–60. 184: 998–1003. Gynecol Obstet Invest 2004. 49. Jauniaux E. Trophoblastic oxidative stress in relation to temporal and regional differences in maternal placental blood flow in normal and abnormal early pregnancies. 40. Frydman R. Hum Reprod 1995. Hempstock J. Hower JF. Jauniaux E. 10: 2676–9. Evaluation of respiratory gases and acid-base gradients in fetal fluids and uteroplacental tissue between 7 and 16 weeks. Bessis R. Coppens M. Am J Obstet Gynecol 1991. 181: 718–24. 166: 585–7. Arheart KL. Cheriex EC. Gasser RF. Influence of human chorionic gonadotropin. Watson AL. Maternal arterial connections to the placental intervillous space during the first trimester of human pregnancy: the Boyd collection revisited. Ultrasound Obstet Gynecol 1996. Loper DL. 82: 78–83. 6: 268–78. De Neubourg F. 33. 128: 251–4. Jouppila P. Am J Obstet Gynecol 2005. Tekay A. 157: 162–8. Longitudinal evaluation of uteroplacental and umbilical blood flow changes in normal early pregnancy. Duvekot JJ. Jurkovic D. Improved prediction of preeclampsia by two-stage screening of uterine arteries using the early diastolic notch and color Doppler imaging. Hum Reprod 1995. The role of relaxin in the development of the uteroplacental circulation in early pregnancy. Jauniaux E. Greenwold N. Effect of ovulation induction on uterine blood flow and oestradiol and progesterone concentrations in early pregnancy. Hustin J. Spiral artery blood flow in the central and peripheral areas of the placental bed in the second trimester. Hempstock J et al. Hisser J. 7: 1467–73. 27. Watson AL. Fertil Steril 1991. 46. Roberts RM. Neilson JP. Am J Pathol 2000. Low O2 tensions and the prevention of differentiation of hES cells. Am J Obstet Gynecol 2001. Nagy AM. Am J Obstet Gynecol 1964. a possible factor in human early pregnancy failure. Sleamaker RH. J Clin Endocrinol Metab 2002. Dickey RP. 25. 48. McFadyen IR.

56. 83. Hustin J. 80. Leiser R et al. Demir R. Lab Anim Sci 1979. Jauniaux E. 3: 219–27. 181: 455–9.Placental circulations 55 55. Mayhew TM. . 18: 83–7. Placenta 1995. Maroto E. Obstet Gynecol 1997. 134: 221–38. II. Ultrasound Obstet Gynecol 1996. 164: 195–203. 9: 271–86. Kaufmann P. Ridgway LE. Fertil Steril 2003. Apoptosis in human chorionic villi and decidua during normal embryonic development and spontaneous abortion in the first trimester. 76: 851–64. Marquette G. Broughton-Pipkin F. Fetal vasculogenesis and angiogenesis in human placental villi. Ramsay B. Erbengi T. 78. Color Doppler imaging of the uteroplacental circulation in the first trimester: value in predicting pregnancy failure or complication. Changes during normal pregnancy. Teyssier G. Hemodynamic interpretation of the arterial Doppler waveform. Rudolph AM. 176: 695–705. Sly DL. 70. Shikone T. Color Doppler imaging and in vivo assessment of the anatomy and physiology of the early uteroplacental circulation. 38: 262–6. Circ Res 1976. Wladimiroff JW. eds. 91: 761–5. Ultrasound Obstet Gynecol 1993. Nakano R. 90. 25: 114–26. Abramowicz JS. Functional characteristics of small placental arteries. Jauniaux E. Burrell SJ. Fertil Steril 2003. Placenta 2004. Am J Obstet Gynecol 1994. 16: 323–31. Jaffe R. Doppler assessment of normal early fetal circulation. Woods JR. Fouron JC. Zaidi J. 61. Kaufmann P. Jauniaux E. Huisman TWA. 69. Am J Obstet Gynecol 1990. 76. Am J Obstet Gynecol 1994. Jurkovic D. 173: 203–14. The development of capillaries in the villi of early human placentas. Palmer AE. Wilmut I. placental karyotype and intervillous circulation in early pregnancy failure. Jurkovic D. Craven CM. Placenta 1998. Jauniaux E. Am J Obstet Gynecol 1991. 12: 269–76. Rice JM. Symonds EM. Sales DI. 85. 8: 1966–82. Nicolaides KH. Sources of variability in umbilical artery systolic/diastolic ratios: implication of the Poiseuille equation. 62. Mayhew TM. Hughes HC. 1992: 45–64. 136: 190–203. Jones CPJ. Jauniaux E. Lees C. 81. Hum Reprod 1994. Campbell S. Ultrasonographic investigation of placental morphology and size during the second trimester of pregnancy. Am J Obstet Gynecol 1999. 60. Jauniaux E. Kokawa K. 86. Jurkovic D. Hempstock J. Wright JW. Kupesic S. 89: 252–6. Kaufmann P. endocrinology. Nicolaides KH. I. AJR Am J Roentgenol 1995. Jauniaux E. Curing the human embryo . Jaffe R. Molecular recognition. McCarthy AL. The fetal vascularisation of term placental villi. Seidl DC. Dempsey EW. Kaufmann P. Moscoso GJ. Obstet Gynecol 1998. Ultrasound Obstet Gynecol 1997. True pregnancy toxaemia (preeclampsia) in the guinea pig (Cavia pocellus). Jaffe R.curing the placenta. 101: 1003–4. Syncytiotrophoblastic fragments in first-trimester decidual veins: evidence of placental perfusion by the maternal circulation early in pregnancy. 68. Diastolic circulatory dynamics in the presence of elevated placental resistance and retrograde diastolic flow in the umbilical artery: a Doppler echographic study in lambs. Hustin J. Evans BJ et al. Ward K. Placental lakes. Comparison of ultrasound and Doppler mapping of the intervillous circulation in normal and abnormal early pregnancies. Burton GJ. 164: 1255–8. 72. 8: 73–6. Charnock Jones DS. Berman W. Spontaneous preeclamptic toxaemia of pregnancy in the patas monkey (Erythrocebus patas). 67. Jauniaux E. 9: 2432–7. Dorgan A. 2: 300–5. In: Barnea E. Campbell S. Maternal relaxin: a determinant of fetal heart rate? Br J Obstet Gynaecol 1994. Woolfson RG. Burton GJ. 25: 103–13. Best Pract Res Clin Obstet Gynaecol 2004. In vivo investigation of the placental circulations by Doppler echography. Am J Anat 1972. 73. The First Twelve Weeks of Gestation: A New Frontier for Investigation and Intervention. Hum Reprod 1993. Jauniaux E. Campbell S. 29: 472–8. Ultrasound Obstet Gynecol 1992. 65. Bertolet R. Gulbis B et al. 74. 79. Kotowski A. Yu CKH. Carter AM. Development of the early human placenta: a morphometric study. 66. Lessard M. Jauniaux E. 58. The role of uterine Doppler in predicting adverse pregnancy outcome. Maternal and embryonic factors associated with prenatal loss in mammals. Greenwold N. Stewart PA. 163: 1788–91. Hustin J. 82. Ramsay B et al. Hustin J. 89. 64. Doppler assessment of the intervillous blood flow in normal and abnormal early pregnancy. II. Placenta 1991. Fertil Steril 1993. London WT. 84. Greenwold N. 170: 945–51. Pathology and clinical implications of abnormal umbilical artery Doppler waveforms. 170: 130–7. Goodlin RC. When is the maternal placental circulation established in Man? Placenta 1997. 2: 1252–3. Loquet P. 59. Bruns U. 71. Burton GJ. 75. Jauniaux E. Maternal circulation in the first trimester human placenta: myth or reality. 77. Jauniaux E. Am J Obstet Gynecol 1997. Ashworth CJ. 60: 293–7. Heyman MA. Rubin PC. Lancet 1988. Ultrasound Obstet Gynecol 1996. Johnson MR. Charnock Jones DS. 79: 100–6. Hustin J. Campbell S. Lab Anim Sci 1979. Kurjak A. 18: 383–96. 29: 102–6. 57. Blood velocity waveforms and placental vascular formation. Anat Embryol 1985. Kaufmann P. Placenta 2004. 79: 1373–9. Castellucci M. Kingdom JCP. Comparison of color Doppler features and pathologic findings in complicated early pregnancy. Relationships between maternal serum. Maulik D. absent umbilical artery diastolic flow and poor fetal growth in early pregnancy. Relationship between pressure and flow in the umbilical and uterine circulations of the sheep. 7: 141–4. 19: 21–6. Aspects of human fetoplacental vasculogenesis and angiogenesis. 63. 88. The relationship between nitric oxide placental production and umbilical artery vascular impedance in early pregnancy. Jauniaux E. Acta Anat 1989. Lang CM. Early human placental morphology. Intermediate and terminal villi. Aspects of human fetoplacental vasculogenesis and angiogenesis. 87. J Reprod Fertil 1986. Papageorghiou AT. Intervillous circulation in the first trimester: the phantom of the color Doppler obstetric opera. Heidelberg: Springer-Verlag. Jauniaux E.

Weiner Z. 58: 1–7. Pretorius DH. 73: 220–4. Bilardo CM. Daniel-Spiegel E. 14: 262–6. 103. Doppler ultrasound for fetal assessment in high risk pregnancies. Weiner CP. 12: 45–9. 99. placental and yolk sac haemodynamics in preeclampsia and preterm labour. 19: 729–33. Shen O et al. Neilson JP. Reverse end-diastolic umbilical artery velocity in a case of intrauterine fetal death at 14 weeks’ gestation. Ariyuki Y. Oxygen and placental villous development: origins of fetal hypoxia. Am J Obstet Gynecol 1993. Imaging of placental vasculature using three-dimensional ultrasound and color power Doppler: a preliminary study. Neilson JP. Marsal K. Khun P et al. 105. Burton GJ. Jauniaux E. Placental morphometry and Doppler flow velocimetry in cases of chronic human fetal hypoxia. Nelson TR. 95. 93. Weiner E. Shen O et al. 94. Gudmundsson S. Cord blood gases and absence of end-diastolic blood velocities in the umbilical artery. Anteby EY. Am J Obstet Gynecol 1990. 106. Am J Obstet Gynecol 1990. Shalev E. Fetal heart rate and umbilicoplacental doppler flow velocity waveforms in early pregnancies with a chromosomal abnormality and/ or increased nuchal translucency thickness. Hum Reprod 1996. Kingdom JCP. Baergen RN et al. First trimester uterine. 11: 435–9. Kirkinen P. 92. J Matern Fetal Neonatal Med 2005. Ultrasound Obstet Gynecol 1999. 104. Placenta 1997. . Huch A. Bricker L. Umbilicalplacental blood flow gradient during the early second trimester of pregnancy. Cochrane Database Syst Rev 2000. Ultrasound Obstet Gynecol 1999. Bracero LA. 97. Early Hum Dev 1990. Nicolaides KH. Byrne DW. Routine doppler ultrasound in pregnancy. Cochrane Database Syst Rev 2000. Anteby EY. Kaufmann P. Gynecol Obstet Invest 2004. Kitao M. Jouppila P.56 Fetal Cardiology 91. 120: 139–45. Simultaneous multigate spectral Doppler imaging of the umbilical artery and placental vessels: novel ultrasound technology. 102. Acta Obstet Gynecol Scand 1994. Doppler measurements of fetal and uteroplacental circulations: relationship with umbilical venous blood gases measured at cordocentesis. Makikallio K. (2): CD001450. 98. Kuzmina IY. Lindblad A. The relationship between the umbilical artery systolic/diastolic ratio and umbilical blood gas measurements in specimens obtained by cordocentesis. Huch R. Haberman S. Campbell S. 18: 613–21. Spectral Doppler index mapping of the umbilicoplacental circulation and pregnancy outcome. Placental blood flow measured by similtaneous multigate spectral Doppler imaging in pregnancies complicated by placental vascular abnormalities. 100. Alfirevic Z. 14: 256–61. Hum Reprod 2004. 24: 231–7. Eur J Obstet Gynecol Reprod Biol 2005. Yagel S. 169: 1621–2. Blood flow velocities in human intraplacental arteries. Ultrasound Obstet Gynecol 1998. 17: 133–6. Humina-Vakulic GJ. (2): CD000073. Yagel S. 162: 1198–202. 101. 162: 115–20. Tekay A. Hata T. Gavriil P. 96. Kurmanavichius J.

the net resultant pressure change in that region will be less in magnitude than any of the individual pressures. Constructive interference If the high-pressure regions of two waves coincide.2. we shall describe how images are formed by state-of-the-art ultrasound systems. This is called constructive interference. and subsequently show how the image can be optimized using the system controls. lasting about 1 microsecond (one millionth of a second) is transmitted into the body. the resultant sound pressure in a given region is determined by the net inflow of molecules. By ‘focused’ we mean that the pulse is somehow made to travel along a very thin beam. When two or more sources send sound waves across the same volume in the tissue. as well as the system controls that could be used to affect them. Destructive interference If the high-pressure region from one wave coincides with the low-pressure region from another wave traveling across the same region. The wavelength is the distance between two adjacent points along the beams having the same phase. The pulse is ‘focused’ and portions of it are reflected back toward the transducer. Table 6. Beams are formed utilizing the physical phenomenon of interference. The molecules leaving it generate a low-pressure region. the sound pressure at the region where the two waves cross will be higher in magnitude than that of each of the individual waves. Take a snapshot of the wave at any instant. distance between two successive minimums or two successive maximums.1 summarizes the basic concepts that will be used in this chapter to define a ‘good’ image. Molecules crowding into a region in the tissue will generate a high-pressure zone. e. We may also distinguish between destructive and constructive interference. The point in the body where the beam is thinnest will be referred to as the ‘focal point’ and the total length where the beam is still ‘thin’ is called the ‘focal depth of field’.6 The physics of ultrasound imaging Zvi Friedman Introduction The purpose of this chapter is to review the physics of ultrasound imaging and the various parameters that define image quality. Generation of the ultrasonic image In the following. If we assume that we know the velocity of propagation of sound in the tissue. It is therefore determined by the algebraic sum of the individual pressures.and lowpressure regions. a succession of high. . we can assign the received echo at a specific time to a definite location within the body.g. These waves. The quality of the image will be affected by the beam parameters detailed in Table 6. The wavelength is defined as follows. Propagation of sound in the body Interference Sound propagates in the body in the form of pressure waves. We shall later in this chapter see how these depend upon the various system parameters. travel through the tissue at the speed of sound – about 1500 m/s. Image quality Confident and reliable ultrasonographic diagnosis depends on the ability to consistently produce excellent images. A short pulse of ultrahigh frequency sound (between 2 and 20 MHz). This leads to the physical concept of interference.

Specular (mirror-like) reflection occurs when the reflector is a large extensive (relative to the wavelength – 0. as will be elaborated below. the surface of the placenta. An ultrasonic pulse will not persist.) The situation is illustrated in Figure 6. thus defining the frequency range in the signal. a 3. (this is the typical pulse length used in diagnostic ultrasound).44 mm at 3. enough in time. however. Intensity is lost due to absorption. and specular reflection. amniotic fluid or simple cysts) should appear as totally ‘black’ Spurious echoes resulting from a variety of artifacts might cause these areas to appear as echogenic and lead to misdiagnosis Uniformity of all the above throughout the entire image Table 6.or twodimensional transducer arrays.or two-dimensional transducer arrays. The receive focus is the ‘microscope’ with which we examine the lighted area. In order to measure the frequency. scattering.3. fetal lungs should be easily distinguishable from fetal liver) Non-echogenic objects in the image (e. As can be intuitively understood. we measure time differences between all adjacent peaks (maximums or minimums). for example.g. beam forming is accomplished electronically. Beam forming allows us to better visualize minute anatomical details in the patient. Examples of such a surface could be the boundaries of a kidney. we shall find a periodic temporal behavior.5 MHz.5 MHz) surface. two-dimensional arrays can provide much narrower beams. To get a rough idea of the actual frequency content of the signal.2 quality Parameter Beam parameters that affect image Explanation The width of the beam along which the transmitted beam travels The width of the beam along which the received (reflected) beam travels The length (time duration) of the ultrasonic beam transmitted into the body The length (time duration) of the electric pulse generated by the transducer when the ultrasonic beam impinges on it Approximately 1540 meters per second The fact that this velocity is not the same for all tissues is one of the reasons for degradation in image quality Beam width – transmit Beam width – receive Contrast resolution Pulse length – transmit Low noise Pulse length – receive Velocity of sound in the tissue Consistency Beam forming The beam forming in modern ultrasound systems is accomplished using one. Frequency and bandwidth When measuring the pressure at any given point in the body. the wave frequency is f = 1/T MHz (millions of cycles per second). the skin of a fetus.5–3.1 Parameters that define image quality Concept Detail resolution Sensitivity Explanation The ability to visualize minute anatomic detail The ability to adequately visualize minute anatomic detail deep in the patient The ability to clearly and easily differentiate between tissue types (e. The three mechanisms are summarized in Table 6. Thus. Each measurement will result in a slightly different value for the frequency.5-MHz signal lasting 1 microsecond (about three cycles) will contain signals in the frequency range 2. Specular reflection and scattering Only two out of three mechanisms listed in Table 6.3 will eventually result in an ultrasound wave getting back to the transducer: specular reflection and scattering. at any given point longer than 1 microsecond. With one. we simply measure the time difference between two adjacent maximums (or minimums). etc. (The frequency range is inversely proportional to the time duration of the signal. the diaphragm.g. The longer the signal lasts. The above description is quite crude. if the time difference between two adjacent maximums is T microseconds.1. The main reason is that for a time signal to be described as a single frequency signal (a ‘pure’ sine wave). it must be long Attenuation As the signal travels across the body its intensity is continuously decreased. The number of pressure maximums (or minimums) per second at any given location within the body is defined as the frequency of the sound wave.58 Fetal Cardiology Table 6. Specular reflections are . Thus. The ‘transmit beam’ is our ‘searchlight’. the ‘purer’ is its frequency.

the scatter will be nine times more intense. The larger is the reflector.1 Pressure wave generated by a 3. Thus. following an exponential decay law The distance at which the beam intensity is reduced by half is inversely proportional to the frequency Scattering is angle independent Scattered intensity is proportional to the 4th power of the frequency Specular reflection is not frequency dependent Scattering Part of the beam is scattered (in all directions) by small microscopic particles (smaller than the wavelength) Part of the beam is reflected at the surfaces interfacing between two regions that have different acoustic properties Specular reflection Specular (mirror-like) reflection is strongly angle dependent . the absorption coefficient is proportional to the wave frequency.3 Mechanism Absorption The three mechanisms causing sound attenuation in the body Description Kinetic energy of particles moving into high-pressure regions and out of low-pressure regions is converted into heat due to viscous friction Angle dependence Absorption is angle independent Frequency dependence The intensity of the beam is decreased with depth. Unlike specular reflection and scatter.44 mm at 3.4 Pressure [MPa] 0 –0. This means that if the parenchymal architecture that gives rise to echoes in one organ is three times larger than that in another. it does not result in a wave that gets back to the transducer.5 MHz) or smaller. part of the beam is reflected. Absorption Absorption is a significant factor in attenuation. 0.8 0 0.8 t [μs] 1. It occurs when the ultrasonic wave is reflected from particles that are of the size of one wavelength (0. it represents a series of processes through which the energy in the ultrasonic wave is eventually converted into heat.5-MHz probe.The physics of ultrasound imaging 59 highly directional and will occur only if the reflecting surface is precisely perpendicular to the direction of the beam. The intensity of the reflection will depend upon the size of the reflector relative to the wavelength. Rather.4 Acoustic impedance –0. the acoustic properties affecting reflection can be related to a single parameter in the tissue. The amplitude of the reflected beam depends only upon the properties of the tissues on both sides of the surface. The amplitude of the reflected wave will be large for tissues with very different properties and will be very direction dependent.0-MHz wave will lose 50% of its original intensity at a depth that is about half the depth for which a 3. sometimes Table 6. In the frequency range usually used in diagnostic ultrasound. doubling the frequency of the ultrasound (halving the wavelength) increases the relative echo strength by a factor of 4. for example. It has been shown experimentally that the intensity of the reflection is roughly proportional to the square of the ratio between the particle size and the wavelength.5-MHz wave loses 50% of its original intensity.6 2 Figure 6.2 1.4 0. a 7. As a matter of fact. As stated above. the stronger is the reflection. Scattering is generated by parenchymal structures. when an ultrasonic beam meets an interface between two types of tissues. Similarly. The proportion of the sound that is reflected depends on the acoustic properties of the two tissues.

This parameter is a function of the density of the tissue and its stiffness.4 Acoustic enhancement posterior to a cystic mass. Figure 6. This is called posterior enhancement.2 showing a transthoracic image of a fetus.60 Fetal Cardiology referred to as the acoustic impedance of the tissue. Figure 6. The shadows cast by the fetal ribs are clearly seen. an ultrasound image would be one for which the gray-level at a given point on the screen represents unambiguously the reflection coefficient of the corresponding point in the tissue. i. The difference between the two situations is illustrated in Figures 6.5 Acoustic shadowing by a fatty mass. no absorption and no scattering. In reality this is almost never the case due to several artifacts. Because the beams passing through the cyst are less attenuated than the rest of the beams that pass through tissue with higher attenuation.e. and is sometimes used to differentiate between fat (highly absorbing) and liquid (nonabsorbing). Figure 6. Note the lack of posterior enhancement. as can be seen in Figure 6. Two of these sonographic artifacts will be discussed here while others will be discussed later. The reverse phenomenon will occur when the sound traverses through a medium with relatively small attenuation.3.3 Acoustic shadowing formed by highly absorbing tissue.4 and 6. as in the example in Figure 6.5. an acoustic shadow will be cast. Note the acoustic shadowing formed by the ribs.2 Fetal chest. Figure 6. A similar situation will occur when the attenuation is caused by absorption rather than by reflection. Shadowing and enhancement Ideally. When an echo is so large that the sound continuing into the tissue is highly attenuated. . the area behind will be brighter. Typical examples could be a cyst or a pocket of amniotic fluid.

for example. can also affect the speckle pattern. Modern well-designed systems are sampled densely enough so that sampling will have minimal effects on the image. and is due to the fact that two targets along the beam direction will be separable only if the echo from the first target dies off before echo from the second target starts arriving. The observer performs in his brain an integration process by which all signals resulting from echoes from the target area are summed together. image contrast results from small differences in the acoustic properties of tissue. Thus. Thus. cause interference to occur between different parts of the wave received from the object region that correspond to a given point in the image. When evaluating equipment.g. Speckles One of the essential tasks of diagnostic ultrasound imaging is the detection of focal lesions of low contrast against background tissue. The target lesion will be discriminated from the background if: (1) the number of individual speckles within the target area is large enough (i. Thus. for two equal targets along the beam the resolution will be relatively high. the transducer aperture. the frequency that can be used for imaging at a depth of 100 mm is half the frequency that could have been used for imaging at a depth of 50 mm. and reflectivity of tissue structures within the scan plan. This noise ‘camouflages’ the target. it hignores the effects of the attenuation. circular) depends on the ‘contrast’ between the target and the background. for example. The lateral resolution for a given sound frequency and a given transducer aperture is inversely proportional to the distance. which are on a scale too small to be observed by the imaging system (smaller than the imaging wavelength). Statement 1 can also be phrased mathematically. (2) the average brightness of individual speckles in the target area is different enough from the average brightness of individual speckles in the background area. The ultrasound system translates these minute differences into low amplitude shades of gray. Diagnostically important . Both axial and lateral resolutions depend on the relative intensity (reflection coefficient) of the targets. carried out by the eye–brain system of the observer. concentration. It has been called speckle because of its similarity to the equivalent optical phenomenon. the distance from the transducer.e. on the other hand. This noise is superimposed on the image and appears in the form of gray ‘stains’ of random sizes and random intensities. and is related to the fact that higher frequencies are attenuated faster than lower frequencies. In order to accomplish this objective. A simplified definition of system resolution is the separation between two point test targets that are just resolvable (can be seen in the image as two distinct points). The phenomenon of speckles results directly from the use of coherent radiation for imaging. The image sampling. In most state-of-the-art ultrasound systems the axial resolution is much smaller than the lateral resolution. The main ‘noise’ factor in diagnostic ultrasound imaging is the so-called ‘speckle noise’. if the resolution at a distance of 50 mm is 1 mm. However. i. The lateral resolution on the other hand is determined by the beam width. In actuality the situation will be even worse. it will be degraded to 2 mm at a distance of 100 mm. one should be very careful in examining and comparing resolution numbers quoted by the manufacturers. The degradation of the axial resolution with the depth is a result of attenuation in the tissue. due to the dependence of the attenuation on the depth. the difference in Contrast resolution Contrast resolution is the ability to perceive tissue contrast differences in gray-scale images. where m is the average number of speckles in the target area. In a high resolution image small structures can easily be seen. The axial resolution is limited by the pulse length. the average speckle size is much smaller than a typical target dimension). It is mainly determined by the spatial distribution.The physics of ultrasound imaging 61 Resolution Spatial (detail) resolution The concept of resolution relates to the amount of detail that can be perceived in an image. The speckle intensity will depend on the average number of scatterers per unit volume in the tissue. the number of beams across the field of view that are used to generate the image (image line density). Most modern systems attempt to provide a uniform resolution throughout the entire image. one needs to wait until the echoes from the first target are less than one-thousandth of their peak value. It occurs when structures in the object (tissue). This partly takes care of the dependence of the lateral resolution on the depth. laser speckle. the first target is 1000 times stronger than the second target (a situation that very often occurs in diagnostic ultrasound). as well as the number of samples along the beam. Both axial and lateral resolution degrade with imaging depth. If. sometimes referred to as subject contrast. upon the target size. These numbers must also state the relative intensities of the targets in order to have any meaning at all. the transducer aperture is increased with the depth of imaging. and the pulse length. mainly by ‘breaking’ its contours. since echoes arriving from the second target will be separable from those arriving from the first target after the latter are decreased by a factor of approximately 2. and upon the background ‘noise’. The average speckle size mainly depends on the properties of the ultrasonic beam: the ultrasonic frequency. stating that the signal to noise is proportional to m1/2. The detection of a lesion within background tissue is a psychophysical process. The capability to detect a target of any given shape (e.e.

Since the transducer material has acoustic properties that are very different from those of the patient tissue. however. It will. sampling at a rate which is not high enough. The object is imaged from various aspects. this results in reducing the interference. change abruptly. which is relatively large. the distance between two samples must be much smaller than the average speckle size or the average distance between two speckles in the image. it would be desirable to make the speckles ‘finer’ (decrease their average size). part of it will be reflected. The speckle intensity can be reduced by image compounding. The reflected wave will then hit the face of the transducer. the larger is the change the more disturbing is the effect. In one commercial implementation the object is viewed from eight directions. Another important factor will be the gain setting of the system. Speckles are caused by the interference of echoes from densely packed tiny parenchymal reflecting structures that are closer than the basic system resolution. there is no one-to-one correspondence to any identifiable structure in the body. The amount of the reflected wave will depend upon the intensity of the incident wave. Small movements of the probe will cause continuous small movements of such structures. A large portion of the wave will bounce Figure 6. The amount of change will depend first on the ratio between the sampling rate and the natural resolution of the system.e. Under-sampling and aliasing As previously stated. . the sampling rate (in both axial and lateral directions) must be high enough to comply with the speckle density. The first objective can be reached by increasing the frequency bandwidth of the ultrasonic beam. giving the effect of a moving background noise. Of course. only part of the wave will enter the transducer. also depend on the difference in the acoustic properties of the two tissues. Thus. This allows us to use the speckle information to discriminate between tissues. as well as make them less intense. i. despite the fact that the echo pattern is random in nature. A similar situation occurs in acoustics. as well as on the properties of the beam.6 The effect of image compounding. where the effect will be most annoying at high gain. on the image is called aliasing. the appearance of the speckles is fairly constant for each type of tissue and differs between tissues. The speckle pattern will. Multiple reflections at the opposing mirrors will probably make such an attempt useless. The displayed image is then an average of the images from the various aspects. and the uniformity of their distribution. i.6.62 Fetal Cardiology acoustic properties between the scatterers and the tissue in which they are immersed. Now try to estimate your distance from one of the mirrors by just looking at it. However. since the amplitude and distribution of the speckles are related not only to the ultrasonic beam properties but also to the parenchymal tissue architecture. when the sound beam is perpendicular to an interface. One of the most severe effects of ‘under-sampling’. their size. As discussed above.e. Obviously. Reverberations Consider yourself in a room that has mirrors on two opposite walls. Images taken in compound mode (a) and in non-compound mode (b). the various directions must be far enough apart in order for the speckles to ‘average out’. Speckle reduction Because of the undesirable effect of speckles on the capability to distinguish lesions from the surrounding tissue. The effect of this implementation is illustrated in Figure 6. however.

the same object will be registered in multiple positions in the image. and since the system registers the axial location of an object according to the time of a single round trip. and (2) make the beam as narrow as possible. however. sensitivity. Focusing To generate a high resolution image. pleasant image. Fixed lens focusing The signal-processing chain The signal-processing chain is schematically described in Figure 6. It ends in the eye–brain system of the of the sonographer.7. the origin of each echo must be accurately determined. contrast resolution. such as single-element mechanical sector scanners. The signal-processing functions include: improving the lateral resolution by focusing techniques. adjusting the amplitude of the echoes further to emphasize particular features of interest. In this section the entire system signal-processing chain will described in detail and the effects of the various blocks and system features on the diagnostic image quality will be elaborated.The physics of ultrasound imaging 63 back into the patient. The machine must. Since the wave from the edge of the transducer Figure 6. by the interaction of the ultrasound beam with the tissue. one needs to (1) use high frequency. Since the time for such a multiple round trip between the transducer and the patient is a multiple of the time for a single round trip. used a fixed lens to focus the beam. as shown in Figure 6. very short pulses. usually shaped spherically. and consistency. inserting the results in the right places to generate a smooth. Focusing is accomplished by attaching a lens – a piece of plastic material. For this to be achieved. In order to understand the operation and performance of these beam-forming techniques. estimating the amplitudes of the missing echoes. Modern scanners employ multielement electronic beamforming and focusing. correcting for the attenuation (in order to make the image brightness uniform). let us first review the ‘fixed lens‘ focusing method. . Image formation in premium high resolution diagnostic ultrasound systems Introduction Premium high resolution ultrasound systems have become increasingly sophisticated and include many new signalprocessing features. and rejecting misleading information. It will travel to the reflecting surface and back to the transducer. noise. The signal processing chain begins with the scanned crosssection in the patient.8 – in which the sound wave travels more slowly than in the tissue. do a larger amount of electronic signal-processing before the echo information it receives can be displayed as a clinically useful diagnostic image. the features of which are determined Older technology. The main factors affecting diagnostic image quality have already been identified to be detail resolution.7 Signal-processing chain.

In order to improve image resolution in other areas. a spherical wavefront is produced. Focusing (lens) r2 c2 r1 c1 r0 c0 + = c2 r2 c1 r1 r0 Figure 6. Areas close to the focal point portray good image quality. in some systems.9 To focus the beam. On transmit The pulse is fired from each of the elements separately. similar to that produced by the fixed lens. Other areas are out of focus and have much lower image resolution. Combining the segments around each of the focal points then generates the entire image. one to achieve clinically acceptable frame rates with more than five transmit focal zones. Confocal imaging on transmit With an electronic transducer array the focusing point is determined by the set of delays between the pulses. The effect of curving the wavefront is accomplished by dividing the transducer into a large number of very small transducers (elements).8 Focusing with a fixed focus lens. only a segment close to the focal point is used. the pulses are first applied to the outside elements and then later to the elements that are situated nearer the center of the array. For each beam. To generate an ultrasonic line. Focusing with an array transducer The same principle is used in array transducers. one should allow more time for all the echoes from a pulse to arrive before transmitting the next pulse. The concept is described schematically in Figure 6. the wavefront will bend and become spherical. The edge of the array is fired first and the central element is fired last. Total delay First 1 2 3 4 5 6 7 Last 8 9 10 11 12 13 14 15 First 16 Wave front Figure 6.9. The fact that many beams need to be transmitted in order to generate a single line in the image severely reduces the frame rate. more focal points are used. This sort of technique allows. By properly adjusting the time delays between the elements. This extra time can be saved if consecutive beams are transmitted along separated lines.64 Fetal Cardiology starts traveling in the tissue before the wave from the center of the transducer. The circular wave fronts spreading from the outer elements propagated a greater distance because they were fired earlier. Each element is a center of a circular wave. several consecutive beams are fired. and since the speed of sound in the lens is slower than that in the transducer. The varying distances of propagation result in crossing points (where constructive interference occurs) lying along a circular curve. This results in a beam with a focal point at which lateral resolution is best. since in order to avoid mixing of reflections of the present beam from the near zone with reflections of the previous beam by strong reflectors deep in the body. whereas the rest of the beam is ignored. Each of the beams is focused at a different point. . Sometimes the frame rate is further reduced.

which is the plane perpendicular to the transducer surface and which cuts the center points of all the elements. Table 6. This allows us to focus the beam at some distance from the array. Side lobes and grating lobes One of the more serious problems in multielement array transducers is related to image degrading artifacts referred to as side lobes and grating lobes. Their intensity and angle from the main beam depends on the center to center spacing (pitch) of the elements. As side lobes are mainly an edge effect. Let us consider two examples. One simply has to delay the output of each of the elements properly and then sum them together. The contribution of this source to the side lobes will therefore decrease as we increase the number of elements and decrease their size. high-performance systems should not produce noticeable grating lobes.The physics of ultrasound imaging 65 Dynamic focusing on receive Focusing a transducer array on receive is very similar to focusing it on transmit. face to generate the so-called main beam. If the beam is too wide. since side lobes are caused by interference. These secondary. where the contrast is even further reduced due to the width of the beam. whereas grating lobes appear at a relatively large distance from the main lobe. As can be seen from Table 6. Spatial variation in side lobe intensity occurs because of interference between sound energies arising from opposite sides of the transducer. one of the portions of the beam may be interacting with the fluid (non-echoic) structure while another portion of the beam interacts with adjacent soft tissue. Generally. the beam is relatively wide in this direction. Modern. A second source of side lobes in array transducers comes from the discontinuous nature of the aperture. At the summation point. side lobes will appear close to the main beam. and upon the size and curvature of the array. but keep total aperture fixed (by increasing number of elements) Effect on side lobe intensity Decrease Decrease Decrease Side lobes Although the majority of the sound produced by a simple transducer propagates directly away from the transducer . Since. This allows excellent focusing of the beam in the plane of scanning.5. The effect of each of the above parameters on the grating lobe intensity and position is summarized in Table 6. which we now approximate with a set of linear segments. A second example is when one desires to detect a low contrast lesion within a tissue. and since the focus is fixed. Out-of-plane focusing and beam-width artifacts Most array transducers available in the market today employ a single row of 128–192 very small elements. The results of a wide beam in the out-of-plane may be painful.4 Effect of various array parameters on side lobe intensity Parameter change Increase total array aperture Increase sound frequency Decrease element size. their intensity relative to the intensity of the main lobe will decrease with increasing size of the transducer (the size is most conveniently expressed in units of the wavelength).5. due to practical limitations. Grating lobes Grating lobes are caused by the constructive interference of laterally directed energy from edges of the individual array elements. A simple cyst appears black (non-echoic) since it does not absorb and does not scatter. resulting in destructive interference. It is well known that diagnosing an adnexal mass as a simple cyst may be crucial to determine its benign nature. the transducer aperture in this direction is rather limited (typically 10–15 mm). Interference can be either constructive or destructive. Side lobes mainly disturb the possibility to see a weak target next to a strong target. The beam width in the perpendicular direction is focused with a cylindrical lens along the length of the array. The effect of each of the above parameters on the side lobe intensity is summarized in Table 6. since the side lobe intensity of the strong target might be larger than the echo of the second target.4. side lobes are produced because at the lateral margins of the sound source a portion of the sound energy is transmitted radially away from the beam axis. a small portion of the energy will be concentrated outside the main central beam. one can easily optimize the image by dynamically varying the delays as function of time so that all echo sources are focused (summed in phase to generate a constructive interference). upon the wavelength (or sound frequency). echoes from all other points will not be in phase. Owing to the fact that echoes from deeper points in the patient arrive later at the surface of the transducer. The approximation will naturally improve as we decrease the length of each of the segments and increase their number. on the spacing between edges of adjacent elements. The ideal shape of the aperture is a circular curve. This could create echoes registered within the cystic structure. their intensity will decrease if the pulse length is decreased. The first example comes from gynecology. Also.

In this case. the reduction in intensity is exponential. Dynamic range: gray-scale The acoustic transducers used in diagnostic ultrasound are capable of recording echoes over a range of pressures in excess of 100 000:1 (100 dB). since time is equivalent to depth. . The concept of dynamic range is most easily explained using Figure 6. several steps are usually taken in order to reduce it with minimal diagnostic penalty. This will be most effectively accomplished by noting that the huge dynamic range of 100 000:1 is mainly a result of two effects. Presently available state-ofthe-art monitors are capable of displaying a range of no more than approximately 32:1 (30 dB).10. however. amniotic fluid. The ultrasonic beam will traverse through fat. The range of each of the effects is approximately 300:1. In order to compensate for this type of attenuation. all ultrasound signals start from the zero level and end at the system saturation level. which is negligible at close range but is very high deep in the image. In most modern systems. but keep total aperture fixed (by increasing number of elements) Increase probe radius of curvature 50 100 150 200 In order to extract all the clinically relevant information contained in the image and display it to the diagnostician. a multiple gain adjustment with some 8–12 sliding potentiometers is implemented. The brightest shade (white) represents the value 64 and the darkest shade (black) represents the value 1. The ratio between the saturation and reject level is defined as the system dynamic range. increasing the imaging frequency for a given transducer will produce grating lobes. An ideal ultrasound image is a two-dimensional map of reflection coefficients regardless of depth. whereas the second effect is a result of the differences in reflection coefficients. Generally. The attenuation is extremely heterogeneous and can by no means be represented using a single parameter.11. Since the attenuation coefficient may be patient dependent. This task is achieved using a system function called time gain compensation (TGC). When sound is attenuated by a large homogeneous object such as the liver. the information is displayed in the form of gray-levels as shown in Figure 6. The lower portion of the signal lies in the range within the noise level and is therefore obscured by it. fetal bones and tissues.11.66 Fetal Cardiology Table 6.10 Shades of gray representing a dynamic range of 64:1. muscles (e. to be described below. As can be seen from Figure 6. one could fairly well compensate for this type of attenuation by adjusting a single parameter. far more complicated.g. the acoustic data must be electronically manipulated so that all the clinically relevant information is faithfully displayed on the limited range monitor. The system reject level is set just above the noise level so that only the echoes above the reject level are displayed. It will be convenient to base the operation of the system on the concept of dynamic range. The image is then optimized by trial and error. Figure 6. one would need to increase the gain logarithmically with time. The first effect is the result of tissue attenuation. The situation in obstetrics is. Our first task would therefore be to minimize the effects of attenuation by the tissue. Since the system cannot display this whole range of information. placenta.5 Effect of various array parameters on grating lobe intensity Parameter change Increase total array aperture Increase sound frequency Effect on grating lobe intensity and position Increase Increase: grating lobes will appear closer to the main lobe Decrease: grating lobes will be shifted far away from the main lobe Decrease Decrease element size. 250 50 100 150 200 250 Time gain compensation The effect of attenuation by the tissue can be at least partly compensated for by progressively increasing the gain of the system as echoes come deeper from the body. etc. the dynamic range of the echoes exceeds 100 dB (a ratio of 100 000:1). from the maternal abdominal wall).

13. For the second application. without any . Some of the low-level signals fall in the region of the background noise and will therefore be obscured.12 Logarithmic compression of the dynamic range.e.13. when imaged at such a setting. which eliminates both system noise and the low-intensity echoes that lie just above the noise level. Image appearance that is best suited to emphasize outlines of objects is different from the image appearance that would maximize the contrast between similar tissues. All ultrasonic signals begin at a zero signal level and can increase in amplitude until they reach the system ‘saturation level’. so that the fact that they are saturated (appear as white on the screen. On the other hand.g. All modern systems have a built-in system-reject. Most ultrasound systems use some kind of logarithmic function for compression of the dynamic range. an additional post-processing function is usually applied to the resultant curve. Noise 300 250 200 Output level 150 100 50 0 Post-processing In addition to the logarithmic compression. as described in Figure 6. large. the dynamic range needs to be further compressed until a displayable range of 30:1 is reached. the information required could be extracted even if the 30 shades of gray provided by the display are equally divided between the 300 input levels. as they mainly result from scattering from tiny scatterers. The selection of the post-processing function is crucial to obtaining the desired appearance of the image. the tissue part of the image will appear ‘grainy’. Echoes from such tissue structures are very weak to begin with. is that the gain for low amplitude signals should be relatively high. On the other hand. The dynamic range of the system is the ratio between the system saturation level and the system reject level. high contrast speckles. This function is application specific. Zero level Noise Signal dynamic range Compression After having reduced the original 100 000:1 input dynamic range by the TGC to a value of about 300:1. The first example is characterized by large echo differences (e. Examples of such functions are shown in Figure 6. between the heart chambers and the myocardium).12.11 The dynamic range of a representative ultrasound system. we are less interested in the highly echogenic structures. so that similar tissue structures with only slight differences in echo signal would still look different on the screen. the requirement is to differentiate between tissue structures with only slight difference in structure. 0 1 2 3 4 5 6 7 8 9 10 Input level (arbitrary units) x 105 Figure 6.The physics of ultrasound imaging 67 Saturation level Displayed brightness Signal dynamic range Figure 6. Different image appearances are optimally tuned for different diagnostic requirements and applications. as described in Figure 6. i. The requirement. In this case. therefore.

and S-shaped. This distortion is a manifestation of the deviation of the wave propagation in tissue from linearity. weights are allowed to change automatically according to the rate of change of image data between successive frames. is low enough so as not to interfere with the information that needs to be displayed. linear.11). A simplistic way of describing the distortion is given below. even very slight movements of the probe could change the speckle pattern in tissue texture echoes. Ultrasound propagates in a fluid or soft tissue as longitudinal waves of alternate compressions and rarefactions. side lobes. During rapid search scanning or when imaging very fast moving objects such as heart valves. This second application presents a challenge. Recent developments Tissue harmonic imaging Reverberations. The gain for these levels should thus be kept very low. The overall gain curve will therefore have a shape of a logarithmic S-curve as shown in Figure 6. For high enough acoustic pressure.g. Reverberations. since blood is a very weak reflector. Tissue harmonic imaging is based on the fact that an ultrasonic wave propagates in a fluid (or tissue). . it is distorted. but will cause much slighter changes in appearance in the larger outlines. showing the same image taken at four different settings of the dynamic range. grating lobes. high dynamic range system.68 Fetal Cardiology capability to resolve any structures in them) is less important. side lobes. with the amount of distortion continuously increasing as the wave propagates deeper into the tissue. and calls for a high quality.13 Examples of three possible postprocessing functions: logarithmic.13. e.14. and grating lobes will generally appear as image noise and clutter that will obscure the boundary between blood and tissue. Weights are also made to vary between different regions in the image. this temporal averaging is undesirable. and grating lobes are very obvious and disturbing in obstetrics and in echocardiography. the averaging is an iterative process in which the new pixel (picture element) information is averaged with the old information. This might have an effect on the image ‘cosmetics’ but will not affect the diagnosis. The chambers should be presented as black. user selectable weights. In these two applications. In most common implementations. etc. Averaging of successive frames will therefore smooth the speckles but will hardly affect important outline information. In some implementations. and if the acoustic amplitude is sufficiently high. however. with White Brightness Black S Shaped Linear Logarithmic Signal amplitude Figure 6. in which the noise level due to image artifacts such as side lobes. reverberations. All these are substantially reduced with tissue harmonic imaging. Note. one desires to accurately outline borders between tissue and large fluid compartments. the compressional Persistence As seen above. The effect of the dynamic range control is illustrated in Figure 6. This calls for an implementation that would allow full control of the degree of averaging. that we still do not wish to emphasize the extremely low signals that usually represent noise. the borders between the endocardium and the heart chambers. or are obscured by noise (see Figure 6.

In terms of frequency content. side lobe. whereas the rarefactional phases travel at a speed that is lower than the average speed of sound in the medium. This is a simple result of the fact that the speed of sound is larger for the more compressed (denser) regions of the tissue. Unfortunately. with a resultant signal which is several times larger than the basic signal. As a result. high (lower left). increasing the amplitude of the transmitted pressure wave can compensate for this effect. The effect is illustrated in Figure 6. low (upper right). This limitation was recently greatly relieved by one of the manufacturers. the harmonics generated will be much weaker. In theory.15. comparing fetal images in the linear and second harmonic modes of imaging. We thus get a distortion that will cause a waveform that is initially sinusoidal to become more like a sawtooth. The decoded signal amplitude is proportional to the . Coded excitation Higher resolution imaging can be accomplished at high ultrasonic frequencies.14 Image taken at very low (upper left). The second harmonic image is then generated by filtering out the fundamental component in the receive signal. high frequency ultrasound is highly attenuated by the tissue. as described in Figure 6. the waveform distortion is equivalent to second as well as higher harmonics generation (integer multiples of the original frequency). the amplitude of the transmitted wave is limited in order not to cause adverse biological effects in the tissue. and very high (lower right) dynamic ranges. However. Since the distorted and scattered energy caused by reverberation. and grating lobe artifacts is much weaker than the transmit energy. Coded excitation encodes a special ‘signature’ to the ultrasonic beam by repeating a specific pattern of ones and zeros. the tissue harmonic image contains minimal noise and clutter compared to fundamental imaging. applying coded excitation to ultrasonic imaging. The amount of distortion is proportional to the distance (measured in units of the wavelength) and to the 2nd power of the acoustic pressure. phases travel at a speed that is higher than the average sound velocity in this medium.The physics of ultrasound imaging 69 Figure 6. The received beam is then decoded accordingly.16.

4 0. length of the code. . The effect of coded excitation is illustrated in Figure 6.8 0.2 0 –0. due to inherent physical and technological limitations an ‘ideal system’ does not exist. Figure 6.4 –0.6 0.8 –1 0 1 2 3 4 ωt/2(rads) 5 6 7 Figure 6.6 –0.2 –0. and the user will generally face painful compromises.4 –0.4 Amplitude 0. deep abdominal imaging can be obtained at frequencies that are double the frequencies obtained using conventional techniques.2 0 –0.70 Fetal Cardiology 1 0.6 0. Gray-scale system controls and their effect on image quality As already discussed above.2 –0.15 Initial waveform (top) and distorted waveform (bottom).8 –1 0 1 2 3 4 ωt/2(rads) 5 6 7 Amplitude 1 0.16 A fetal image in tissue harmonic-imaging mode (a) and in linear mode (b). comparing coded excitation and conventional non-coded excitation imaging.17. Using this technique.6 –0.8 0.

The situation is described schematically in Figure 6.6 Gray-scale system controls Favorable effect See weak echoes Optimize gain at all depths See more detail in the near field Improve orientation Improve the ability to measure large objects Reduce speckle and temporal noise Increase image area that can be ‘seen better’ Improve image resolution of object of interest Optimize image for the desired diagnosis Lose the image at the far field Reduce frame rate. the imaging parameters are preset for the various applications in most expected conditions. The frequency change of the reflected beam will be positive (the received reflected beam will have a higher frequency than the incident beam) if the target is moving toward the transducer and will be negative (the received reflected beam will have a lower frequency than the incident beam) if the target moves away from the transducer. Doppler motion analysis is performed within a ‘sample volume’. e. This would mean that for ‘heavy patients’ one needs to image at lower frequencies in order to gain penetration.g. high frequency ultrasound is highly attenuated. The way the motion is described is called a ‘Doppler spectrum’. for example. Each of the red blood cell clusters. myocardial motion. When a certain ‘sample volume’ in the body is insonated. the reflected wave changes its frequency. although the scattering from blood cells is extremely low. Table 6. and the objective is to provide a statistical description of the motion of the various constituents within this region. Doppler Introduction Ultrasonic imaging allows us not only to see how organs look. we can still ‘see’ them based upon the differences in velocity compared with the surrounding tissue. Some adjustments by the operator will therefore almost always be required.17 An image in coded excitation mode (left) and in conventional (non-coded-excitation) mode (right). In most modern systems. We can easily detect and analyze motion of blood (down to the arteriole level) or tissue. The system controls and their effect on overall image quality are summarized in Table 6. Motion analysis is based upon the physical fact that when a sound wave is reflected from a moving target. In the first example. one would like to image at the highest possible frequency in order to improve resolution. The term ‘sample volume’ relates to a certain region of the image. always be somewhat different from those for which the presets were determined. will scatter an Figure 6. but also to a certain degree to analyze how they function. The frequency change is proportional to the velocity component of the moving target along the direction of the beam. and to the beam frequency. However. however. A Doppler spectrum is actually a histogram that describes the distribution of velocities (along the insonating beam) within the ‘sample volume’.The physics of ultrasound imaging 71 Thus. moving at a certain velocity.18. The situations in ‘real life’ will. there will generally be a multitude of reflected waves at a variety of ultrasonic frequencies. and smear of fast moving objects Smearing of fast moving objects Reduce frame rate Adverse effect Strong echoes will be saturated Control change Increase gain Set time gain compensation Increase imaging frequency Increase system field of view Increase persistence Add transmit focus Move transmit focus to region of interest Select required gray-scale curve .6.

Another difference between the two modes is that unlike CW Doppler where insonation is continuous. Since the vessel walls are usually extended in size. Pulsed wave Doppler In this mode of operation. however. Only the signals returning from the ranges of the sample volume will be analyzed. A Doppler signal is therefore composed of a range of frequencies with varied amplitudes. ultrasonic beam with the corresponding Doppler shift. This type of operation was used in obstetrics more than 20 years ago. This means that only the received signal at a predefined time gate (which corresponds to a specific range gate) is processed and included in the spectral analysis. This difference in frequency is called the Doppler frequency shift. intensities of signals from the vessel walls might be as much as 100 000 times stronger than reflections from the blood cells themselves. it is no longer found in modern imaging equipment for obstetrics and gynecology. the same transducer elements are used for both transmitting and receiving the ultrasonic beam. In other words. There are.18 An ultrasonic beam. Another example is capillary flow. which means lower hazard to the patients. the reflected signals from them could be much larger in amplitude than those of blood cells. The transmitted signal is a relatively short pulse. however. is reflected by moving red blood cells. In practice this means that the operator cannot determine the precise origin of the Doppler signal along the ultrasonic beam. Unfortunately. Contrast agents are. The change in frequency of the wave is proportional to the velocity of the red blood cells. Continuous wave Doppler Continuous wave (CW) Doppler is performed using a transducer composed of two separate elements. insonation in PW Doppler generally lasts less than 1% of the time. There is practically no way to measure blood flow in the coronaries using conventional Doppler ultrasound techniques. in most cases the velocities of the arterial walls are significantly lower than the velocities of the blood cells. However.72 Fetal Cardiology Moving reflector Transmitted beam Reflected beam Figure 6. The first element functions continuously as the transmitter . although it may be several times longer than the pulse in gray-scale imaging. while the second element functions continuously as the receiver. not permitted in obstetric applications and are therefore beyond the scope of this book. and Doppler color flow imaging. The intensities of the individual waves at given Doppler shifts will be proportional to the number of red blood cells moving at this particular velocity. This allows the use of lower signal intensities. The sample volume can be positioned anywhere on the image and may assume almost any desired size. The Doppler measurement starts by positioning a sample volume on the two-dimensional gray-scale image prior to activating the PW Doppler beam. The returned signal is ‘gated’. In both cases. novel methods utilizing ultrasonic contrast agents are employed. and in most cases also for gray-scale imaging as well as for color flow imaging (see explanation below). The magnitude of the Doppler frequency shift is dependent on the angle between the ultrasonic beam and the red cell velocity vector. When the reflected signal reaches the transducer. emitted by the transducer. Doppler modes Three modes of operation will be discussed: continuous wave (CW) spectral Doppler. when the beam is perpendicular to the wall. not only the red blood cells but also surrounding tissues such as the arterial walls are also moving reflectors that contribute to the Doppler signal. In extreme situations. the difference between the transmitted and the returned frequency is measured. Luckily. cases such as coronary flow where this is not true. the ‘sample volume’ in this case consists of the entire beam. It is still used in monitoring equipment and in echocardiology scanners. This fact is used to discriminate between the two types of signal. pulsed wave (PW) spectral Doppler. In CW Doppler all blood flow velocities detected along the axis of the transducer will contribute to the Doppler signal.

meaning that large velocity ranges may be measured at close image ranges only. . sometimes referred to as power Doppler or ‘ultrasound angio’. Slowly moving tissue will also cause a Doppler signal. A detailed spectral analysis at each gate would be too long if reasonable accuracy is required and is therefore not applicable. the transducer will function as a receiver until the signal has returned from the specified depth.The physics of ultrasound imaging 73 Pulse repetition frequency After a pulse is transmitted. as illustrated in Figure 6. sometimes called color flow imaging. for a given PRF. which must be filtered out in order to eliminate ‘flash’ artifacts.19. This defines a minimum time between consecutive pulses or a maximum rate or repetition frequency. As a result. the analysis is usually limited to the determination of the average velocity at each gate. in which the Doppler signal amplitude. and the second. Actually. High pulse repetition frequencies (PRFs) are required for measuring high velocities. is displayed as a color map on top of the grayscale image. however.19 Spectral folding caused by aliasing. When the Doppler frequency shift exceeds PRF/2 a folding artifact will occur. which depends upon the range. cannot be made too short. The baseline is the Doppler frequency shift (corresponding to the velocity component along the ultrasonic beam) which corresponds to the center of the spectrum. Between six and 12 ultrasonic beam transmissions are required at each sample volume in order to distinguish blood flow from slowly moving underlying tissue. The PRF is determined from the time difference between two consecutive pulses. the Doppler shift signals in each of them. a two-dimensional distribution of Doppler shift signals can be obtained. The Doppler frequency shifts and the amplitude of the Doppler signal are then analyzed in each individual gate. There are therefore two forms of presenting real-time blood flow information: the first. which is proportional to the density of the blood cells. The data are used to assess either the average velocity (magnitude and direction) or the total amplitude of the signal in each of the gates. resulting in high Doppler frequency shifts. Color flow imaging is based on a Doppler color flow imaging By placing multiple gates (sample volumes) across an entire area in the gray-scale image and separately processing Figure 6. the ‘interrogation time’ at each gate is rather limited. It can be shown mathematically that only when the Doppler frequency shift is less than one-half the sampling rate (PRF/2) can the velocity be properly assessed. This imposes constraints on the maximum range: high velocities can only be measured at close ranges.19 can be ‘unfolded’ by moving the ‘baseline’. Only then can the next pulse be transmitted. employing relatively sophisticated filtering. The interrogation time at each gate. The depth range thus limits the pulse repetition frequency. In order to achieve high enough frame rates. Color velocity imaging In this mode a real-time image representing blood-flow velocities is displayed as an overlay on top of the gray-scale two-dimensional image. the measured Doppler frequency shift may be anywhere between –PRF/2 and PRF/2. The maximum measurable velocity range is therefore dependent on the PRF. in which a colorcoded velocity map is displayed on top of the gray-scale image. This defines a measurable velocity range. The spectrum in Figure 6.

This generally results in a three-fold increase in sensitivity in the capability to demonstrate low flow. The task of determining these relatively minute frequency shifts is generally accomplished by ‘mixing’ (multiplying point by point) the received signal with a reference ‘master clock’ signal that is of the frequency of the transmitted wave. since we are not trying to trace the fast velocity changes during the systolic phase. but only the Doppler intensity. blood flowing toward the transducer is colored red while blood flowing away from the transducer is colored blue. According to this scheme. Most physical phenomena affecting Doppler ultrasound are not relevant in this case. The problem we are trying to solve in detecting Doppler ‘frequency shifts’ (the frequency difference between the incident ultrasonic wave and the waves reflected from the moving objects) is actually quite complicated.21 shows a side-by-side comparison of renal flow as imaged by color velocity mapping and by power Doppler. . There is almost no angle dependence and no aliasing effect. values do not change significantly over the heart cycle. The scale is designed to optimize high or low velocity display.21 Color velocity images of a native kidney (a) and power Doppler image of the same kidney (b). These are to be compared to the carrier frequency (the frequency of the incident ultrasonic beam) which is of the order of several (2–10) MHz.20. Figure 6.20 A color velocity image of a native kidney. An example of a color velocity image is provided in Figure 6. Since power Doppler is velocity independent. The first function in the signal processing chain is called ‘demodulation’. The frequency shifts involved are usually in the range of a few hundred Hz to tens of thousands of Hz. Similar to the gray-scale presentation used in two-dimensional imaging. the velocities measured at each Doppler gate are assigned a corresponding shade of color. The first component is Figure 6.22. Doppler signal processing The chain of the Doppler signal processing is described in Figure 6. This is essentially the intensity of the reflection from moving blood particles. In particular. Power Doppler imaging This is a relatively new way to image blood flow. What is displayed in this case is not the average velocity in the sample volume. This will result in a signal that could be written as a sum of two components having two main frequencies. demonstrating blood perfusion in the kidney. temporal averaging techniques can be employed to reduce noise. so that very low pulse repetition rates can be used to detect very low flow without risking the artifacts associated with too low PRFs.74 Fetal Cardiology color scale. Figure 6. Usually a red and blue color scheme is used.

This overlap is required in order to update the spectra often enough. the direction of the flow can also be determined. f3 … etc. or 256) are used. Usually there is also a certain overlap between data used for consecutive spectra. we also ‘mix’ the signal with a second ‘master clock’ reference signal. T must be small enough so that the frequency content of the acoustic Doppler signal remains constant for the period of measurement T. By passing the resultant signal through a low pass filter. The second component is a low frequency signal that equals the difference between the frequencies of the transmitted and reflected signals. The height along the vertical axis is proportional to the velocity (or frequency Transducer Spectral analyzer Spectral display Figure 6. In situations like these. T ∼ 1/Δf. Thus. as is the case for example in color flow imaging. In order to get also the sign of the shift. Spectral analysis In practice. when the velocity of the particles changes rapidly. Modern systems usually repeat the calculation every 2 milliseconds. the high frequency component is removed (filtered out). f2 = 1/t2. . The spectral density function at each velocity is proportional to the number of blood cells moving at that velocity (and reflecting a sound wave with the corresponding frequency shift). The method commonly accepted in all commercial systems is described below. and the higher is the accuracy. for consecutive periods of time 2–10 seconds. The reason is that the frequency content of the signal changes over time. f3 = 1/t3 … etc. 128. t2.24.The physics of ultrasound imaging 75 a high frequency signal. However. A high pass filter is also almost always used in order to filter out every frequency component that originates from arterial wall and tissue motions. t1. In most practical cases. With the simplistic approach described above. the frequency resolution becomes so coarse that a spectral analysis of the signal becomes meaningless. The estimate is quite accurate even for a short measuring period T. The second signal is identical to the first master clock signal. One actually obtains for each velocity (or Doppler frequency shift) the spectral density function. For detailed spectral analysis. One can obtain a rough idea of the frequency content of the signal simply by measuring all time differences. Its frequency equals the sum of the transmitted and reflected signals. As a rule of thumb. the more data points there are. between adjacent peaks. as the number of frequency measurements (N = Τ<f>. The longer is the ‘measurement time’ T. Sometimes. T is severely limited by other system requirements. Spectral analysis is mostly performed using a highly efficient computer coded algorithm. It is therefore ‘non-real-time’ in nature. in order to follow quick changes such as during systole. It can be shown that using the resultant two demodulated signals. As illustrated in Figure 6. during systole. where Δf is the required resolution in frequency.23. only the average frequency over a given period of time is measured. It can be shown that this process will give us only the magnitude of the frequency shift. called FFT (fast Fourier transform). From these one can then calculate the corresponding frequency shifts f1 = 1/t1. spectral Doppler analysis is usually performed at a single Doppler sample volume (‘gate’).22 Schematic description of the Doppler processing chain. In this case. the methods employed for spectral analysis are much more sophisticated and accurate. for example. Transmitter Oscillator Receiver Demodulator The display of spectral results The information obtained from each spectrum is quite extensive. t3 … etc. we simply average out f1. As discussed above. in order to include several heart cycles. Our next step is to determine the frequency content of the demodulated signal. the more ‘peaks’ there are in the signal. This process must take place over a long enough period of time T. f2. T cannot be made too large. much larger values of T (N = 64. but shifted by 90° (delayed in time by exactly one-quarter of the cycle of the signal). N is of the order of 10. This results in a huge amount of extensive information that must be properly displayed so that it can be easily interpreted by the sonographer. as will be explained below. each ‘spectrum’ (the spectral density function as a function of the frequency shift) is displayed as a narrow vertical bar. The number of particles in each frequency shift range is approximately proportional to the number of measurements with frequencies in that range. An example of such a signal is described in Figure 6. where <f> ιs the average frequency shift) may still be quite large.

6 0.6 -0. The spectrum is displayed as a column of points and is just a different way of displaying the velocity (in the direction of the beam) distribution in the left-hand side of this figure.8 -1 0 (c) 1 0. and the demodulated signal (c).23 Examples of the transmitted signal (a). The brightness of each bar is proportional to the number of blood cells moving at this velocity. The point at the bottom represents the lowest velocity whereas the point at the top represents the highest velocity.6 0.6 -0. 4 140 120 Spectral density (arbitrary units) Figure 6.4 0. The entire velocity range in this illustration has been divided into 128 velocities.4 0. the signal from a reflector moving toward the transducer (b).8 0.2 -0.8 -1 0 20 40 60 80 100 120 140 20 40 60 80 100 120 140 Figure 6.2 0 -0.8 0.4 -0.8 0.76 Fetal Cardiology (a) 1 0.6 0.6 -0. Velocity (arbitrary units) 100 80 60 40 20 0 –4 –3 –2 –1 0 1 2 3 4 Velocity (arbitrary units) 0 –4 .8 -1 0 20 40 60 80 100 120 140 (b) 1 0.24 An illustration of a display of a spectrum (right-hand side).2 -0.4 -0. The vertical position of each of the bars represents its velocity.2 0 -0.2 0 -0.2 -0.4 0.4 -0.

the entire overwhelming amount of information is not actually used for clinical diagnosis. Wave form analysis Note that Doppler frequency shifts do not measure velocity. the end-diastolic frequency shift (D). They are all based on the peak systolic frequency shift (S). In doing so we are giving up the pretension to measure volume flow. The indexes defined below are mostly relevant for evaluating downstream resistance. This is distinctly different from the flat constant velocity waveform usually found in veins. The data provided by the Doppler shifts alone is however extensive enough to allow clinical assessment of the hemodynamic system. as provided automatically by the system. It is called the envelope because historically this curve was manually inscribed by the sonographer as an envelope to the spectral display (after the image was frozen). In some cases. and the temporal mean frequency shift over exactly one cardiac cycle (A) (Figure 6. In today’s modern equipment. The amount of pulsatility carries a lot of information regarding the vascular system.26). When this happens the next spectrum will be displayed in the leftmost position. The spectral ‘envelope’ Often. is marked in red. Figure 6. Several different indexes have been defined in the literature. and so on. . shall concentrate on the hemodynamic information that can be extracted from these frequency shift measurements. the next one will be displayed to its right.The physics of ultrasound imaging 77 shift) and the brightness at each point is proportional to the number of particles moving with this velocity. the Doppler data allow a simple recognition of the existence of flow.25 is an example of a spectral display describing the flow at a certain point of a carotid artery. The envelope.25 Spectral display and envelope. such as is often found in arteries. This is repeated until the whole area that is dedicated to the spectral display is full. Obviously. Perhaps the most important feature of the Doppler waveform is its pulsatility. The pulsatility of the waveform can be represented quantitatively in many different ways according to the vascular property under investigation. In the following we shall show how the information embedded in the spectral envelope (maximum frequency shift curve) can be used in the evaluation of the vascular system. In the following we Figure 6. They are only proportional to velocity. replacing the ‘old’ spectrum displayed there. and so on. The next spectrum is now displayed next to the previous one. Doppler waveform analysis and the Doppler indexes As discussed above. Pulsatile flow. the envelope is automatically computed online (during the scan). The envelope actually describes the velocity of the fastest cluster of cells as a function of time. only the envelope is required in order to provide the relevant hemodynamic information. the maximum frequency shift curve (the spectral envelope) represents the temporal changes in the velocities of the fastest moving blood cells during the cardiac cycle. In order to obtain this function one needs for each spectrum to mark the highest velocity (frequency shift) for which the spectral density function is not zero. is characterized by a pulse-like shape with a systolic peak.

associated with an obliterative process. uteroplacental.g. these sites are conveniently selected in locations where the Doppler signals are relatively easy to acquire.26 Typical time–velocity waveforms. Doppler-only-based techniques are limited due to the angle dependence of the signal. Applications include studies of regional function. However.8 0.2 Velocity (arbitrary units) 1 0. Due to technical limitations. e. slow diastolic flow might be more affected by the filter than the relatively fast systolic flow. and fetal cerebral circulation. S is the maximal value of the maximum frequency Doppler shift.78 Fetal Cardiology 1. the uterine artery. Following recent advances in computer and multimedia technologies. extremely high frame rates are required. and A is the average of the maximum frequency Doppler shift over a heart cycle. Umbilical–fetoplacental circulation State of the art Umbilical Doppler waveform analysis was suggested as a non-invasive tool for the assessment of the fetoplacental . fetal cardiac stroke volume. Recent developments Tissue Doppler and two-dimensional strain (speckle tracking imaging) One of the most important recent technical developments in the field of Doppler imaging is the invention of tissue Doppler (spectral and imaging). Although the principles are not different from those used to analyze blood flow. originally designed to filter out the arterial wall and tissue signals. the flows at the center of the arteries are always faster than the flows closer to the vessel walls. Although all Doppler indexes defined above do not depend explicitly on direction. In the case of flow in the umbilical artery.4 1. this mode of imaging has been implemented only recently. This has stimulated a wealth of new promising ideas and methods that could potentially improve the accuracy of Doppler waveform analysis as a clinical tool for evaluating various vascular systems. fetal heart rate. mainly due to the high pass filter. Until recently. D is the minimal value of the maximum frequency Doppler shift. resulting from variations in the position and orientation of the beam and Doppler sample volume relative to the vessel walls. at the umbilical artery. especially during diastole. When a vessel is approximately perpendicular to the flow direction.4 0. Pulsatility values depend also upon additional factors such as the fetal cardiac output. or the middle cerebral artery. it has become possible to perform simultaneous acquisition and analysis of a multitude of Doppler waveforms in a very short period of time. there still is a residual dependence. As a result. mainly cardiological. certain myocardial areas are excluded. two-dimensional motion analysis through tracking of magnetic tags.2 0 0 1 2 3 4 5 6 7 8 9 10 Time (arbitrary units) A D S circulation in 1978. only major changes in the waveforms such as reverse or no end-diastolic flow in the umbilical artery or a diastolic notch in the uterine circulation are really clinically significant. It has been found that elevated values of these indexes are associated with increased downstream resistance. Tissue Doppler is used for the analysis of moving tissues. etc. The clinical diagnostic value of the method is still questionable due to the following reasons: 1. Figure 6. As a result. One therefore expects a distribution of pulsatility values in the umbilical artery. 2. It has been shown that only when two-thirds of the placental microvasculature is affected will a significant change in the pulsatility index be noticed. The main reason is that for tissue Doppler imaging. It has been shown that an elevated pulsatility index in the umbilical artery is indicative of increased downstream resistance within the placenta. and more. Usually. especially in clinically relevant applications. Such high frame rates have recently become feasible with the introduction of systems with digital beam forming. These systems allow the required high frame rates by receiving several beams for each beam transmitted. cardiomyopathy. The limitations of the present methods and some of the new ideas will be reviewed in the context of fetoplacental circulation. This has led to a multitude of clinical studies aimed at the development of clinical tools for evaluation of the fetoplacental. typically 1–3. diastolic function analysis. 3.6 0. only magnetic resonance imaging (MRI) could provide full. cardiac resynchronization imaging. all presently available methods are based on measuring the Doppler indexes in a very small number of sites.

Thus. and color. high velocities will result in aliasing. fetal arrhythmias. and other malformations of the fetal cardiovascular system and malformations impacting on cardiac function (e. myocardial motion is characterized in terms of tissue velocity and tissue deformation parameters. that allows objective analysis of the complete myocardial motion throughout the entire cardiac cycle. the overall measurement time will be short so that the measurement resolution will be low. 2D strain is in a way a natural extension of one-dimensional Doppler motion analysis. however. tissue and vessel wall motion need to be filtered out. similar in concept to MRI tagging. for example. such as strain and strain rate. and other fetal/maternal problems that may affect the fetal heart. Similar to one-dimensional Doppler. ultrasound’s new acoustic markers keep coming in as some of the previous markers fade out. These natural markers are used in a way similar to the magnetic tags in MRI. Machine controls An optimal Doppler signal requires optimization of quite a large number of system parameters. and its inability to analyze beat-to-beat variability. congenital heart defects such as hypertrophic cardiac myopathies. Fine adjustments are usually for the individual cases. which defines at each gate the number of data points that are available for the calculation. reduces button pushing. that presets are usually designed for average set of conditions in the given application. fetal anemia and other causes of hydrops fetalis. Similar in concept to MRI tagging. Myocardial motion and velocities are then analyzed by calculating frame-to-frame changes.The physics of ultrasound imaging 79 However. most systems will use presets for all clinical applications. twin-to-twin transfusion syndrome.7 below lists the various controls used in the operation of the various Doppler modes. MRI is not widely available for clinical use because it is expensive and time-consuming. one cannot expect the natural acoustic markers to persist throughout the entire cardiac cycle. if the PRF selected is too low. Table 6. Increasing the ‘quality’ will improve the accuracy of the calculation. in which the entire tagging fades out and limits the analysis time to only part of the heart cycle. the tags are short-lived. As seen from Table 6. . Color flow imaging still presents more problems. unlike in MRI. Too much filtering will also eliminate ‘legitimate’ low velocity blood flow signals. Note. maternal diabetes. difficulties in analyzing the whole cardiac cycle due to the short persistence of the tagging. mainly due to their movement in and out of the imaging plane. However. This is illustrated in Figure 6. on the other hand. In order to allow rapid optimization of the gray-scale image. General Electric introduced into the market a new diagnostic tool. ischemia. placental dysfunction/insufficiency. two-dimensional (2D) strain analyzes motion by tracking ‘tags’ (natural acoustic markers) in the ultrasonic image in two dimensions. Indications for functional evaluation of the fetal heart include: intrauterine growth restriction. Figure 6.g. and ensures diagnostic results. If. the PRF selected is too high. As with tagged MRI.27. This again is quite tricky. New features (blue circles) keep coming into the image as old ones (yellow circles) fade away. Doppler. vein of Galen aneurysm). The ‘quality’ of the measurement depends on the number of ultrasonic beams that are transmitted in a given direction.7. but will also adversely affect the frame-rate. the number of controls that need to be optimized is quite large.27 ‘Natural acoustic tagging’. Proper use of the presets saves time. where cardiac failure may be present. Other limitations include relatively low spatial (∼2–5 mm) and temporal resolution (at best ∼30 ms) of the magnetic tags. As also discussed above.

Zerhouni EA. but signal from slowly moving blood particles will not be observed Shifting the baseline will increase the measurement range of positive (negative) velocities and reduce the measurement range of negative (positive) velocities Defines the precise location on the anatomical gray-scale image from where Doppler data are sampled NA Increasing the sample volume will increase signal intensity. Computed sonography. Kremkau WF. 1984. Friedman Z. Moore CC. Woodcock JP. The maximum PRF is depth range dependent. for computation of the velocities along the vector. Noise is also amplified Decreases the signal from slowly moving objects such as vessel walls or surrounding tissue (irrelevant in tissue Doppler). Bluemke DA. not applicable. but will result in reduced frame rates Pulse repetition frequency (PRF) The PRF determines the maximum velocity that can be measured without causing aliasing. 23 (Spec No): S127–40. IEEE Trans Sonics Ultrasonics 1978. 36: 211–20. Orlando: Grune & Stratton. 1–16. 156: 1267–72. but signal from slowly moving blood particles will not be observed Shifting the baseline will increase the measurement range of positive (negative) velocities and reduce the measurement range of negative (positive) velocities. Haberman S. Castillo E. eds. Leitman M. Not provided by some manufacturers. NA. Aliasing can sometimes be recognized when colors are mixed within the same vessel Defines the number of beams transmitted in each direction. 17: 1021–9. In: Sanders R. Scanlan KA. Ultrasound Annual 1985. London: Academic Press. J Am Soc Echocardiogr 2004. Quantitative tagged magnetic resonance imaging of the normal human left ventricle. AJR Am J Roentgenol 1991. Instrumentation and Exercises. Two-dimensional strain – a novel software for real-time quantitative echocardiographic assessment of myocardial function. Gynecol Obstet Invest 1993. Increasing the ‘quality’ may be required for demonstrating very low flow. Irrelevant in power Doppler mode NA Wall motion filter Baseline Sample volume position ROI position Sample volume size Defines the part of the image where blood flow will be imaged NA Size of the ROI The size and position of the area on the image where flow will be demonstrated. Sonographic artifacts and their origins. 11: 359–71. 1982. Lysyansky P. Noise is also amplified Decreases the signal from slowly moving objects such as vessel walls or surrounding tissue (irrelevant in tissue Doppler). 25: 1–6. Effect will increase with higher setting. McVeigh ER. Hill M. .7 Control Total gain The system controls available to the operator for optimizing Doppler measurements Pulsed wave (spectral) Doppler Increases signal received from moving objects of low reflectivity. The maximum PRF is depth range dependent. Sidenko S et al. Diagnostic Ultrasound: Principles. Lima JA.80 Fetal Cardiology Table 6. Bibliography Atkinson P. Top Magn Reson Imaging 2000. Regional myocardial function: advances in MR imaging and analysis. Increasing the size of the ROI will reduce the frame rate The PRF determines the maximum velocity that can be measured without causing aliasing. Doppler Ultrasound and Its Use in Clinical Measurement. Burckhardt CB. as long as the sample volume size does not exceed the boundaries of the vessel NA Color flow imaging Increases signal received from moving objects of low reflectivity. New York: Raven Press. Speckles in ultrasound B-mode scans. Radiographics 2003. A new technique for improved diagnosis of local placental abnormalities: Fourier analysis of intraplacental waveforms. Aliasing can sometimes be indicated by ‘cropped’ spectral displays NA Color ‘quality’ ROI. Effect will increase with higher setting. 1985. region of interest. Maslak S.

Halliwell M. 1988. Lopez H. Clinical Applications of Doppler Ultrasound. . In: Taylor KJW. Wells PNT. Lopez H. New York: Raven Press. 30: 156–63. Burns PN. Wagner RF. Wells PNT. Speckle in ultrasonic imaging. Smith SW. Wagner RF. 26–45. Ultrasonics 1981. Low contrast detectability and contrast detail analysis in medical ultrasound.The physics of ultrasound imaging 81 Smith SW. Statistics of speckles in ultrasound B scans. Instrumentation including color flow mapping. eds. IEEE Trans Sonics Ultrasonics 1983. IEEE Trans Sonics Ultrasonics 1983. Wells PNT. 30: 164–73. Sandrick JM. 19: 225–9. Sandrick JM.

.

1 Lateral resolution of 0. and display Electronic focusing and two-dimensional matrix array transducers Lateral resolution describes the minimal distance between neighboring objects in the plane of the ultrasound beam that can be resolved. that the wave fronts they generate converge at the level of the selected depth (or focus).5–9 Screening for fetal heart disease and detailed diagnosis of fetal heart disease require technologies of different complexity.10 This chapter attempts to review recent and imminent technical advances that have impacted or may affect fetal echocardiography. Figure 7. Using matrix array transducers. but prenatal detection rates vary considerably.1 shows an ultrasound phantom insonated with good and poor lateral discrimination. it affects the image only in the plane of the ultrasound elements. processing. focusing perpendicular to the long axis of the transducer is possible.7 Technical advances in fetal echocardiography Boris Tutschek and David Sahn Introduction Structural congenital heart disease (CHD) is among the most frequently missed anomalies in prenatal ultrasound studies. Resolution depends on the distance between adjacent transducer elements and also on the lateral width of the ultrasound beam (slice thickness) used to interrogate a certain depth.1–4 Various technical advances in ultrasound systems have promised improvements specifically for fetal echocardiography.2 shows a phantom with cystic objects insonated with either a conventional one-dimensional or a new two-dimensional array transducer capable of electronic focusing in two planes. Figure 7. Focusing of the ultrasound beam can be achieved by electronic ‘steering’: adjacent elements are activated at different time points so (a) (b) Figure 7. GE Healthcare). Improved acquisition. avoiding artifacts from structures immediately adjacent to the insonated plane.3-mm nylon threads in a phantom insonated with good (a) and poor (b) lateral resolution (image kindly provided by H Dudwiesus. Technical improvements in transducer technology and image processing as well as a basic description of three-dimensional (3D) technology are complemented with an outlook on the techniques for studying fetal cardiac mechanics. Perpendicular to the plane of the ultrasound elements the focus zone cannot be changed in one-dimensional arrays: the ‘focus’ of the beam perpendicular to the long axis of the elements is fixed. However. Recently. two-dimensional matrix array transducers that can overcome this limitation have become available. and depend on factors such as operator training and experience. Prenatal detection can significantly improve perinatal outcome at least in certain types of CHD. . This electronic focusing has been used in one-dimensional array transducers over many years.

however. In general the diagnostic results were equal. Real-time compound imaging In conventional ultrasound imaging. If this beam hits the interface between structures of different echogenicity (a reflector) at a 90° angle. They studied 50 women in three groups receiving fetal echocardiography with either fundamental or harmonic imaging. consisting of the original and of additional waveforms of smaller amplitude.3). Using a matrix transducer. Paladini et al17 performed a detailed study of harmonic imaging in fetal echocardiography. and the aortic and ductal arches were better using harmonic imaging. the ultrasound beam is emitted at right angles to the elements. Harmonic imaging can be implemented without increasing power output. Tissue harmonic imaging (THI) shows a significant improvement in gray-scale imaging. At other angles reflections are diverted away from the transducer. These lower amplitude. therefore. Reflections or signal losses occurring in particular on cystic structures or irregular shaped borders are reduced using compound imaging (CI. temperature. harmonic frequencies. drops. emitted sound waves themselves also exceed pressure and. leading the authors to the conclusion that in pregnant women of normal weight. Decreasing the insonation frequency. also lowers image resolution. especially in difficult scanning conditions as for example in obese patients. GE Healthcare). harmonic imaging performed better (‘rescue’ modality). and are received. one-dimensional array transducer (a). the more their waveform changes from the emitted pure sinus wave to a compound waveform.11 Signals recorded at twice the insonation frequency are called first harmonic frequencies. fundamental imaging is still the technique of choice for fetal echocardiography. as well as better side-lobe suppression.4). reducing the received signal intensity. yielding better axial and lateral resolution. Real-time spatial compound imaging uses electronic beam steering of a transducer array to acquire several overlapping scans of an object from different view angles (Figure 7.14 THI improved the resolution in half the pregnant patients studied by Treadwell et al. Individual views of the same plane. for an example of compound imaging of the fetal heart see Figure 7. providing better border definition and contrast as well as decreasing artifacts.12 Tissue harmonic imaging can improve imaging of fluid-filled structures. focusing is possible also perpendicular to the transducer’s longitudinal axis. The further the ultrasound waves travel through the tissue. valves. The authors concluded that harmonic imaging improved the image quality and that it was a useful adjunct to fundamental imaging in fetal echocardiography. Kovalchin et al16 compared fundamental and harmonic imaging in a group of fetuses from mothers.13. little or no reflection occurs. Image quality and visualization of the ventricles. The small cystic structures in the phantom are resolved better and over a greater depth field (image kindly provided by H Dudwiesus. when the beam is parallel to the interface. because views are acquired from Tissue harmonic imaging Lower frequency ultrasound signals have a better tissue penetration. 71% of whom had difficult scanning conditions. are reflected. Tissue penetration of ultrasound waves also depends on tissue properties such as tissue density. change as they travel through tissue. but resolution from harmonic imaging was poorer.15 in particular in obese women.2 Improved resolution using a two-dimensional matrix array transducer (b) versus a conventional. but acquired from slightly different angles (up to nine different angles are being used). are compounded into a multiangle image in real-time. In obese women and those with inadequate imaging in fundamental imaging. In the worst case.84 Fetal Cardiology (a) (b) Figure 7. and others. higher frequency waves occur at multitudes of the insonation (fundamental) frequency and are called . however. Harmonic imaging removes the fundamental frequencies using filters and typically shows a narrower main lobe. however. but higher frequency. most echoes are returned to the transducer. The actual frame rate. pressure.

20 Compound imaging alters contrast ratios. promising improved contrast resolution and tissue differentiation. then compounded (computed) before the image can be displayed. By electronically steering the ultrasound beams to ‘look’ at objects from different angles. (a) (b) Figure 7.18 In a clinical study of breast tissue. but also increased the apparent target width. involving different types of filters. averaging multiple frames for CI invariably reduces the frame rate. In addition. matrix-array transducers.4 A normal fetal heart (at 23 weeks’ gestation) imaged without (a) and with compound imaging (b) (but otherwise with identical settings).23 Speckle reduces the spatial and contrast resolution in ultrasound images and results in an artificial ‘sono-texture’ that becomes apparent when highest magnifications are used (Figure 7. Speckle reduction Speckles are artifactual ultrasound signals caused by the interference of reflected ultrasound energy from scatters that are too small and too close to be resolved with the frequency used. CI increased signal-to-noise ratio. GE Healthcare: CrossX Beam™ imaging).5). and (3) postprocessing approaches. techniques for (1) improved resolution. is to be based on subjective contrast impression using harmonics and compound imaging. CI has been marketed under various names (e.3 Conventional ‘single line of sight’ (a) and real-time compound imaging (b). true reflectors can be better differentiated from artifacts because their reflections appear consistently in several of the angled views. limiting its practical use. for example of fetal bowel echogenicity or cardiac valve or myocardium appearance. Philips: SonoCT.21 similar to using higher frequency transducers. Speckles appear as a pseudo-structure in histologically homogeneous tissue due to wave interference of reflected signals.19 A study of compound and harmonic imaging in fetal diagnostic scanning at 11–14 weeks’ gestation reported improved visualization in various anatomical regions for the combination of compound and harmonic imaging. and harmonic imaging. including higher frequency transducers. for example.Technical advances in fetal echocardiography 85 different angles. coded excitation. (2) temporal averaging and spatial compounding. CI reduces ultrasonic artifacts such as speckle.g. For the fetal heart there was a small (but probably not significant) improvement over conventional B-mode versus compound plus harmonic imaging. (a) (b) Figure 7. A prominent speckle structure in a diagnostic image can obscure true objects with little contrast to the neighboring tissue. . To reduce speckles.22 which must be kept in mind when qualitative assessment.

Germany). a fact that has been noted as well regarding other subtle anatomical details such as nuchal translucency or smallest distance measurements. ContextVision: GOPView.23 It should be mentioned that. Siemens Medical Ultrasound. for example.26 The use of colors in B-mode imaging The physiology of human visual perception involves two types of receptors in the retina. utilizing color. . Post-processing algorithms in modern ultrasound systems reduce speckles by real-time image analysis and interposition of gray tones to ameliorate the speckled appearance. gray levels. Rods typically resolve between 20 and 60. fundamental imaging combined with speckle reduction. small structures may appear different in size from fundamental imaging. but cones. and intensity. enable differentiation of up to seven million colors in photopic vision.7). (a) High magnification shows the artifactual pattern of the tissue (speckles). using harmonic and also compound imaging. the wide dynamic range available from raw ultrasound data Figure 7. but without compound and harmonic imaging.24 Post-processing algorithms for speckle reduction have been introduced under different names. the so-called scotopic vision. Interference of the reflected ultrasound waves at the level of the resolution limit causes an artifactual pattern appearance (see alternating white and dark areas indicated by the arrows) without histological correlate in the tissue. The rods are specialized for perception of low light intensities.25 GE Healthcare: speckle reduction imaging (SRI). Erlangen.6). Philips: XRES.23 Speckle reduction algorithms aim to remove the distracting speckle pattern without reducing the detail in the ultrasound image (Figure 7.5 Ultrasound speckles. saturation.27 So-called ‘photopic ultrasound imaging’ was available on a commercial ultrasound system also used for fetal studies (Elegra. which are the dominant receptors in the fovea centralis where they provide highest resolution perception. In this implementation.6 Speckle reduction. providing sharp vision in low light surroundings. The techniques of geometric filtering to reduce speckle were first applied to radar images and later also to ultrasound images. under ideal circumstances up to 250. showed the most accurate distance measurements (Figure 7. (b) speckle reduction applied to the identical still image.86 Fetal Cardiology have been employed. In an ultrasound phantom designed to measure distance below 1 mm. (a) (b) Figure 7. Photopic vision is performed by the cones.

Toshiba Europe Medical Systems) there is a palette using several colors in substitution for the conventional simple gray-scale. Rendered 3D images are often produced using a colored display.02 cm Measuring accuracy of fundamental and tissue harmonic imaging in an ultrasound phantom with membranes 0. Heling et al32 studied ‘advanced dynamic flow (ADF)’.30 Volpe et al31 showed that B-flow in combination with STIC was superior to 2D gray-scale and color Doppler alone in the detection of small pulmonary vessels (major pulmonary collateral arteries). When compared to fetal Doppler ultrasound. was transformed in real time from the conventional shades of a gray-scale (scotopic) into a photopic image.28 two separate beams: the reflection of the first is used to reconstruct the cross-sectional imaging. Video clips 7. Another way of exploiting colors to display ultrasound data has been implemented for three-dimensional ultrasound in a new commercial ultrasound system (iE33. a sensitive Advanced methods for motion detection B-flow Conventional Doppler techniques in vascular imaging tend to exaggerate the real size of a vessel (‘bleeding’ of the color). for example using red and blue denoting flow towards and away from the transducer and velocity encoded in the brightness of the color signal. such as total anomalous pulmonary venous return. for example the cells in the moving blood. In one commercial system (Aplio. whereas the amplitude of the reflected signal correlates with the reflected energy. displays the amplitude component of the reflected signals.Technical advances in fetal echocardiography 87 a b c 1 1 Figure 7. offering a potential for the detection of small cardiac vessels. however. enabling much finer delineation over a wide dynamic range. for example monochromatic coloring of B-mode images. The combination of both yields a sensitive motion display with directional information. B-mode imaging can be extended to detect blood flow independent of Doppler-derived signals by using digitally encoded ultrasound. the morphological information detected and displayed by gray-scale B-mode imaging may be lost by the overlying color or power Doppler signals.3 mm apart. This ultrasound system providing photopic imaging is now no longer in production. Traditionally. but color to enhance the visual perception. which may crucially aid in diagnosing complex cardiac malformations with pulmonary vascular involvement. c) harmonic imaging at 14 and 10 MHz (reproduced with permission from reference 26). Power Doppler. is available in many ultrasound systems. B-flow alone or in combination with spatiotemporal image correlation (STIC) allows visualization of fine small vessels with low velocity.03 cm 1 L 0. but no fetal studies have been published. Directional power Doppler The reflected signal from Doppler insonation contains different information that can be used in various display modalities.29. Photopic imaging has been studied in internal medicine. (b. amplified beam are analyzed for the moving blood particles. color Doppler has been used to encode motion direction. such as pulmonary veins. and disadvantages of Doppler-based flow detection such as aliasing and signal dropout at orthogonal scanning angles are avoided. The frequency shift of the reflected signals indicates the velocity of the reflectors. while the reflections of the second. In addition.7 14 MHZ nativ Abb 14 14 MHZ THI 10 MHZ THI 1 L 0. Philips Medical Systems) to enhance the depth perception: ‘dynamic colorization’ is a 3D display modality coloring elements in the foreground (‘closer to the viewer’) differently from those in the background.2 show the application of dynamic colorization for fetal 3D studies. (a) Fundamental imaging at 14 MHz. Both signals are displayed in the same gray-scale image. An ultrasound beam is encoded in . B-flow has higher resolution and frame rates.1 and 7.

8 Multiplanar imaging of a normal fetal heart at 23 weeks’ gestation. and en face view of the interventricular septum (IVS). LV/LA. all structures can be viewed at different times in the cardiac cycle. Other approaches have also been studied (for reviews see references 37 and 38). Navigation and display of ultrasound volume data Reconstructed fetal 3D echocardiography Reconstructed 3D echocardiography currently is the dominant clinical technique for fetal 3D echocardiography. overcoming some of these limitations. Clinical systems commercially available today use STIC for gray-scale and color Doppler. and inversion modes. be difficult due to an unfavorable fetal position. right ventricle/atrium.88 Fetal Cardiology flow modality based on power Doppler imaging with directional information for fetal studies including fetal echocardiography. One (or several) cross-sectional plane(s) can be placed anywhere within a volume. Multiplanar imaging describes a display with three orthogonal planes (Figure 7. reconstruction of 3D views (so-called ‘rendering’39. Failure to be able to do so may even hint at structural heart disease. left ventricle/atrium. pulmonary vein(s) PV. IAS. we will only briefly describe the display modalities available from volume echocardiographic data. if a fetal (a) (b) Diaphragm RV MS RA Pulm. This may. a stored volume (sequence) can be manipulated digitally.35. either in the plane of acquisition (highest spatial resolution) or in any other (reconstructed) plane.48 In contrast. Three-dimensional sequences of the fetal heart can be displayed in various forms. views of the great arteries can be derived virtually from the standard four-chamber view. Therefore. pulm. and higher sensitivity. enabling interactive scrolling through the heart offline. or lack of operator experience. STIC is described extensively in Chapter 14. but are not available yet for broad clinical use. movements.41–44 and the ability to quantify fetal cardiac volumes45–47 have provided fascinating insights into and very graphic three-dimensional representation of the fetal heart. STIC was first introduced into a commercially available ultrasound system by Kretz Ultrasound (now GE Healthcare).v LA Pulm. They concluded that ADF was superior to conventional color Doppler vascular imaging with higher resolution. A sequence of sonographic data from a series of crosssectional images is acquired while an automated probe sweeps the ultrasound plane across the fetal heart. and is today incorporated into other ultrasound systems (Philips. Because of the given anatomical position and relation of the normal structures. good lateral discrimination. (a) The display shows three orthogonal cross-sections with four-chamber view (top left).40). and by rotation and slight angulation to obtain views along the vessels. provided that the volume covers these structures.36 but the reconstructed nature of the volume leaves room for motion artifacts. RV/RA. pulmonary valve. Volume data from STIC displayed as multiplanar imaging. Inspecting adjacent cross-sectional planes in a volume scan is particularly useful for examination of the great arteries.v dAo RV MS LV Left lung Stomach MS RA μs PV Figure 7. Offline reconstruction then generates a virtual cardiac cycle comprising data from multiple heartbeats which are rearranged using an algorithm that composes volume data from multiple cross-sectional images33 (for review see reference 34). If a whole cardiac cycle has been reconstructed.v. the outflow tracts can be visualized by moving the transducer toward the fetal head to obtain cross-sections.36. however. providing flow information almost in B-mode image quality. the combination with color Doppler.8). Using two dimensions only. interatrial septum (see also corresponding Video clip 7. For example. Medison). (b) Annotated image with the intersection of all three planes represented by the red dot. short-axis view (top right). tomographic.3) . grey shaded area: IVS.

apparent tissue transparency.50 A semiautomatic algorithm helps to display the relevant cardiac structures in one or several tomographic or multiplanar panels. Using particular settings for display thresholds.Technical advances in fetal echocardiography 89 cardiac volume has been acquired in the four-chamber view insonation.or longitudinal sections of great arteries can be extracted from the data set following a simple algorithm (e. In tomographic imaging several parallel cross-sectional planes through the same volume are displayed similar to a typical CT or MRI study (Figure 7.51–53 In the current release of the Voluson ultrasound system (GE Healthcare). while primarily digital imaging modalities (magnetic resonance imaging (MRI) and computed tomography (CT)) display structures in relation to standard anatomical orientations (sagittal.9 and Video clip 7. ‘iSlice’). coronal. Tomographic imaging of 3D fetal echocardiographic data can elegantly display normal or structurally abnormal planes from one insonation angle (Figure 7.10 and Video clip 7. 2: origin and branching of the main pulmonary artery. the ‘spin’ technique48). Panel –3 shows the fetal stomach. ‘tomographic ultrasound imaging/TUI’. cross.49. transverse) and in parallel planes or cross-sections. The top left image shows a sagittal section orthogonal to the other views. three-dimensional representations can be displayed on a two-dimensional computer –3 –2 7. Typical 2D echocardiography aligns its standard planes with one of the ventricular axes. . Rendering displays either external or internal surfaces of organs from volume data. *: indicates the level of the four-chamber view. Tomographic displays of ultrasound volume data have been given different commercial names (‘multislice’. Tomographic imaging has successfully been applied to the heart at different gestations and to CHD. indicating the spatial relationship of the horizontal sections (see also Video clip 7.54 Rendering modes are an alternative to the cross-sectional modes. VCAD) is now available commercially. an automated approach (‘volume computeraided diagnosis’. Volume echocardiography can combine the conventional 2D imaging and orientation standards.4). 1: origin of the aorta. –2: hepatic veins converging toward inferior vena cava (IVC).0 mm –1 1 2 3 Figure 7.9 Tomographic display of a fetal volume containing parallel cross-sections of the upper fetal abdomen and thorax. 3: cross-section of the head and neck vessels in the upper mediastinum.4).g. and shading techniques.5). –1: IVC passing through the diaphragm.

RV.57.63–67 The first report on 3D measurement of fetal cardiac ventricles by Chang et al.61. generating a more plastic impression than the thin slice usually displayed in a conventional B-mode image (Figure 7. which combines the anatomic detail of fluid-filled structures in inversion mode with the dynamic information of color Doppler (Figure 7.62 B-flow combined with STIC as well as these rendering techniques provides even more detailed visualization of small normal and aberrant vessels.58 STIC can be combined with color and power Doppler.and late-gestation fetal hearts.65 In vitro experiments using STIC demonstrated acceptable accuracy for volume and even mass estimations in the range comparable to mid.45–47. power Doppler can also demonstrate extracardiac structures such as pulmonary vasculature. left ventricle. monitor or in a printed image. LV. aorta.11). Surface rendering was originally developed to display the outer surfaces of solid objects such as the fetal face or the skeleton in three dimensions. Initial experiments using non-gated 3D mode from free-hand acquisition confirmed the correlation between ventricular chamber volumes and gestational age.43.45.46 3D inversion mode sonography combined with STIC represents another and possibly more reproducible method for estimating fetal cardiac ventricle volume. Semiautomatic quantification Functional and quantitative analysis of the fetal heart is yet another promising new area of volumetric fetal echocardiography research.56 including cutting the heart in halves at the level of the four-chamber view. Philips Medical Systems).11f).42 The most recent extension to the color rendering modes from STIC is the color inversion mode (HD11 XE.40. Ao.90 Fetal Cardiology Normal 4CV Normal 4CV Ao arises from LV and above 4CV (a) PA originates cranially and from RV (b) PA arises from LV and above 4CV Ao ventrally and from RV (wide arch) Figure 7.63 using a fast automated sweep. pulmonary artery. teratoma. right ventricle. generating ‘digital casts’ of the fetal heart and vasculature in both normal and structurally abnormal cases. Technically different 3D technologies have been applied to measure cardiac ventricular volumes and masses. Rendering in inversion mode shows only fluid-filled spaces (Figure 7.5). cropping into the heart will display its internal surfaces.31.36.30. generating structural and functional cardiac information and angiography-like images. or looking from an en face view of the valve plane.66 Live 3D .55 For rendered images of the fetal heart. PA.10 Tomographic imaging of a normal heart (a) and a dextrotransposition of the great arteries (b). For the same gestational age.59. demonstrated a linear increase of the cardiac volume between 20 and 30 weeks and that 3D volumetry had a better reproducibility than 2D measurements. identical settings of slice distances at three different levels at and above the four-chamber view (4CV) display the diagnostic sections (see also Video clip 7.42.60 Due to its high sensitivity for low velocities. and chorioangioma.64.12).39.

e. semiautomated virtual casting of fetal heart using live 3D ultrasound data algorithms have been developed that can expand ‘seeds’ placed into the cardiac cavities in a 3D volume automatically. rendered images: (c) surface-rendered view. i. Cross-sections: (a) B-mode. (e) ‘glass-body’: region of interest as in (d) but with transparent tissue rendering over color Doppler. cropped to display only the ‘cranial half’ of the heart. Preliminary data show that non-reconstructed fetal cardiac volumetry and STIC technology exhibit reasonable concordance between these two measurement approaches.11 Cross-sectional and rendered views of a normal fetal heart at 29 weeks’ gestation (all images derived from a spatiotemporal image correlation (STIC) volume acquired in a four-chamber view plane). (c) (d) (e) (f) mode using newer transducer technologies. (f) inversion mode (thin volume slice cropped to display only the four-chamber plane) (reproduced with permission from reference 28). twodimensional matrix array transducers.67 Finally. (d) color-only rendering of the whole volume with display of hepatic veins and inferior vena cava (blue) and ventricular inflow.Technical advances in fetal echocardiography 91 (a) (b) Figure 7. . holds the promise of measurements in non-reconstructed data sets. performing both a segmentation and a volumetric analysis of cardiac chambers (Deng. (b) color duplex with biventricular inflow (in red). personal communication).

1997). Sparse array matrix transducer The first 2D array transducer available was a sparse 2D array (Volumetric Medical Imaging Inc. has been introduced (Philips Medical Systems.13).12 Color inversion mode (normal fetal heart. and rapid new technique. .73 Quantification of cardiac ventricles was also attempted using such a system (Figure 7.74 Routine application of the initial devices to fetal echocardiography remained limited.92 Fetal Cardiology (a) (b) Figure 7. miniaturized multiplexers built into the probe handle and capable of generating in excess of 20 B-mode volumes per second. comparing 3D with 2D visualization rates. NC. Live three-dimensional ultrasound using matrix array transducers Matrix array transducer Transducers with more elements and broader frequency bandwidths improve both resolution and tissue penetration. image quality was inferior to 2D mode. Full cardiac volume rendering (viewed cranially) in diastole (a) and systole (b) (reproduced with permission from reference 28). and even color duplex volumes. facile. enabling full beam steering in all three dimensions. 29 weeks). while coarse volumetric assessment of the fetal cardiac ventricles was feasible. a full 2D array transducer with some 2800 individual elements. While standard planes could be visualized more easily in three dimensions. Durham.69–71 Sklansky et al72 were the first to evaluate a (sparse) array matrix transducer for real-time threedimensional fetal echocardiography to evaluate fetal cardiac anatomy and function in 10 human fetuses (four with congenital heart disease). A major goal in transducer technology has been the development of an electronic two-dimensional array. and suggested that live 3D fetal echocardiography could be a significant tool for prenatal diagnosis and assessment of congenital heart disease. They concluded that the system allowed comprehensive visualization of fetal cardiac anatomy and color Doppler flow unattainable by two-dimensional approaches. resolution even in the third trimester was generally insufficient for structural diagnosis. 2001).68 system for real-time 3D fetal echocardiography prospectively in 13 fetuses at mid-term. Another group also used this 3D scanner Full matrix array transducer Recently.. They concluded that fetal real-time three-dimensional echocardiography is a feasible. The system used displayed the volumetric data as a series of four simultaneous planes. Maulik et al75 used such a 4-MHz 2D array transducer in 12 fetuses between 16 and 37 weeks.

and enabled offline retrieval of views not visualized during the actual scans. or computation of cardiac output from flow velocities and vessel diameters (for a recent review see reference 78). estimating size and function of the right and left ventricles. and evaluating mechanism of valvular regurgitation and pulmonary obstruction. 2. a section perpendicular to it (through the aortic outflow) in the lower right. (reproduced with permission from reference 28). locally.79–84 Figure 7. Philips Medical Systems and Vingmed/GE Healthcare) or pediatric (7 MHz.17 shows an example of normal and abnormal fetal cardiac rhythm. Di Salvo et al87 .7 shows a normal cardiac pulsed wave TDE study.g.67 It remains to be seen how they can be applied to the diagnosis of structural fetal cardiac disease. Philips Medical Systems) volumetric imaging. contractility Motion of the fetal heart as a whole and of its individual regions. focusing on rendered images of the fetal heart in comparison to conventional cross-sectional imaging. At least three other transducers have been introduced for adult (3 MHz.14 and 7. for example for localizing multiple cardiac tumors.77 Preliminary data using these systems show that they can be used for fetal cardiac volumetry (Figures 7.13 Real-time fetal 3D echocardiography using the first generation two-dimensional matrix array (Volumetrics. Tissue Doppler Fetal tissue Doppler echocardiography (TDE) has been used to assess regional diastolic and systolic wall excursions and for functional assessment of the fetal heart.Technical advances in fetal echocardiography 93 Figure 7. e. mechanical cardiac intervals. 3D was helpful.5-MHz matrix transducer). Volume-rendered displays identified all major abnormalities. In pathologic fetal hearts. Blood flow Doppler studies provide data from velocity waveform analysis across the atrioventricular valves. can be studied to describe the cardiac mechanical function globally or Strain and strain rate. or in the veins close to the heart. also enabling diagnosis of fetal cardiac rhythm disturbances.6 show color tissue Doppler imaging of a normal fetal heart. as described above) or on velocities of intracardiac blood flow or cardiac structures (tissue Doppler echocardiography). The accompanying Video clip 7. Sonographic measurement of cardiac function can be based on ventricular dimensions (fractional shortening calculated from M-mode or B-mode imaging to estimate the ejection fraction or from 3D measurement. and two perpendicular images from the elevation plane (cross-sections through the two ventricles in short axis)). Cardiac mechanics Motion detection. as well as intracardiac blood flow.84–86 Figure 7. quantitative data include acceleration times. Realtime 3D echocardiography should provide an accurate means of determining chamber volumes and cardiac mass. volume-rendered displays had only slightly inferior image quality compared with conventional two-dimensional images. speckle tracking Because overall tissue velocities alone have not been sufficiently sensitive in fetal cardiac studies.16 and the accompanying Video clip 7. Quantitative information can be obtained by using either color or pulsed wave Doppler TDE. Acar et al76 reported their experience with bi-plane (simultaneous display of two different cross-sections.15). in the great arteries. Fetal echocardiography at 25 weeks’ gestation shows the acquisition orientation (four-chamber view on the top right. Sklansky et al55 studied 30 fetuses with this matrix system. simultaneous display of both outflow tracts) and live 3D (rendered) imaging with the same system in 60 fetuses between 22 and 34 weeks.

. WA).94 Fetal Cardiology Figure 7.14 Real-time fetal 3D echocardiography using a fully electronically steerable two-dimensional matrix array transducer (x3-1/iE33 and software QLab 3DQ Advanced. LV Septum Se pt um Apical view Cranial LV Apex Cranial Lateral view Base Diastole Systole Figure 7. WA). Phillips Medical Systems. Philips Medical Systems. Bothell. generated using semiautomatic volume calculation software from a volume loop acquired using real-time 3D echocardiography (QLab 3DQ Advanced and x3-1/iE33.15 Virtual endocardial casts of the normal human left cardiac ventricle in a 28-week fetus in diastole and systole and endocardial volume curves. Bothell. The ultrasound images represent three cross-sections orthogonal to each other and a 3D niche view of a normal 26-week fetal heart.

Strain rate imaging by speckle tracking imaging (STI) is less affected by these factors. fetuses and one fetus with aortic valve atresia. Strain rate imaging has been used in various adult cardiac conditions including ischemic heart disease and myocardial dyssynchrony (for an extensive review see reference 90).6). Individual segments of the beating heart. The systolic (S’) and diastolic (E’ and A’) peaks can be seen and the segments of the mechanical action of the heart can be measured in milliseconds.7 cm/s Figure 7. In addition. showing severely reduced strain rates in the affected ventricle. and found an increase of longitudinal deformation with gestational age.e. can be tagged and tracked by a semiautomatic image analysis algorithm due to their ‘speckled’ appearance (Figure 7.Technical advances in fetal echocardiography 95 studied regional myocardial deformation (strain and strain rate) using tissue Doppler in 75 normal human fetuses. Systole (b) S‘ .8). or.18).7 –10. STI can be used to assess longitudinal strain in the human fetus91 (see Video clip 7. Doppler-based measurements. see also Video clip 7. STI exploits the sonographic texture of the cardiac walls in a two-dimensional image sequence of the beating heart. The deformation (or stretching) of tissue. the faster the myocardium contracts or relaxes. the faster the deformation occurs between individual points in the myocardium. regions along the myocardial walls in long. They described the characteristics of ventricular filling.00ms/s Systole S‘ E‘ A‘ Diastole 1sec (a) Diastole E‘ A‘ Extrasystole Figure 7. i. however.16 Color tissue Doppler image of a normal fetal heart (reproduced with permission from reference 83.17 (a) Normal pulsed wave (PW) tissue Doppler (TDE) tracing of the right ventricular valve annulus and (b) PW TDE showing an extrasystole (reproduced with permission from reference 83).00ms 0. normalized to its original size or shape. Larsen88 obtained tissue Doppler sequences to calculate fetal strain and strain rate in three normal +10.89 The higher is the strain rate.or short-axis views. is called cardiac strain. suffer from the limitation of angle dependency. One intriguing 35 175 55 95 110 12 6 cm/s 95. and is the only sonographic measurement that allows rotational or twist analysis of the left ventricle of the heart. quantitative analysis based on Doppler data is hampered by measurement errors and by its angle dependency. in other words. The rate at which this tissue deformation occurs is called the strain rate.00ms 0.

9). similar to wringing a towel dry. larger aperture. strain. A break point will occur that clearly sets different data format and analysis formats for fetal cardiac imaging.18 Speckle tracking uses the artifactual. Figure 7. more complex. or even curved plane matrix arrays will be developed. as distinct from the other 3D methods used in perinatology.92 STI can also be used to study left ventricular torsion in the human fetus91 (see Video clip 7.96 Fetal Cardiology Figure 7. The growing interest and understanding of the potential for changing the natural history of heart disease with prenatal cardiac interventions will lead to developments of miniaturized devices for intrauterine applications. These will provide 4D near-field imaging and potentially offer integrated therapeutic options such as radiofrequency. aspect of the complex architecture of the left ventricle is the ability to produce torsion of the ventricle. twist. laser. This fundamental component of left ventricular function is caused by the helical arrangement of myocardial layers. Prospective studies are required to establish reproducibility of fetal STI and possible uses. Automated and tailored methods for edge detection should substantially improve the ability to integrate cardiac volume measurements into the clinical realm of fetal echocardiography. and the assessment of right and right ventricular interaction taking into account the effect of through-plane motion. high frame rate 3D/4D acquisition of high resolution for speckle cluster mechanics should yield methods for defining cardiac mechanics. and torsion measurements of the fetal ventricles. for denser acquisitions. A bifurcation in the pathway for matrix array implementation It is likely that matrix array development will proceed in two directions: for transabdominal imaging of the fetal heart. Likewise.19 Capacitive micromachined ultrasound transducers (CMUTs) (scanning electron microsope image). They will incorporate the latest non-ceramic . high-resolution fetal cardiac imaging will correct aberrations which occur along the imaging pathway. Future developments Automated 3D/4D studies of ventricular volumes and chamber mechanics It is likely that in the near future the ability to acquire and compute and measure 3D/4D images of the fetal heart will continue to improve. or high-intensity focused ultrasound therapies. but region-specific pattern generated by wave interference to track smallest tissue regions (magnified from a cross-sectional image sequence throughout the cardiac cycle).

The resulting PW tracing shows the tissue excursion with the typical two-peak diastolic and the systolic (E’ and A’. and timing of the mechanical segment of the cardiac cycle can be measured from the PW tracing. the foramen ovale flap and how the septum moves with the free wall of the right ventricle. Dynamic colorization refers to the differential coloring of fore.8 Longitudinal strain in the fetus measured using speckle tracking in an apical four-chamber view. The graph on the right shows circumferential strain over time during one cardiac cycle. (Reproduced with permission from reference 28. The possibilities of volume ultrasound.) Video clip 7. the main factor likely to improve the detection rate of fetal cardiac defects remains the operator’s ability to obtain diagnostic views. By adjusting the center of the volume in the middle panel and selecting the appropriate distance between the sections several relevant planes can be displayed dynamically and at once (top left panel: short axis view displaying the “slice” levels).1 Dynamic colorization. are becoming available now.Technical advances in fetal echocardiography 97 ultrasound sensors – such as capacitive micromachined ultrasound transducers (CMUTs) (Figure 7. Especially offline image or volume analysis methods and novel ways of analyzing and presenting fetal echocardiography data will improve the general level of scanning techniques and personal expertise. The top left panel (A plane) is the acquisition plane.6 Color tissue Doppler of a normal fetal heart at mid-trimester.2 High-resolution real-time three-dimensional view of a beating normal fetal heart (22 weeks’ gestation). The next breakthrough in this sense will be the introduction of full aperture matrix transducers ideally suited for transabdominal scanning. Video clip 7. . highest-frequency transvaginal probes.) Video clip 7. adjusted to display the four-chamber view. together with the general trend toward earlier prenatal diagnosis. In this example of a real-time 3D study of a normal fetus at 16 weeks’ gestation the fetal legs and umbilical cord in the foreground are colored in amber while the structures further away from the viewer are shaded in a blue.) Video clip 7. the top right panel B the short axis view of the ventricles and the bottom left panel C displays the interventricular septum en face. The wall filter and pulse repetition frequencies as well as the gain are lowered and the sweep speed increased.and background objects in threedimensional rendering.) Legends for the DVD Video clip 7. (Reproduced with permission from reference 28. in particular together with offline processing. right panel. By reducing the wall filter.4 Tomographic imaging of a virtual cardiac cycle (from a STIC acquisition). instantaneous or real-time threedimensional scanning with sufficiently high spatial and temporal resolution is possible. automatic assessment of relaxation and contraction.5 Tomographic imaging of a normal heart (left panel) and a heart with d-transposition of the great arteries (TGA. The graph on the right shows longitudinal strain over time during one cardiac cycle. movement away from transducer. (Reproduced with permission from reference 28. (Reproduced with permission from reference 84.9 Application of speckle tracking to measure circumferential strain of the fetal right ventricle in a short axis view.19) – which have the agility for controlling frequency and power output93 far beyond what exists today. Using a two-dimensional matrix array transducer. Pushing the limits at another front. It can be expected that these developments will be accompanied by further automation. but continuing advances in array technology will eventually lead to non-reconstructed. Note the synchronicity of the atria and ventricles. Video clip 7. Arrhythmias show distinctly altered typical patterns. Both hearts are displayed using the same settings to demonstrate the striking differences in analogous planes. including detection algorithms that enable tracking of tissue borders. negative velocities) and the systolic peak (S’. have pushed the field forward tremendously. Video clip 7. The colors in the top left panel indicate longitudinal shortening and lengthening (tissue deformation. The colors in the top left panel indicate circumferential shortening and lengthening (tissue deformation.7 Pulsed-wave tissue Doppler study of a normal fetal heart. reduced playback speed). strain). STIC is the clinically dominant 3D modality. towards the transducer. In this video three orthogonal sections through a looping virtual cardiac cycle are shown. Video clip 7. Despite all the technological improvements. Conclusion Recent technological ultrasound advances have increased the image quality available for fetal cardiology. pulse repetition frequency and gain the tissue motion is shown in color. Currently. and their use in fetal echocardiography is imminent. the intersection of planes is denoted by little yellow or red dots. Video clip 7. positive velocity). truly live 3D imaging. strain). A stack of sections parallel to the four-chamber view is displayed from a virtual cardiac cycle (captured using STIC). and automatic volume and mass quantification and flow measurements.3 Multiplanar reconstruction (MPR) of a virtual cardiac cycle acquired using spatial-temporal image correlation (STIC). The fetal heart is imaged in an apical four chamber view and the PW sample volume placed to cover the excursion of the lateral part of the tricuspid valve annulus.

15: 1–9. Stanziano A et al. J Ultrasound Med 1996. Kuhling-von Kaisenberg H. compound. Vincoff NS. 31. Fetal transabdominal anatomy scanning using standard views at 11 to 14 weeks’ gestation. Shapiro RS. Winborn RC et al. 31: 51–9. Bonnet D. 9. Trahey GE. 28. Tutschek B. The RADIUS Study Group. Bollmann R. Porter BA. Quantitative assessment of fetal bowel echogenicity: comparison of harmonic. Floreby L. gehealthcare. Grenier N. Sklansky M. 27: 252–65. 1995: 1353–6. Schmitt KJ. Circulation 2001. Crimmins TR.pdf. 4. analysis and display. Crane JP. 29: 1–5. Carvalho JS. Harmonic imaging in fetal echocardiography. Ultrasound Med Biol 1999. Real-time spatial compound imaging: application to breast. Butera G et al. accessed June 2007. 103: 1269–73. Semin Ultrasound CT MR 2001. J Am Soc Echocardiogr 2001. Improved surgical outcome after fetal diagnosis of hypoplastic left heart syndrome. Three-dimensional echocardiography for studies of the fetal heart: present status and future directions. 29: 81–95. 14. Eik-Nes SH. Sillesen HH et al.contextvision. Detection of transposition of the great arteries in fetuses reduces neonatal morbidity and mortality. Categorial Course in Diagnostic Radiology Physics: CT and US Cross-sectional Imaging.98 Fetal Cardiology References 1. Results of a 2 1/2-year study in the South East Thames Region. A pilot study of its applicability. Screening for congenital heart disease. accessed June 2007. Ultrason Imaging 2004. Reinehr T. Yagel S. Oak Brook. Sharland GK. Callen PW.a new method of vascular imaging in prenatal medicine. Williams W. Campobasso G. Fetal imaging: a brief history of the future. 99: 220–5. Eik-Nes SH. Clinical evaluation of combined spatial compounding and adaptive imaging in breast tissue. vascular. Becker D. In: Goldman LW. 3. Ultrasound Obstet Gynecol 2007. Averkiou M. Tworetzky W. 26: 203–16. 29. 20. A randomized trial of prenatal ultrasonographic screening: impact on the detection. 27. 31: 302–7. Abdul-Khaliq H et al. Fortschr Röntgenstr 2007. . Pooh RK. Am J Obstet Gynecol 2000. Paladini D. Parsons RB et al. Mavrides E. 16. LeFevre ML. Seattle. Br J Obstet Gynaecol 1992. 21. The examiner’s ultrasound experience has a significant impact on the detection rate of congenital heart defects at the second-trimester fetal examination. Benefits associated with harmonic tissue imaging in the obstetric patient. Lewin MB. 15. A model based approach to improve the performance of the geometric filtering speckle reduction algorithm. 2000: 77. Shapiro I. J Ultrasound Med 1999. McElhinney DB. Wolfe HM. Ultrasound Obstet Gynecol 2006. 5. Shinebourne EA et al. Clinical use of ultrasound tissue harmonic imaging. 2005. The role of tissue harmonic imaging in fetal echocardiography. Blaas HGK. 28: 8–14. Heart 2002. 42: 17–27. Dudwiesus H. Photopic Ultrasound Imaging – Ein adaptives Verfahren basierend auf unserer Sehphysiologie. 15: 163. Tranquart F. [Prospective comparison of different ultrasound modalities to measure thicknesses less than 1 mm]. New application of B-flow sono-angiography in perinatology. Heling KS. Sutcliffe P. 19. Contrast and tissue harmonics imaging. Immediate and long-term outcomes in children with prenatal diagnosis of selected isolated congenital heart defects. Valsky DV. 22. 25: 889–94. Ishrak SO. Sahn DJ. 11. Tegnander E. Vassallo M. 171: 392–9. AJR Am J Roentgenol 1998. management. Cohen S. 6. Ultraschall Med 2004. 26. Busse LJ. GE Medical Systems.com. Goldstein RB. and fundamental sonographic images. IL: RSNA. Simpson LL. Muller H. 22: 50–64. 23: 159–64. Obstet Gynecol Clin North Am 2004. Fowlkes JB. Breitkopt P. 2. 7. 68: 59–64. Cardiol Clin 2007. Screening for congenital heart disease prenatally. and outcome of anomalous fetuses. 25: 280–4. Dahl JJ. 29: 38–43. 18: 799–803. 171: 1203–6. Fuchs IB. Electromedica 2000. www. Treadwell MC. Bleck J. 30. Ultrasound Obstet Gynecol 2007. Zador I. Pourcelot L. Presented at 1995 IEEE Ultrasonics Symposium. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Three-dimensional echocardiographic evaluation of fetal heart anatomy and function: acquisition. WA. 32. Altman CA. Tissue harmonic imaging sonography: evaluation of image quality compared with conventional sonography. eds. Cho JY. Fritzer E et al. Volpe P. 18. 17. GOPView™ US-Ultrasound image enhancement in real-time. Fowlkes JB. Goyert GL. Ultrasound Obstet Gynecol 2006. marketing versus clinical reality. Fienup JR. Circulation 1999. 25. 2003. 25: 341–55. [in German] Gebel M. 13. Milkowski A. Tartaglione A et al. Entrekin RR. Novel application of 4D sonography with B-flow imaging and spatio-temporal image correlation (STIC) in the assessment of the anatomy of pulmonary arteries in fetuses with pulmonary atresia and ventricular septal defect. Ultrasonic biomedical technology. 88: 387–91. Am J Obstet Gynecol 1994. Kovalchin JP. 99: 916–18. Ultrasonics 2004. Chaoui R. Hagen-Ansert S. Bezold LI. Ayres NA. 24. Reddy VM et al. Nelson TR. Smith-Bindman R. 33. Johansen OJ. Seubert DE. Ultrasound Obstet Gynecol 2000. 7–10 November. Am J Obstet Gynecol 2005. 192: 535–42. Pretorius DH. www. Advanced dynamic flow . J Clin Ultrasound 2003. Lee HJ. Ville Y. Wagreish J. 95CH35844. 14: 1025–9. Speckle reduction imaging. 12. Ultrasound Obstet Gynecol 2004. Soo MS. von Kaisenberg CS. Tegnander E. Improving the effectiveness of routine prenatal screening for major congenital heart defects. Prenatal detection of heart defects in a nonselected population of 30 149 fetuses – detection rates and outcome. 8. Ultrasound Obstet Gynecol 2007. Ultrasound Obstet Gynecol 2006. 182: 1620–2. 179: 65–71. Coltri A.com/euen/ultrasound/docs/education/ whitepapers/whitepaper_SRI. 28: 40–6. Effect of ultrasound transducer frequency on the appearance of the fetal bowel. Allan LD. and musculoskeletal ultrasound. Yadong L. 23. 10. Eder V. Wunsch R. Forsberg L.

Romero R. The role of spatio-temporal image correlation (STIC) with tomographic ultrasound imaging (TUI) in the sequential 50. Detailed assessment of fetal ventricular septal defect with 4D color Doppler ultrasound using spatio-temporal image correlation technology. Romero R. 49. Paladini D. Schmidt S. Am J Obstet Gynecol 2006. 24: 696–8. Pretorius DH. Yagel S. 61. 64. Wong PC. An automated approach to visualize standard views of the fetal heart. 36. Espinoza J. A new approach to fetal echocardiography. 40. 27: 555–61. 48. 23: 283–9. Inversion mode spatiotemporal image correlation (STIC) echocardiography in three-dimensional rendering of fetal ventricular septal defects. Automated multiplanar imaging: a novel approach to ultrasonography. 18: 177–84. 34: 39–55. DeVore GR. Validation of volume and mass assessments for human fetal heart imaging by 4-dimensional spatiotemporal image correlation echocardiography: in vitro balloon model experiments. DeVore GR. Sklansky MS. Espinoza J. 44. Bhat AH. Rendering in fetal cardiac scanning: the intracardiac septa and the coronal atrioventricular valve planes. Prenat Diagn 2004. Cera E. Vassallo M. Fetal ventricular mass determination on threedimensional echocardiography: studies in normal fetuses and validation experiments. 28: 26–31. 25: 947–56. Sklansky M. J Ultrasound Med 2006.net/xtcommerce/ pdfs/300-06-U026E. A novel algorithm for comprehensive fetal echocardiography using 4 dimensional ultrasonography and tomographic imaging. Digital casts of the fetal cardiac chambers and great vessels for detection of congenital heart disease. Sglavo G. Giuliano A. Ultrasound Obstet Gynecol 2005. Riccabona M. 25: 307–21. 24: 201–7. Martinelli P. accessed June 2007. Romero R et al. Messing B. J Ultrasound Med 2001. Falkensammer P. Three-dimensional ultrasound display modalities in obstetrics. Ultrasound Obstet Gynecol 2005. Fetal heart volume assessment by three-dimensional ultrasound. Cohen SM. Corbett V. Bhat AH. Espinoza J et al. 62. Three-dimensional quantitative echocardiographic assessment of ventricular volume in healthy human fetuses and in fetuses with congenital heart disease. 58. Ultrasound Obstet Gynecol 2006. Three-dimensional power Doppler (3DPD) ultrasound in the diagnosis and follow-up of fetal vascular anomalies. Lee W. 23: 1151–9. 9: 42–8. Liu R et al. J Ultrasound Med 2004.and four-dimensional ultrasonography and ‘inversion mode’. 42. Heling KS. Prenatal diagnosis of pulmonary sequestration using three-dimensional power Doppler ultrasound. 25: 97–8. 23: 535–45. Advances in fetal cardiac imaging. Examination of the fetal heart by four-dimensional (4D) ultrasound with spatiotemporal image correlation (STIC). Espinoza J. Abuhamad A. Sciaky-Tamir Y. Ultrasound Obstet Gynecol 2004. Espinoza J. Hochner-Celnikier D et al. J Ultrasound Med 2004. 25: 128–33. Ultrasound Obstet Gynecol 2004. 39. VCAD Volume-aided diagnosis. 20: 317–27. Ko HC et al. Goncalves LF et al.geultrasound. J Ultrasound Med 2005. 59. Ruano R. 52. Current applications of fetal cardiac imaging technology. 51. Budorick NE.Technical advances in fetal echocardiography 99 34. 3D inversion mode combined with STIC: a novel technique for fetal heart ventricle volumetry in congenital heart disease. Espinoza J. Goncalves LF. Ultrasound Obstet Gynecol 2006. Lee W et al. Goncalves LF. 24: 415–24. 194: 274–81. Sholler GF. 35. Chang FM. Deng J. Pacheco V. 63. Ultrasound Obstet Gynecol 2006. 37. J Ultrasound Med 2004. Curr Opin Obstet Gynecol 2006. Romero R. Espinoza J. Ghi T. Abuhamad AZ. Circulation 2004. Falkensammer P. Three and four-dimensional reconstruction of the aortic and ductal arches using inversion mode: a new rendering algorithm for visualization of fluid-filled anatomical structures. Ultrasound Obstet Gynecol 2005. Goncalves LF. Valsky DV. 55. Cohen SM et al. analysis of fetal congenital heart disease. http://brochureorder. 23: 573–6. 28: 359–411. Inversion mode: a new volume analysis tool for 3-dimensional ultrasonography.pdf. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatio-temporal image correlation (STIC). Goncalves LF. Spatio-temporal image correlation (STIC): new technology for evaluation of the fetal heart. 26: 679–86. . Goncalves LF. Hoffmann J. Nelson TR. Ultrasound Obstet Gynecol 2004. Fourdimensional ultrasonography of the fetal heart using color Doppler spatiotemporal image correlation. Espinoza J. Lee W. Ultrasound Obstet Gynecol 1997. 56. 23: 473–81. Lapadula C. Reichartseder F. Ultrasound Obstet Gynecol 2006. 22: 380–7. Cole A. 28 (Suppl): 397. Meyer-Wittkopf M. Automated sonography: defining the spatial relationships of standard diagnostic fetal cardiac planes in the second trimester of pregnancy. J Perinat Med 2006. Romero R. Benachi A. Bonnet D et al. Yagel S. Segata M et al. Lee W Mazor M. Ultrasound Obstet Gynecol 2006. 24: 72–82. Benachi A. 53. 43. 27: 336–48. Kusanovic JP. Hsu KF. J Ultrasound Med 2005. Cooper SG. Carpenter N et al. 24: 1092–103. 57. Austria. Four-dimensional ultrasonography of the fetal heart using a novel Tomographic Ultrasound Imaging display. Fetal atrioventricular valve junction in normal fetuses and in fetuses with complete atrioventricular septal defect assessed by 4D volume rendering. 54. 47. 38. DeVore GR. Messing B. 41. 45. New fetal cardiac imaging techniques. 46. Vinals F. Real-time 3-dimensional fetal echocardiography with an instantaneous volume-rendered display: early description and pictorial essay. Lee W. Polanco B. Valsky DV. A novel method to improve prenatal diagnosis of abnormal systemic venous connections using three. Johnson D. Sklansky MS. Sklansky MS. 110: 1054–60. 60. Rodeck CH. Pediatr Cardiol 2004. Ultrasound Obstet Gynecol 2005. 25: 428–34. Ultrasound Obstet Gynecol 2006. 25: 157–67. Mazor M. Chaoui R. Deng J. Ultrasound Obstet Gynecol 2003. The ‘spin’ technique: a new method for examination of the fetal outflow tracts using three-dimensional ultrasound. Goncalves LF. Falkensammer P. GE Healthcare. J Ultrasound Med 2004. 28: 266–74. Rodeck CH. Goncalves LF. Aubry MC et al. Platt LD. Romero R. Zipf. J Clin Ultrasound 1997. Platt LD. Corbett VN.

87. 89. Chikada M et al. Fiering TA. 66. 24: 1081–91.100 Fetal Cardiology 65. Rein AJJT. 90. Tissue Doppler imaging of the fetal heart. Bender HG. Singh V et al. Buck T. J Cardiovasc Diagn Procedures 1998. Larsen LU. 48: 2012–25. Tutschek B. Stetten G. Tutschek B. Yagel S. Teodoro A et al. 25: 370–5. 18: 745–52. Prenat Diagn 2005. 68. Sklansky MS. Nir A. Eur J Echocardiography 2000. Tsuda A. Use of high-frame rate imaging and Doppler tissue echocardiography in the diagnosis of fetal ventricular tachycardia. 31: 1159–62. Jamal F et al. threedimensional echocardiography: feasibility of dynamic right ventricular volume measurement with saline contrast. Harada K. Ultrason Imag 1998. 81. 24: 229. Application of free-hand three-dimensional echocardiography in the evaluation of fetal cardiac ejection fraction: a preliminary study. Real-time 3D ultrasound: a new look at the heart. 77. 137: 958–66. Ultrasound Obtet Gynecol 2007. 86. Buck T. Ultrasound Obstet Gynecol 2004. 27: 210–13. Utility of tissue Doppler echocardiography (TDE) in analysing fetal arrhythmias. Norrild K et al. Circulation 1998. Strain rate imaging. Oralkan O. Heimdal A. J Am Soc Echocardiogr 2001. Ushakov FB. Tutschek B. 16: 128–32. 53: 1202–11. Fetal cardiac ventricular volumes derived from 3D/4D echo: definitive data from two different types of 3D ultrasound systems. Acar P. Lamberti A. Steinborn A et al. Paladini D et al. 82. 92. Levine JC. Schmidt KG. Regional strain and strain rate measurements by cardiac ultrasound: principles. . Tuschek B. Strachan M et al. Ultrasound Obstet Gynecol 1999. J Am Coll Cardiol 2007. 88. Ultrasound Obstet Gynecol 2004. 20: 235. Myocardial motion imaging: a new application of power color flow and frequency-based color flow 80. 74. Echocardiographic evaluation of cardiac function in the fetus. Cohen SM. 3D real-time imaging of the fetal heart. Welson T.and C-scan images with the use of real-time 3-dimensional echocardiography: studies in an in vitro model. 79. O’Donnell C. Wygant IO. Imbar T. Ultrasound Obstet Gynecol 2000. Real-time. Fetal cardiac ventricle volumetry in the second half of gestation assessed by 4D ultrasound using STIC combined with inversion mode. Smith SW. 16: 530–5.no/ stoylen/strainrate/. Twining P. Assessment of left ventricular function by cardiac ultrasound. Tissue Doppler imaging in the normal fetus. Ultrasound Obstet Gynecol 2003. Fleishman CE. Maulik D. 71. 13: 255–9. 85. Strain rate derived from color Doppler myocardial imaging for assessment of fetal cardiac function. Stoylen A. 91. Hui L. Davies CH et al. 3-D ultrasound imaging using a forward-looking CMUT ring array for intravascular/ intracardiac applications. Rein AJJT. Real-time three-dimensional fetal echocardiography using matrix probe. Fetal Diagn Ther 2000. 75. Mori Y. Geva T et al. 97: 1897–900. O’Donnell M. Hui L. Russo MG et al. 14: 149–51.ntnu. 106: 1827–33. Valsky DV et al. 69. J Ultrasound Med 1999. 15: 267–74. Russo MG. Popovic ZB. 70. Comparison of ventricular volume and mass measurements from B. Orino T et al. IEEE Trans Ultrason Ferroelect Freq Contr 2006. Esh-Broder E. Rusk RA. Ultrasound Obstet Gynecol 2000. Pacileo G. Scharf A. Shiota T. Zimmermann T. Ota T. Yeh DT. Am Heart J 1999. Ohazama C et al. Tutschek B. Circulation 2002. 15: 73–84. Ultrasound Obstet Gynecol 2006. Real-time three-dimensional fetal echocardiography: initial feasibility study. 23: 546–51. Strub M et al. implementation and limitations. Use of tissue velocity imaging in the diagnosis of fetal cardiac arrhythmias. JACC 2006. Simpson J. Prenat Diagn 2004. Real-time threedimensional fetal echocardiography. Paladini D. Quantification of regional left and right ventricular longitudinal function in 75 normal fetuses using ultrasound-based strain rate and strain imaging. 30: 142–51. Real-time threedimensional echocardiography for determining right ventricular stroke volume in an animal model of chronic right ventricular volume overload. J Am Coll Cardiol 2007. 1: 154–70. Thomas JD. Hruschka TA. Fetal tissue Doppler echocardiography: detection rates of cardiac structures and quantitative assessment of the fetal heart. 16: 53–4. 83. Taktak A et al. Ultrasound Med Biol 2005. 49(Suppl A): 254A. Reihs T et al. Live three-dimensional echocardiography of the human fetus. 84. Ota T. Echocardiography 2003. 21: 26–32. Doppler in fetal echocardiography. 93. 71: 227–34. 73. Fetal cardiac ventricular volumes derived from 3D/4D echo: definitive data from two different types of 3D ultrasound systems. Khuri-Yakub BT. Int J Cardiol 1999. J Am Soc Echocardiogr 2000. 72. 67. Davidson JO. Geka F. Di Salvo G. Ultrasound Obstet Gynecol 2000. Progress in 2-D arrays for real-time volumetric imaging. Jones M. Robertson PA et al. Robertson PA et al. Light ED. 49 (Suppl A): 254A. Paladini D. Nanda NC. Dulac Y. Messing B. D’hooge J. Petersen OB. Echocardiographic assessment of ventricular filling pressure during the second and third trimesters of gestation. 20: 715–21. http://folk. 76. 13: 910–17. accessed June 2007. 78.

so that the right ventricle moves to the right side of the left ventricle and the normal asymmetry of the heart is initiated. As defined. pathogenesis.1. the more its malfunction will affect major cardiac architecture.2 Cardiovascular development involves a series of complex processes. such as double inlet left ventricle (single ventricle).3–5 Many of these have been found in studies of chick and mouse heart development. lefty. and hypertrophic or dilated cardiomyopathy. and the movement of the atrioventricular and semilunar valve rings. CHD is one of the most common serious congenital anomalies. and in an even higher percentage of fetuses.14–16 The proteins that drive the cilia – kinesin and dynein – are subject to mutations that can alter normal asymmetry. The two cell populations fuse early in development. In general. l-loops. some have been identified in humans. the left ventricle. occurring in up to 2% of liveborn children. Anterior to these cells are myocardial precursors that form the secondary or anterior heart field. or abnormal connections between great vessels and the heart. although these usually present later in childhood or adulthood. situs abnormalities (heterotaxies). studies have revealed that the earliest asymmetry occurs in Hensen’s node. A few children are born with arrhythmias (mainly conduction defects). but vast numbers are yet to be discovered. and is produced by the unidirectional swirling of the monocilia that dip from the nodal cells into the extracellular embryonic fluid. zic3). and truncus arteriosus. These changes allow the right and left atria to connect to the right and left ventricles.18 After normal d-looping has occurred. nodal. double outlet right ventricle. and the ventricles to communicate with their respective great arteries.12.17. respectively. and other abnormalities of these cushions lead to the relatively gross distortions of architecture found in atrioventricular septal (endocardial cushion) defects. fibroblastic growth factor 8. so that their malfunction produces gross structural anomalies indicating developmental arrest at a primitive stage.8 Epidemiology of congenital heart disease: etiology. .13 This fused heart precursor begins as a straight tube that contains in sequence from the caudal end segments that will develop into the atria. cardiovascular disease present at birth. and this is connected to the right ventricle which leads to the truncus arteriosus. the future atria are still connected to the left ventricle. Recently. Even though asphyxial heart disease is present at birth. the precursor of the future aorta and main pulmonary artery. Most CHD is due to gross structural developmental cardiovascular anomalies such as septal defects. but remain clonally distinct. This asymmetrical rotation affects some early expressed genes (for example. each component of which has to occur at the right time under the orchestration of a cascade of genes and gene products. stenosis or atresia of valves.12. and the truncus arteriosus. by definition. faulty expression of these genes will produce dextrocardias. The next major developmental changes are the formation of the primary atrial and ventricular septa. The tube elongates and twists into a d-loop. and complex heart disease.6–11 The mammalian heart develops partly from the primary heart tube formed by myocardial and endothelial cells that have differentiated from splanchnic mesoderm and will form the atria and left ventricle. The formation of the heart begins with the differentiation of two specialized groups of cells – the primary and secondary heart fields. and incidence Julien IE Hoffman Introduction Congenital heart disease (CHD) is. the further upstream (closer to initiation of development) a gene is. Integration of the secondary and primary fields depends in part on the influence of cells in the neural crest. it is not included as CHD. and cells from this field develop into the outflow tract and right ventricle. the right ventricle. hypoplasia or absence of one or other ventricle.19 These abnormalities are in part related to abnormal function of the endocardial cushions. Coordinating these changes are genes involved in cell migration and the formation of the extracellular matrix.

28 This makes screening for genetic causes of CHD complicated and expensive. so that some mutations might be expected to be more common than others. and 13 and the sex polysomies (e. and abnormalities of the radius or thumb. Many gene mutations have been found in CHD. Therefore. and monosomy X that is inversely related to maternal age. and there is no reason why one particular gene should always be involved.4% (median 4. Genetic abnormalities Chromosomal abnormalities involve excesses or deficiencies of multiple genes.29–31 These are not usually regarded as syndromes. patients with one congenital abnormality often have abnormalities of other systems.g. in first trimester fetuses aborted for non-medical reasons. the Holt–Oram syndrome. Infants with these abnormalities are survivors of a much larger cohort of fetuses with chromosomal defects. This figure is likely to grow rapidly. ranging from 15. and for X monosomy.1 These and other less common abnormalities are listed in Table 8. and hypoplastic left heart syndrome. and function early in development to affect several systems. Nevertheless. it would not be surprising to find that non-specific factors and random errors in cell migration produce various defects or influence the severity of others.67%) of all live births are associated with chromosomal defects. Therefore. and some of these may act by interfering with the action of specific genes. In addition. XXX and XYY) that increase with maternal age. nor should we expect them to be. these are usually the aneuploidies: trisomy 21. and function later in development to affect only one organ or part of one organ. are specific or ‘downstream’. due to mutations in the Tbx5 transcription factor. Some gene mutations characteristically cause defects in more than one organ system and produce recognizable syndromes. CHD in liveborn infants. Among spontaneous abortions.5%) have chromosomal defects.7 to 69. however.102 Fetal Cardiology Given the importance of precisely orchestrated spatial and temporal gene expression in cardiovascular development. 2.3.24 Other genes. About 0. The percentage of aborted fetuses with chromosomal anomalies is greatest in the youngest fetuses. It is about 1% for trisomy 21. there are gene mutations with major effects that are modified by other genes with subtle differences in timing or level of expression that will influence the phenotype. and decreases to 6–15. fortunately.27% (median 0.2). For example. single gene defects are also common. usually not as part of a general syndrome. multiple genotypic abnormalities may converge on a similar phenotype.1. It is likely that the genes responsible for these syndromes are general or ‘upstream’. is associated with left-sided defects including septal defects. In addition. sex polysomies do not have an increased incidence of CHD. and 13. shows characteristic body habitus and facies as well as dysplastic pulmonary valves and hypertrophic cardiomyopathy. 18. the incidence of chromosomal abnormalities is much higher. or isolated point mutations. As a group. Almost certainly. there are also well known environmental causes of CHD. abnormalities of the genitourinary tract occur in about 30% of patients with congenital heart disease. only about 15% of fetuses with chromosomal defects survive to birth. a more detailed list is published by Burn. the earlier the echocardiographic study is done. Finally. The chances of survival to birth depend on the defect. atrioventricular septal defects. it is likely that any given defect can arise from mutations in one of many genes. On the other hand.22 The Noonan syndrome. For example. about half due to a mutation of the PDPN11 gene on chromosome 12.30–2. In general.7%) beyond 20 weeks’ gestational age. These considerations indicate the desirability of karyotyping all early fetuses found to have CHD.23 Some of the better known genetic syndromes with CHD are listed in Table 8.5 None of these genetic defects to date are very frequent. with the complexity of cardiovascular development.6%). chromosomal defects account for about 6% of all . microdeletions (both of which affect many genes). One other aspect of chromosomal anomalies is important for the echocardiographer. These abnormalities may take the form of chromosomal rearrangements.6–6. Because all these chromosomal defects except for sex polysomies have a very high association with CHD (Table 8.8% (median 46. and patients with tetralogy of Fallot often have omphaloceles.9% (median 11. and it is not Chromosomal defects Chromosomal defects vary from gross abnormalities such as trisomies or monosomies to deletion syndromes and microdeletions that involve contiguous genes. 18. Mutation in any one of the members of the cascade would cause a specific cardiac defect. some genes will be more susceptible to mutations than will other genes in the cascade.21 but this figure could change dramatically depending on increased survival of the affected fetuses on the one hand or the frequency of therapeutic abortion on the other.1). In the development of any part of the cardiovascular system there is usually a cascade of genes and gene products involved. the greater is the chance of finding a chromosomal abnormality in a fetus with CHD. the incidence of CHD in general depends on how many of these fetuses are conceived by older mothers and how many of the affected fetuses reach term alive. but as high as 80% for the sex polysomies (Table 8. and currently account for at least 3% of all CHD. it is not surprising that specific genetic abnormalities are being found in increasing numbers in CHD.

PDA. ASD AS. total anomalous pulmonary venous connection. patent ductus arteriosus. truncus arteriosus. ventricular septal defect. ToF High 25 Supravalvar AS. coarctation. . Based on reference 20 Incidence/1000 live births Percentage with CHD Predominant types of CHD Chromosomal defect Trisomies 21 (Down syndrome) 18 (Edward syndrome) 13 (Patau syndrome) Duplications 3q26–27 duplication (Cornelia de Lange) 4p 5p (Opitz) 8 9p 10q 11p 12p (Pallister–Killian. PPS PDA. VSD. ToF Variable VSD. peripheral pulmonic stenosis.3 (Rubinstein–Taybi) 1–1. AS. VSD. ToF. ASD. ASD. pulmonic stenosis (valvar). PDA VSD. tricuspid atresia.3 0. VSD VSD.2 50–60 95 90 AVSD. PPS. ToF 0. PS 50 60 30 30–50 60 50 10 Low High PPS Aortic arch anomalies.2 50 Coarctation of the aorta.1 Chromosomal defects and congenital heart disease (CHD). interrupted aortic arch. PPS. ASD. Fryn) 22p duplication (cat eye) Monosomy X (Turner) Deletion syndromes 4p− (Wolf–Hirshhorn) 4q− 5p− (cri-du-chat) 9p− (CHARGE) 11q− (Jacobsen) 13q− 18p− 18q− 20p11− (Alagille) Microdeletion syndromes 22q11 (DiGeorge: CATCH-22). PDA. TAPVC.5 0.23 (Williams) 16p13. AS. Shprintzen (velocardiofacial) 7q11. PDA. ASD VSD 10–15 10 20 Low 50 Low 25 VSD VSD+. aortic stenosis or bicuspid aortic valve ASD VSD. coarctation.1–0. AVSD.Epidemiology of congenital heart disease 103 Table 8. atrial septal defect (secundum). absent pericardium TAPVC.2–0. tetralogy of Fallot. atrioventricular septal defect (endocardial cushion or atrioventricular canal defect). PS. PDA.1–0. aortic stenosis.

1.4.3 16p3.4 107 7601 7708 1. QT3–3p21.3 3. metavinculin.5. Based on data from references 24–27.3 CHD caused by gene mutations (syndromic and non-syndromic). 5q33. 5q1–13. 11p15 ventricle Osler–Rendu–Weber Long QT 9q34. AVSD. phospholamban.38 127 31 158 80.104 Fetal Cardiology Table 8. actin. caveolin. 3p25.3 1q32.1. 15q14. 12p12. tafazzin. 11p11. 19q13. 12q11 QT1–11p15. PS ASD+.2. 3p. α-actinin. peripheral pulmonic stenosis. VSD.1–12. 15q14. AVSD Dilatation and rupture of aorta. cardiac troponin T.2. PDA. In many of these disorders the specific gene mutations have been identified Syndrome Noonan Apert Holt–Oram Ellis–van Creveld Marfan Ehlers–Danlos type IV Pseudoxanthoma elasticum Hypertrophic cardiomyopathy Chromosome 12q24. 10q22–q23. PPS AS. PPS. and myosin binding protein C Osteogenesis imperfecta 17. 18q12. 15q22.71 52. 6q22. desmin. 2q31. restrictive cardiomyopathy Aortic or mitral incompetence. Xq27. coarctation of aorta. 2q24. muscle LIM protein. 7 Mucopolysaccharidoses 4p16. sudden death Supravalvar AS Supravalvar AS.2.1 10q26 12q24. 1q42–43. Dilated (congestive) cardiomyopathy due mainly to abnormal cytoskeletal but also sarcomeric proteins: dystrophin.23 Cardiomyopathy Pulmonary arteriovenous fistulae Long QT interval.4 Table 8. ASD. aortic stenosis. QT6–21q22.1 4p16 15q21 2q31–32.1. 2q35.3. tropomyosin. truncus arteriosus Single atrium. lamin A/C.5 38 330 368 10.1. aortic or mitral incompetence MVP. titin. 10q212. MVP. aortic or mitral regurgitation Arterial rupture Aortic root dilatation.3 Dilated cardiomyopathy Xp21. 11p11.1.1. PS.2.3. atrioventricular septal defect (endocardial cushion or atrioventricular canal defect). α-tropomyosin Non-compaction of left Xq28. QT4–4q25–27. also 5q31. 1q32.3–q23. 15q22 protein C. mitral valve prolapse.0 205 489 694 29. myosin binding 11p15. pulmonic stenosis (valvar). AVSD. patent ductus arteriosus. QT2–12p11.1 CHD Pulmonic stenosis. 20q13. cardiac actin. coronary arterial disease. 12q23–24.3. Reproduced with permission from reference 1 Group Aneuploidy Sex chromosomes Autosomes Polyploidy Euploidy Miscellaneous Percentage total Total pregnancies aberrations 1 2848 2849 0.2 Survival of fetuses with chromosomal defects. Xq28.4 70 000 30 000 1 00 000 Monosomy Polysomy Trisomy Liveborn Spontaneous abortion Total Percentage survival 17 4434 4451 0. coronary artery narrowing Asymmetric ventricular hypertrophy due to a large variety of mutations in sarcomeric proteins: myosin heavy and light chains. δ-sarcoglycan. 14q12. hypertrophic cardiomyopathy VSD.1 0.1. troponin. . arrhythmias. titin. VSD. β-myosin heavy chain. 14q12. QT5–21q22. 1q21.1 7q11. 7q31. atrial septal defect (secundum). ventricular septal defect.

ASD. septal defects. TAPVC. random migration. With abnormally increased adhesiveness. SV. Much has been written about multifactorial inheritance in CHD24. DA. ASD. complete transposition of the great arteries. and 38 Environmental factor Rubella virus Mumps Lithium CHD PDA. could explain why CHD could be genetically caused without manifesting classical Mendelian genetic frequencies. single ventricle.34 The concept is that certain lesions may be due to the interaction of several genes (polygenic inheritance). PS. then. the atrioventricular canal region developed normally in their model.36 During endocardial cushion development.33. ASD. there is an increased risk for CHD in the children. TAPVC 5–10 AS. just as in an atrioventricular canal defect. double outlet right ventricles. ventricular septal defect. if one parent (especially the mother) has had CHD. some of the atrioventricular canals were abnormally formed.4 Environmental causes of CHD. AVSD.12 the development of the aorticopulmonary and truncal septa depends on the migration into the embryonic heart of cells from the cranial neural crest. The quintessential lesion related to endocardial cushion defects. This combination of events. for example. Ebstein syndrome. but its expression depended in part upon random events in cell migration. TGA. the atrioventricular septal defect. PPS. VSD. Thus. atrioventricular septal defect (endocardial cushion or atrioventricular canal defect).24 Recurrence risk to sibs and offspring is approximately the square root of the population incidence. how do we explain the other forms of CHD? One possible answer lies in experiments reported by Kurnit and colleagues. PDA. ToF. PPS Truncus arteriosus. An example of a link between environmental and genetic events occurs with retinoic acid and its metabolites.32 and in cardiovascular disease in general. TGA. ASD Outflow tract lesions. coarctation of the aorta Complete heart block PDA. The risk to sibs is comparable to the risk to offspring. a genetic defect itself. there is a high incidence of conotruncal defects such as ventricular septal defects. VSD. Phenylketonuria. but this risk is usually no more than 5–10%. For example. coarctation. PS. 37. cited in reference 24) If this is true. patent ductus arteriosus. Data taken from references 24. Environmental factors CHD has been associated with several environmental toxic or infectious factors (Table 8. especially TGA. total anomalous pulmonary venous connection. hypoplastic left heart syndrome. PS. SV. As Kirby first demonstrated.35.4). ToF. Criteria needed to indicate a multifactorial model have been described by Burn. We do not know whether most of these act by affecting gene expression directly or by blocking the action of the gene product. and certain rules for when migration and cell division would cease. HLH. HLH PS. VSD Endocardial fibroelastosis Mitral and tricuspid incompetence. With normal cell adhesiveness. ToF. If these cells are removed experimentally. coarctation. failure to fit classical Mendelian genetics is not an argument against a genetic cause of CHD. coarctation of the aorta VSD. The only form of CHD shown to fit the multifactorial model well is the patent ductus arteriosus (Zetterqvist. is particularly common in trisomy 21. PPS. pulmonic stenosis (valvar). affects the fetus through the increased maternal blood levels of phenylalanine and phenylpyruvic acid.Epidemiology of congenital heart disease 105 clear whether these associations stem from a common genetic origin or from some other disturbance during fetal development. PDA. peripheral pulmonic stenosis. ASD Trimethadione Phenytoin Systemic lupus erythematosus Coumadin Thalidomide TGA. atrial septal defect (secundum). Studies have shown that in trisomy 21 fibroblasts cultured from the lung have abnormally increased adhesiveness. and truncus arteriosus. Since these neural crest cells also aid in the formation of the pharyngeal arches and pouches . the outcome being modulated by environmental factors. AS. ToF Conotruncal anomalies 25–50 15–30 2–3 20–40 3–5 Frequency of CHD (%) > 35 Diabetes in pregnancy Alcohol Retinoic acid 25–70 Phenylketonuria ToF. What was important in their study was that not all the canals were abnormal. unlike the 50% risk found for autosomal dominant disease or the 25% risk associated with recessive disease. Therefore the abnormality produced in their model was due to the abnormal adhesiveness. Table 8. certain cell adhesion molecules such as platelet endothelial cell adhesion molecule are downregulated when endocardial cells undergo mesenchymal transformation. None of them are known to affect the genome itself. aortic stenosis. tetralogy of Fallot. Kurnit and colleagues developed a computer model of embryological development in which they programmed differing degrees of adhesiveness.

So do certain arrhythmias such as the long QT syndrome. children with very mild lesions such as minimal pulmonic stenosis. Patent ductus arteriosus of prematurity. and trisomy 21 is more frequent in mothers over 35 years old.25. The other is how many pregnancies of fetuses with trisomy 21 will be terminated medically. About 35% of dilated cardiomyopathies are familial and genetically determined. 3. thus reducing the incidence of this lesion at birth.62 The prolapse does not appear to be present at birth.53–55 and those associated with arrhythmogenic right ventricular dysplasia. Another major problem is whether or not the bicuspid aortic valve is included in the series. One of the changes in recent years is the availability of echocardiography for diagnosis. Atrioventricular septal (endocardial cushion canal) defects are disproportionately frequent in children with trisomy 21. or all-trans-retinoic acid produces lesions resembling those found in Kirby’s experiments. Therefore. Although most of these cardiomyopathies are not present in the fetus or even in the neonate. hypocalcemia. and the majority of hypertrophic cardiomyopathies are also due to isolated gene mutations43–49 (Table 8. Here. retinoic acid. These conditions are met in several countries today. and the parents may choose to abort these fetuses. Even if the above criteria are met. many reports do not specifically exclude this form of patent ductus arteriosus.64 In some series65. Even though these mild lesions seldom cause clinical problems. thereby inflating the incidence of the anomaly. the incidence of ventricular septal defects and indeed of all CHD (because ventricular septal defects are the most common form of CHD) depends upon how many of Incidence of congenital heart disease To determine the true incidence of CHD there must be a medical system in which pediatric cardiologists can diagnose CHD accurately and objectively.61 If even a few patients with this lesion are included under the title of aortic stenosis it will alter the apparent incidence of the latter lesion drastically. or small atrial or ventricular septal defects. In communities where this practice occurs we have to take account of these aborted fetuses if we wish to determine the true incidence of CHD. The frequency of this lesion in any series. abnormal facies.56–58 A normal echocardiogram in the face of an abnormal family history of one of these diseases does not exclude the genetic defect.59 Conversely. Many serious forms of CHD are now detected by fetal echocardiography. failing to include them in a . Recent studies of infants in the newborn nursery have shown that as many as 4–5% of them have tiny muscular ventricular septal defects. resulting from 22q 11 deletions) syndrome.60 Unfortunately.66 as many as 50% of pregnancies with fetuses with CHD were electively terminated. cleft palate. is very common. if subjects with bicuspid valves and small pressure gradients across them are not classified as aortic stenosis. Most series do not include prolapse as a congenital lesion. and has been thought to occur in 4–5% of the population. the incidence of aortic stenosis will appear to be smaller than it should be. Not only is this very accurate in experienced hands. 4. and without an autopsy examination the incidence of these serious malformations might be seriously underestimated.3). so that their frequency will also be underestimated.106 Fetal Cardiology (from which the thymus and parathyroid glands are derived) and the aortic arches. A specific diagnosis might not have been made in these infants by the time of death. Two barriers exist. and this may well differ in different societies. the incidence of CHD will be determined accurately only if ascertainment of the disease is complete. giving bis-diamine. phenotypically and perhaps genotypically mimics DiGeorge syndrome. they will inflate their incidence. Alternatively. 5. One is how many older mothers there are in any series. the environmental agent. a maturational disorder rather than a cardiovascular anomaly. might never be included in any cardiologist’s practice.50–52 Brugada syndrome. but it can be applied to children with minimal heart disease who previously would not have had the diagnosis proven because they would not have been submitted to cardiac catheterization. There are several other factors that influence our ability to estimate the true incidence of CHD: 1. 6. Thus. but obviously if even a few with mitral valve prolapse and mitral regurgitation are included under the heading of mitral valve lesions. isotretoin. and in which the whole population has easy access to these cardiologists. Some CHD causes death in the first few days after birth. Some data suggest that the bicuspid aortic valve occurs in 1% (possibly more) of the population.39–42 Certain chemicals are now known to interfere with migration of these neural crest cells. depends on two factors. An unknown number of these may be included in any collected series. thymic hypoplasia.5 It is thus possible that some outflow tract lesions might be due either to genetically or to environmentally determined defects of neural crest cell migration. therefore. About 95% of these defects close by themselves within 6–12 months after birth. this may explain why aortic arch anomalies and truncus arteriosus are so frequently associated with DiGeorge (CATCH-22: cardiac defects. database reduces our ability to study the factors that cause these lesions. Mitral valve prolapse is also common.67 Many of these infants had no murmurs. they do have a genetic origin.63 although this does not exclude a developmental or genetic origin for the prolapse. 2.

6 14. over the last 10 years the incidence of CHD has been rising slowly. the incidence drops to about 1% of live births.2 16.8 1.7 3. the incidence of all forms of CHD might be 5–6% of all live births.8 8.0 0.8 23.0 9.8 4.8 5.4 3.6 0. 8. It should be clear.0 0. rather than to describe its incidence as a proportion of all CHD.7 5.2 0.8 7.6 Median 32.0 2.2 2. Fetal echocardiography has shown that certain lesions.2 1. consider that in a series examined after a year of age that excludes patent ductus arteriosus in premature infants.3 12. respectively.6 1.Epidemiology of congenital heart disease 107 these trivial defects are included in the series.6 25% 27. too.1 2.3 11.0 1.4 3.4 0.4 5.4 2.2 2. This study would give a high value for VSD of 50 000 per million liveborn children.7 5.6 1.9 Lowest 987 60 135 85 160 40 0 176 167 0 0 0 0 51 0 Per million liveborn children 25% 1773 324 372 242 355 161 289 226 302 61 154 105 54 82 52 Median 2829 567 564 340 532 256 356 303 356 94 226 160 85 115 91 75% 4482 782 1059 396 836 388 492 364 577 136 279 224 136 188 120 Highest 6616 2108 2112 791 1155 1425 620 560 633 344 347 347 277 263 155 ∗25% and 75% are the lower and upper quartiles.0 3. In one study.9 14. the bicuspid aortic value occurs in about 10–12 per thousand live births. If they are excluded.72 Studies such as this will then underestimate the incidence of pulmonary stenosis which is often mild.4 10. the investigators specified that they did not include mild pulmonic stenosis with systolic gradients across the pulmonary valve of under 25 mmHg. ∗∗Excludes one study in which 5% of all neonates had a small ventricular septal defect (VSD).9 4.3 0. From these caveats it is clear that it is very difficult to assess accurately the incidence of CHD at birth.8 5.0 14. For example. In terms of the need for medical services this is a very desirable outcome.0 0.2 1.1 75% 42.5 19.6 7.4 1.0 2. .0 0.4 7.6 3. and 550 children with all other forms of CHD.8 5.2 8. that it might be better to describe the absolute incidence of each specific lesion per 1000 or per 100 000 of the population. Data on percentages from reference 73.7 4.3 2. but other forms of CHD were not studied.5 3.2 1.5 Incidence of CHD in liveborn children.4 6.68–71 7. From these data. particularly ventricular septal defects.8 1.5 0. 400 children with a ventricular septal defect. and about 93% of all CHD.5 3.2 6.1 5.8 0.8 5.7 2. and bicuspid aortic valves there are 50 patients with CHD who have tetralogy of Fallot.5 1.8 2.7 3.8 7. the incidence of tetralogy will be 5% and of ventricular septal defects will be 40% of all CHD. data on incidence per million from reference 2 Lesion Percentage of all congenital heart disease∗ Lowest Ventricular septal defect** Patent ductus arteriosus Atrial septal defect Atrioventricular septal defect Pulmonic stenosis Aortic stenosis Coarctation of the aorta Transposition of the great arteries Tetralogy of Fallot Persistent truncus arteriosus Hypoplastic left heart Hypoplastic right heart Double inlet left ventricle Double outlet right ventricle Total anomalous pulmonary venous connection Miscellaneous 16. may be detected in utero but have disappeared at the time of birth.6 1.9 3. In addition to the data shown above.0 0. not because of true increase in all congenital heart lesions but because of an increasing number of patients with small ventricular septal defects who are now being included in the series. mitral valve prolapse.8 3. In fact. If in that same series the investigators decided Table 8. equivalent to the incidence of all other forms of congenital heart disease combined.2 If they are all included. but in terms of understanding frequency and etiology this leads to the underestimation of the incidence of these lesions.0 10.2 0.0 10.3 2.8 5.6 Highest 50.

the ultimate goal should be to prevent CHD or reduce its incidence. McCulley DJ. Somlo S. Cilia propel the embryo in the right direction. Makova S et al. McAnally J et al. a ventricular septal defect is still the most frequent form of CHD. Prenatal incidence. It would be better to focus on the 50 with tetralogy of Fallot and relate them to the more accurately determined numbers of live births. a regulator of right ventricular heart development. 97: 258–70. In: Emery AE. Poelmann RE. Mechanisms of segmentation. 16. De Val S et al. Heart Development. 16: 155–65. 18. References 1.23 the total incidence of CHD is likely to increase slowly. Rasmussen TL. Two populations of node monocilia initiate left-right asymmetry in the mouse. and ventricular septum comprise a restricted expression domain within the secondary/anterior heart field. Brueckner M. 41: 307–19. In: Moller JH. In: Harvey RP. Schneider H. 2000: 3–14. 1997: 767–828. . Hoffman JIE. 101: 339–44. 1999. Pediatr Cardiol 1995. Waldo KL. then the same 50 children with tetralogy of Fallot would be included in a group of 5000 children with CHD (4400 of whom would have ventricular septal defects). Dev Biol 2005. 114: 61–73. Conclusions Worldwide. Rosenthal N.5. Rosenthal N. 3. San Diego: Academic Press. Kaplan S. Casey B. eds. 55: 485–92. Heart Development. San Diego: Academic Press.74 Today. Am J Med Genet 2001. Molecular inroads into the anterior heart field. Meilhac SM et al. Trends Cardiovasc Med 2005.2 Even excluding the studies of neonates with an extremely high percentage of ventricular septal defects. Kelly RG. 15: 51–6. Pediatric Cardiovascular Medicine. 281: 66–77. Normal and abnormal cardiac development. 6. Dev Biol 2005. Heart Development. Phan D. Left-right asymmetry and cardiac looping. Heart Development. In: Harvey RP. This progress comes at considerable economic cost. Mjaadvedt CH. Bristow J. Yamagishi H. 4. Development 2005. 15. furthermore. 1999: 159–77. CHD remains a major cause of morbidity and premature death. BOP. Dev Biol 2005. The incidence of congenital heart disease. 10. 281: 78–90. both as percentages of all CHD and as numbers per million live births. Rosenthal N. because of advances in its treatment. Ward CC et al. Overbeek PA.5 so that it is reasonable that many different ways of interfering with cardiac development will end up with a ventricular septal defect. Kelly RG. Parisi L et al. New York: Churchill Livingstone. Nakagawa O et al. This probably does not carry any great implications. eds. 9. Kirby ML. eds. Because the children of these parents have an increased incidence of CHD. 95: 261–8. 21. 2. Hoffman JIE. Cardiac neural crest is necessary for normal addition of the myocardium to the arterial pole from the secondary heart field. In: Harvey RP. 1999: 391–402. Maeda J. 13.108 Fetal Cardiology to include another 4000 children with small ventricular septal defects present at birth. but a ventricular septal defect is the most frequently observed congenital cardiac defect in subjects with chromosomal and genetic defects. Hutson MR. Acknowledgments I thank Dr Harold Bernstein and Dr James Bristow for valuable advice and criticism. most subjects with CHD survive to become adults and to raise their own families. Cell 2003. The right ventricle. 1999: 479–89. Not only is this true in general. eds. San Diego: Academic Press. Waldo KL. and remodeling of the tubular heart: endocardial cushion fate and cardiac looping. eds. Rimoin DL. septation. Verzi MP. Hutson MR. Contribution of neural crest to heart and vessel morphology. 20. San Diego: Academic Press. eds. Tbx1 is regulated by forkhead proteins in the secondary heart field. The search for genetic mechanisms of congenital heart disease. 12. Edinburgh: Churchill Livingstone. Extracardiac abnormalities in infants with congenital heart disease. Right ventricular myocardium derives from the anterior heart field. Secondary heart field contributes myocardium and smooth muscle to the arterial pole of the developing heart. Goodship J. 1999: 1179–93. Dev Dyn 2006. Harvey RP. Burn J. Heart Development. 19. 7. 14. Am J Med Genet 2000. Wessels A et al. 287: 134–45. outflow tract. Brueckner M. Kosaki K. Therefore. 17. Rosenthal A. Stadt HA et al. CHD is estimated to occur in about 1 500 000 live births annually. eds. Principles and Practice of Medical Genetics. Rosenthal N. Greenwood RD. In: Harvey RP. is a direct transcriptional target of MEF2C in the developing heart. Malumder K. Yamamura H. Of mice and men: dissecting the genetic pathway that controls left-right asymmetry in mice and humans. 11. Rosenthal N. Gittenberger-De Groot AC. generation by generation. for a percentage of 1% tetralogy of Fallot. Circ Res 2004. 132: 2669–78. and this can be accomplished only after we have a better understanding of its genetic and environmental causes. 8. Congenital heart disease. San Diego: Academic Press. Incidence of congenital heart disease: II. Pediatrics 1975. 39: 1890–900. J Am Coll Cardiol 2002. Cell Mol Biol Res 1995. Genetics of human left-right axis malformations. The ventricular septum is formed from multiple portions of the developing heart. Zaffran S. Some of the current data on the incidence of CHD are given in Table 8. 5. Hoffman JIE. 235: 701–10. McGrath J.

eds. 42. 20: 385–99. 48. Jackson BT. Abu-Harb M. 55. Molecular basis of congenital and acquired long QT syndromes. 23. Mutations in the cardiac transcription factor NKX2. 33. 37. 8: 143–55. Seidman JG. 26. 24. a knowledgebase of human genes and genetic disorders. The DiGeorge anomaly. Pediatr Cardiol 1999. Kurnit DM. Riccardi VM. Allen HD. 2002: 141–213. Ackerman MJ. 45. Bowles NE. Cardiovasc Res 2005. 99: 646–55. Towbin JA. Nat Med 2004. 60. 113: 1136–9. 49. Curr Mol Med 2001. Molecular genetics of hypertrophic cardiomyopathy. 25. Circulation 2006. Including the Fetus and Young Adult. J Electrocardiol 2004. Expert Rev Mol Diagn 2006. 91: 791–3. Baker EJ. Yang Z. Missense mutations in the rod domain of the lamin A/C gene as causes of dilated cardiomyopathy and conduction-system disease. Macartney FJ et al. 6: 365–74. 341: 1715–24. Ching CK. 72: 361–3. Patterns for diagnosis. 65: 366–73. Associated abnormalities in children with congenital heart disease. Nava A et al. Dilated cardiomyopathy: a tale of cytoskeletal proteins and beyond. Orphanet J Rare Dis 2006. Bruneau BG. MacRae C. Molecular genetics of left ventricular dysfunction. Clin Rev Allergy Immunol 2001. Children and Adolescents. 52. Aortic arch anomalies associated with chromosome 22q11 deletion (CATCH 22). Layton WM. Nava A et al. Circulation 1968. Regulatory mutations in transforming growth factor-beta3 gene cause arrhythmogenic right ventricular cardiomyopathy type 1. Blair E. Increased adhesiveness of trisomy 21 cells and atrioventricular canal malformations in Down syndrome: a stochastic model. Cardiovascular genomics.Epidemiology of congenital heart disease 109 22. Hey E et al. Br Heart J 1995. 10: 463–4. 36. Burn J. eds. 27. 340: 573–5. The inheritance of hypertrophic cardiomyopathy. 29. Genet Med 2006. 56. Eur J Hum Genet 2003. Coronel R. Rimoin DL. Scherer SE et al. 54. Hong R. Familial hypertrophic cardiomyopathy: man. Benson DW. mouse and cat. 50. diseases. chance. Burch M. Online Mendelian Inheritance in Man (OMIM). Paediatric Cardiology. Brugada syndrome. Circ Res 2006. Wilson DI. 53. Goodship JA. Principles and Practice of Medical Genetics. Baty C. Riemenschneider TA. 38: 604–17. Seidman CE. Fatkin D. Ackerman MJ. Cardiovascular malformations associated with extracardiac anomalies and malformation syndromes. Hamosh A. Greenwood RD. Congenital heart defects and 22q11 deletions: which genes count? Mol Med Today 1998. Kramer HH. Momma K. Pacing Clin Electrophysiol 2006. N Engl J Med 1999. 1990: 1207–37. Genetics. 57. Piano MR. 41: 979–95. 1989: 614–28. J Am Coll Cardiol 2004. London: Churchill Livingstone. Towbin JA. Watkins H. Heart Disease in Infants. 29: 1130–59. Kurnit DM. QJM 1998. 38. The geneticenvironmental interaction. Davis N et al. Opthof T. 59. 47. The pathogenesis of congenital cardiovascular anomalies. 1: 81–90.5 affect diverse cardiac developmental pathways. Aldridge JF. Malformation patterns in children with congenital heart disease. Edinburgh: Churchill Livingstone. Erickson JD. Rampazzo A. Sabatine MS. Baldini A. Antenatal diagnosis of congenital heart disease and Down’s syndrome: the potential effect on the practice of paediatric cardiology. van Asselt KM. 44. 211: 100–8. 20: 43–60. Majewski F. 13: 46–58. Am J Dis Child 1987. Bowles NE. Kavanaugh-McHugh A et al. N Engl J Med 1968. Arrhythmogenic right ventricular cardiomyopathy type 1 (ARVD1): confirmation of locus assignment and mutation screening of four candidate genes. and genetic causality. 32. Brugada syndrome. Genetics of dilated cardiomyopathy: more genes that kill. The long QT syndrome family of cardiac ion channelopathies: a HuGE review. Gutgesell HP. 46. In: Emery AE. Chamber-specific cardiac expression of Tbx5 and heart defects in HoltOram syndrome. 104: 1567–73. 11: 69–76. 35. 31. 37: 1–6. 41. Lehmann MH. Antzelevitch C. Familial hypertrophic cardiomyopathy. Deletions within chromosome 22q11 in familial congenital heart disease. The aetiology of congenital heart disease. 28. 2: 134–40. Occhi G. 1: 35. 30. Beffagna G. Matsuoka R. 113: e450–5. 2nd edn. 74: 192–8. 17: 919–26. Rikken A et al. Michels VV. 107: 281–9. Priori SG. 141: 789–95. Role of genetic analyses in cardiology: part II: heritability estimation for gene searching in multifactorial diseases. 5th edn. eds. Burn J et al. Clin Pediatr (Phila) 1984. 4: 350–7. Patent ductus arteriosus and ventricular septal defect: trends in reported frequency. 23: 145–51. J Cardiovasc Nurs 1999. . Sistermans EA. Congenital long QT syndromes: clinical features. Tan EC. Silberbach GM. 51. Nucleic Acids Res 2005. Am J Med Genet 1985. Cardiovasc Res 2006. Kok HS. Towbin JA. Why the Brugada syndrome is not yet a disease: Syndromes. 33(Database issue): D514–7. Am J Hum Genet 1987. Genotypic and phenotypic characterization of Noonan syndrome: new data and review of the literature. Multifactorial inheritance hypothesis for the etiology of congenital heart diseases. Beffagna G. and morphogenesis. Edmonds LD. Napolitano C. Takao A. 39. Circulation 2006. Zahka KJ. Am J Epidemiol 1978. Jongmans M. van der Schouw YT et al. Solaro RJ. Scott AF. 134: 165–70. Baltimore: Williams & Wilkins. J Clin Invest 1999. Lancet 1992. Anderson CE. Congenital heart defects. Berecki G. Curr Cardiol Rep 2000. 20: 313–16. Desmosomal dysfunction due to mutations in desmoplakin causes arrhythmogenic right ventricular dysplasia/cardiomyopathy. Lindsay EA. 43. 44: 2041–3. 34. Sasaki T et al. Towbin JA. Wyllie J. 20: 97–102. Matthysse S. Amberger JS et al. Trampisch HJ et al. Bowles NE. 58. Cardiac channelopathies: it’s in the genes. molecular genetics and genetic testing. 279: 80–9 concl. Logan M. In: Anderson RH. J Cardiovasc Electrophysiol 2006. Modell SM. Nora JJ. Am J Med Genet A 2005. Dev Biol 1999. Matsuoka R et al. 40. Pediatr Cardiol 1999. In: Emmanouilides GC.

Tidsskr Nor Laegeforen 1990. 80: F49–53. Am Heart J 2004. Allan LD. 73. Pastor P. Cook A. 62. 68. Kaplan S. Otterstad JE.006 consecutive cases of congenital heart disease in the fetus. Hoffman JIE. Levine RA et al. Is mitral valve prolapse a congenital or acquired disease? Am J Cardiol 1997. 74. 72. Cook AC et al. 2nd edn. Teratology 1988. 337: 959–61. Incidence of congenital heart disease in Blackpool 1957–1971. Nascimento R. Circulation 1990. and future of pediatric cardiology. Teixeira F et al. J Am Coll Cardiol 1995. Circulation 1998. 65. Sharland GK. Levy D. Milburn A et al. Richmond S. 69.110 Fetal Cardiology 61. 81: 137–42. Wren C. 1970–1983. 341: 1–7. 71. Liberthson RR. Paediatric Cardiology. Prevalence and clinical outcome of mitral-valve prolapse. Macartney FJ et al. present. Reflections on the past. Prevalence of congenital heart disease. Chamberlin M et al. Fixler DE. 2002: 111–39. Freed LA. N Engl J Med 1999. Hook EB. Trends in congenital heart disease in Dallas county births 1971–1984. High prevalence of muscular ventricular septal defect in neonates. In: Anderson RH. Du ZD. Lancet 1991. Br Heart J 1977. Pulmonary atresia with intact ventricular septum: impact of fetal echocardiography on incidence at birth and postnatal outcome. Meberg A. Roguin N. Hypoplastic left heart syndrome: effects of fetal echocardiography on birth prevalence. 4: 208–23. Barak M et al. [in Norwegian] Spooner EW. 63. 70. mortality. UK and Eire Collaborative Study of Pulmonary Atresia with Intact Ventricular Septum. Allan LD. 66. Sharland GK. Increase in the incidence resulting from improved diagnostics methods]. Arch Dis Child Fetal Neonat Ed 1999. Donaldson L. Bound JP. Farina MA et al. Hoffman JIE. Freitas A. Sullivan I et al. Cardiol Young 1994. Daubeney PE. Presentation of congenital heart disease in infancy: implications for routine examination. 23: 1452–8. 110: 354–7. J Am Coll Cardiol 1994. 98: 562–6. 39: 445–50. Logan WF. 64. Hoffman JIE. Incidence. eds. Evaluation of a temporal increase in ventricular septal defects: estimated prevalence and severity in northeastern New York. [Children with congenital heart defects in Vestfold 1982–88. . 79: 226–7. 147: 425–39. London: Churchill Livingstone. and natural history. Baker EJ. Froland G et al. 37: 21–8. 26: 1545–8. Prospective diagnosis of 1. 67.

Fetuses with lesions that require intervention early in the neonatal period might especially benefit from closer care and the planning of delivery. In the Baltimore–Washington Infant Study between 1981 and 1989. the timing of the examination.9–11 Furthermore. and patients considered at high risk have been offered detailed fetal echocardiography. However. which is established for antiarrhythmic treatment but is still in the experimental stage for others.2 by the end of the first month. the true incidence among fetuses is difficult to evaluate. As some cardiac lesions can evolve in terms of their echocardiographic characteristics with the progress of pregnancy. and this has profound medical. A referral center should offer up-to-date information about current methods and results of treatment with the backup of a pediatric cardiologist and surgeon. The reported incidence rate per 1000 liveborn infants increases from 3. Owing to widely different levels of the obstetric scanning experience of examiners. Congenital heart defects constitute a major segment of birth malformations. 60% of cases with CHD were diagnosed by 4 weeks of age. might further improve fetal outcome12 (Chapter 33). and gestational age-related differences in national laws regarding termination of pregnancy. Screening in the low-risk population has been reported to have lower accuracy than in the high-risk population. The perinatal management will depend on associated extracardiac malformations or chromosomal disorders. After accurate diagnosis. socioeconomic.0 at the end of the first week of neonatal life. Specific indications and risks have been classified. There are several controversial issues in this field. 80% by 12 weeks. there is a large discrepancy in reported study results.9 Indications for fetal echocardiography: screening in low. obtaining a high level of sensitivity and specificity. particularly if it is performed in early pregnancy.4–6 To perform a complete and accurate examination. and 90% by 24 weeks.8 by the end of the first year. In recent years. most children are born to mothers who have no known risk features during pregnancy. and the usefulness of fetal echocardiography as a screening instrument in unselected populations has remained a matter of debate. and knowledge of the natural history of CHD.3 The likelihood of detecting a fetal cardiac defect is closely related to the experience of the ultrasonographer. The in utero development of some types of congenital heart defect is still partially unknown. and medicolegal consequences. .1 Cardiac defects are seen four to five times more frequently in stillbirth. and is currently under intensive study. as well as the wishes of the patient. possibilities for treatment. and the equipment used.2 However. To improve expertise in the diagnosis and management. the possibility of intrauterine cardiac intervention. the prognosis and outcome of the cardiac defect.7. 5. and 7. commonly at a center with appropriate expertise and facilities.8 The impact of these screening programs on the mortality and morbidity of children with prenatally diagnosed cardiac anomalies and of the general population has been evaluated with different results. highresolution ultrasound with pulsed and color Doppler imaging capabilities is required.3 at birth to 4. a certain group of defects cannot be reliably excluded by a single scan. tertiary centers have been created. psychological. and management options.and high-risk populations Ulrich Gembruch and Annegret Geipel Introduction Technical advances in ultrasound technology over the past decade and the introduction of fetal echocardiography into the prenatal ultrasound examination have further improved the antenatal detection of congenital heart disease (CHD). Nevertheless. cardiac anomalies are the most frequently overlooked lesions during prenatal ultrasound evaluation. efforts have predominantly focused on developing screening programs for prenatal diagnosis of congenital cardiac anomalies. counseling of the patient is mandatory about the kind of anomaly.

however.17 With advances in molecular genetics.27 While nearly all cardiac anomalies can be diagnosed by two-dimensional echocardiography.41 In a series of 6640 pregnancies with a first-degree family history of CHD.42. have a slightly higher recurrence rate than others.32 Although the majority of detailed cardiac examinations is performed between 20 and 24 weeks’ gestation. the most sensitive period is approximately 6 weeks.7%. animal experimental.14.21. The recurrence rate is about 2–4% for parental CHD and 2% for siblings. and should therefore always be followed by a second trimester examination33–37 (Chapter 13). including first trimester evaluation. High-resolution ultrasound with color Doppler facilities should be used in addition to two-dimensional echocardiography. diagnostic and prognostic information.20. at present the etiology for the majority of congenital cardiac anomalies is still under research. specialized centers provide increasingly the possibility of early fetal echocardiography performed from around 12–14 weeks of gestation.112 Fetal Cardiology Risk groups and indications for fetal echocardiography Recommendations on indications for fetal echocardiography are closely linked to the recognition of possible etiologic factors and risk groups in respect of CHD. Using this technique. and multifactorial inheritance (90%). and Examination of the high-risk population There are sufficient numbers of referral centers for highrisk patients where detailed echocardiography can be performed. a recurrence was seen in 2. Three main etiologic groups have been identified on the basis of empirical. and spatiotemporal image correlation (STIC).14. and therefore each risk category should be assessed independently. the concept of multifactorial etiology is being increasingly questioned.43 Furthermore. 3%). primary exogenous factors (maternal infections.40. 5%. This includes a genetic predisposition and the influence of environmental triggers at a vulnerable period of embryogenesis. Exact concordance was seen in 37% and group concordance in 44% of cases. valvular stenoses. B-flow imaging. insufficiencies.18–20 However.39.13–18 These are primary genetic factors (chromosomal aberrations. Concordance was highest (55%) in families with two or more recurrences and in cases with isolated atrioventricular septal defects (80%) and laterality defects (64%).1).3. a parent or a second-degree relative with a heart defect. the detection rate of congenital cardiac anomalies in referral centers is about 80–90%. as numbers of congenital cardiac defects have been identified as a result of inherited or new mutations of single genes and of inherited or spontaneous microdeletions. monogenic inherited defects. or an occurrence of a certain familial genetic syndrome (Table 9.17. and intracardiac shunts. the latter can provide essential . examination early in pregnancy has lower detection rates compared to second trimester echocardiography. Indications are the same as described for second trimester fetal echocardiography.44 whereas other studies found no statistical differences compared to siblings. especially through enhanced demonstration of the great arteries. Some lesions.17 imprinted genes may influence cardiac development. Extended echocardiographic examination includes visualization of International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) guidelines26: • the standard four-chamber view • the left and right ventricular outflow tracts with crossing of the aorta and the pulmonary artery • the ‘three-vessel–trachea view’ • the short-axis view of the ventricles and great arteries • the aortic arch and ductus arteriosus • the superior and inferior venae cavae. detailed fetal echocardiography. Indications for fetal echocardiography Although only approximately 10% of fetuses with cardiac anomalies have known risk factors. The proportion of fetuses affected is likely to be risk factor-specific. In some cases.42 More congenital heart defects occurred in the offspring of affected mothers (3–5%) than in those of affected fathers (2%).1). However. The most common indication is a family history of CHD – with a previous child affected. teratogens/maternal metabolic diseases.40 In general.43 Paternal CHD has been considered a stronger risk factor in one study. especially left heart obstructive lesions and heterotaxy syndromes. should be offered to this high-risk population (Table 9. 1%.28–31 A novel approach in the assessment of the fetal cardiac anatomy are four-dimensional (4D) sonography. and therefore in clinical practice the identification of high-risk populations is based on patient history and sonographic anomalies or markers (Table 9. some cases might be exclusively diagnosed by color-coded Doppler echocardiography. the rate of miscarriage in offspring of affected women (20%) was significantly higher than in those of affected fathers (9%). the recurrence risk of CHD as an isolated multifactorial disorder is obtained by taking the square root of the incidence in the specific population. In cases of multifactorial inheritance the cardiac anomaly is considered to result from the interaction between unspecified genetic and environmental factors.43 Besides the possibility of mitochondrial inheritance. and genetic data. 1%).2). the pulmonary veins.

Diagnosed chromosome anomaly (see Tables 9.39 A number of drugs (Table 9. Positive family history of CHD • one previous child affected • two previous siblings affected • maternal CHD • paternal CHD • syndromes or malformations associated with cardiac anomalies (see Table 9.1 Indications for fetal echocardiography (high-risk group for congenital heart disease (CHD)). often resulting in death.43 Furthermore.14. uE3. and also with coarctation of the aorta (2.5%. amphetamine. warfarin • intrauterine infections: rubella. However. Abnormal obstetric sonogram • suspected cardiac anomaly. Risk (%) 2* 10* 4* 2* 4–6** 12–16** 2–3 50–60 13–14 3–6 1–2 1–2 30–40 15–20 15–20 6 10 5 25–30 7 normal heart formation might depend on the maternal copy of the gene. hCG. retinoic acid (vitamin A). hydantoin.3 and 9. 0%. and parvovirus • high doses of ionizing radiation Documented fetal anomaly 1. After a normal examination they are calmed. pregnancy-associated plasma protein A.41.6) 3.22 Exposure to these teratogens in the first 6–8 weeks of pregnancy increases the risk of . and face the remaining pregnancy with great relief. and 4%. Twin gestation • monozygous twins • conjoined twins Decline of invasive prenatal diagnosis (‘genetic sonography’) In cases with increased risk due to • advanced maternal age • abnormal biochemical parameters in the maternal serum (AFP. the predominant effect of prenatal diagnosis. lithium. PAPP-A. human chorionic gonadotropin. whereas the risk was significantly lower in mothers with tetralogy of Fallot and transposition of the great arteries. 15. phenobarbital.21. uE3) • familial risk AFP. myocarditis from cytomegalovirus.43 suggesting monogenic inheritance for some defects. respectively). is early reassurance of patients with a previous child born after undiagnosed severe cardiac anomaly. α-fetoprotein.4) • arrhythmia extrasystoles tachyarrhythmia complete heart block • non-immune hydrops fetalis • hygromata colli • increased thickness of nuchal translucency ≥ 3. phenytoin. hCG. and 20–25 Indication Increased risk based on history 1. Exposure to teratogens in pregnancy • teratogenic drugs: alcohol. including heart anomalies. valproic acid). *The risk of recurrence is also dependent on the type of CHD in the index case.1) and high doses of ionizing radiation have been implicated as causes of various malformations. Modified from references 13. abnormal cardiac position • extracardiac malformations frequently associated with cardiac anomaly (see Tables 9.15. unconjugated estriol.5 mm • early (before 32 weeks) and/or predominant symmetrical growth restriction • amniotic fluid abnormality of moderate and severe degrees 2.Indications for fetal echocardiography 113 Table 9. anticonvulsants (carbamazepine. coxsackievirus.2) 2.13. a high recurrence risk of around 14% was found in mothers with an atrioventricular septal defect and aortic stenosis. Maternal diseases • diabetes mellitus • phenylketonuria • connective tissue disease and/or autoantibodies (risk for atrioventricular block) 3. trimethadione.4 and 9. **The risk of CHD is dependent on the metabolic control in the first weeks of pregnancy. and especially of fetal echocardiography. PAPP-A.

aortic stenosis. TAC.56 Fetuses with diagnosed extracardiac anomalies constitute another major group of patients in whom fetal echocardiography is warranted. coxsackievirus. periconceptionally and during the first 8 weeks of pregnancy might decrease the risk of the affected mothers to near the level of the normal population. which may occur in mothers with connective tissue diseases such as systemic lupus erythematosus and Sjögren syndrome. mitral valve prolapse.47. Associated CHD VSD. lithium. the risk does not seem to be increased for gastroschisis. with a higher . PS.5. A much higher incidence of cardiac anomalies is found in fetuses with sonographically suspected cardiac and extracardiac anomalies (Tables 9. alcohol. MvP with MR ASD. ventricular septal defect. AS. tetralogy of Fallot. A variety of malformations of the central nervous system. ToF. yielding by far the most cases of severe CHD. and cytomegalovirus can cause fetal myocarditis. retardation of growth and development. Modified from references 18. ASDII VSD. ASD.45 Coexisting maternal diseases. but good metabolic control of maternal blood sugar and phenylalanine levels. patent ductus arteriosus. DORV. ToF Rhabdomyoma VSD. atrioventricular septal defect. ASD. adenovirus. Sometimes. genital and urinary abnormalities. interrupted aortic arch. MvP. antiepileptics. and retinoic acid are the most important drugs. and gastrointestinal and urogenital systems are frequently associated with heart defects (Table 9. ToF Supravalvular AS. anal atresia. the mediastinum. Chapter 43). VSD. renal abnormalities. double outlet right ventricle. IAA Dilated aortic root. 22. truncus arteriosus communis.46 In cases of poor metabolic control during early pregnancy. TGA. transposition of great arteries. VSD. CoA. The overall incidence of extracardiac malformations with congenital heart disease varies from 27 to 42%. complete heart block. AVSD.114 Fetal Cardiology Table 9. such as insulin-requiring diabetes mellitus and phenylketonuria. CoA. AS PDA. coarctation of the aorta. **Vertebral anomalies. mitral regurgitation. but also in healthy women.48 and in phenylketonuria up to 14%.1 and 9. TAC. the likelihood of a chromosomal anomaly is markedly increased. may be associated with an elevated risk of fetal cardiac lesions. MR. ToF. CHB. VSD. heart defects. are associated with an elevated risk of fetal cardiac lesions.27. tracheo-esophageal fistula. hypoplastic left heart.49 A predominance of conotruncal lesions has been reported in both maternal diseases. If these anomalies are associated with a cardiac malformation. ToF. the risk for fetal CHD in maternal diabetes is between 4 and 6%. aortic arch. ASD ASD. such as maternal rubella infection early in pregnancy. VSD. ASD. CoA. PS. CoA. TGA Dilated aortic root. HLH. pulmonary stenosis. IAA.55 In patients referred for an ‘abnormal’ four-chamber view. VSD. respectively. VSD ToF.3. a cardiac anomaly can be expected in about 50–60% of patients.2 and 38 Syndrome Syndromes and conditions frequently associated with CHD. ASD. CoA Conotruncal defects (VSD. In this context. coarctation. specifically anti-Ro (anti-SSA) and/or anti-La (anti-SSB). MvP with MR AS ToF. VSD VSD. right AA. ear abnormalities and hearing loss. atrial septal defect (type II). Some viral intrauterine infections. IAA. cardiac defect. repetitive echocardiographic examinations of the fetal heart should be performed in the second and third trimesters for early detection of autoantibodyinduced fetal heart block and cardiomyopathy. but an increased risk for fetal structural cardiac lesions is not well documented in other infections. occurrence of CHD.48–50 If there are maternal autoantibodies.24. atresia of the choanae. IAA. ToF. PDA. as the detection of a cardiac anomaly fundamentally affects the prognosis and might influence perinatal care strategies. VSD. ASD. parvovirus B19.3). TAC). AA. CoA. limb abnormalities. PS PS. DORV.23. ToF Risk (%) 10 50 30 50–65 85–95 50 50 20–30 50–85 25–40 60–90 25 55–65 25 30–33 50 100 Apert Bourneville–Pringle (tuberous sclerosis) Cornelia de Lange CHARGE* association DiGeorge/Shprintzen (microdeletion del 22q11) Ehlers–Danlos Ellis–van Creveld Goldenhar (hemifacial microsomia) Holt–Oram Klippel–Feil Marfan Meckel–Gruber Noonan Rubinstein–Taybi Thrombocytopenia–absent radius (TAR) VACTERL** association Williams–Beuren (microdeletion) *Coloboma. ASD. depending on the training level of the referring physicians. dextrocardia HLH.24. DORV.

59–61 Intrauterine congestive heart failure with hydrops and associated extracardiac malformations has been reported as a negative prognostic factor for fetal outcome. where hydrops and hygroma result from a lymphatic drainage disorder.3 Extracardiac malformations frequently associated with a cardiac anomaly. Modified from references 22.58 Cardiovascular anomalies are often observed in fetuses with non-immune hydrops fetalis (Chapter 34). resulting in atrioventricular–valvular insufficiencies. in many cases with non-immune hydrops and a heart defect. frequency if detected during pregnancy than at birth. Ebstein’s anomaly.57. in early pregnancy spontaneous fetal demise frequently occurs in hydropic fetuses often associated with a chromosomal abnormality. and a coarctation of the aorta only may be associated.51. but also for exclusion of a cardiac anomaly which may be present in 30–40% of cases with complete atrioventricular block and in 2–5% of cases with extrasystoles and tachyarrhythmias.Indications for fetal echocardiography 115 Table 9. Furthermore. some cardiac defects can result in intrauterine congestive heart failure with signs of hydrops. First. the incidence is between 15 and 20% of cases.52 There are some types of cardiac lesions where non-cardiac abnormalities are commonly associated. tricuspid valve dysplasia. and obstructions of the right and left ventricular outflow tracts.4 Frequency of extracardiac and chromosomal anomalies associated in single types of CHD diagnosed prenatally. whereas others only rarely occur with extracardiac malformation (Table 9. For an effectively screened low-risk population the incidence of extracardiac malformations is assumed to be lower.* Based on the data of references 22. the incidence of the association between hydrops. at around 20%. 48. If only true cardiac anomalies are included. not only for differentiation. 38. this relationship is coincidental rather than causal. and 51–54 Central nervous system (2–15%) hydrocephalus microcephalus agenesis of the corpus callosum encephalocele Dandy–Walker malformation Mediastinum (10–40%) diaphragmatic hernia esophageal atresia (VACTERL association) Gastrointestinal (12–22%) duodenal atresia abnormal situs visceralis anorectal anomalies Abdominal wall (14–30%) omphalocele ectopia cordis Genitourinary (5–40%) hydronephrosis renal agenesis renal dysplasia horseshoe kidney Vascular (5–10%) single umbilical artery persistent right umbilical vein agenesis of ductus venosus Table 9. chromosomal anomaly. which results in increased right atrial volume and pressure with consequent elevation of hydrostatic venous pressure with resultant edema. as in atrioventricular septal defect.22. the high incidence of extracardiac malformations of 30–40% is biased by a selected referral to the tertiary center of high-risk fetuses with already detected anomalies. the anomaly rate is between 20 and 30%.62 Second. This is for two reasons. If brady.52. Because cardiac defects are more common in this group of fetuses. and is higher the earlier in pregnancy the screening is performed.55 However.4). such as atrioventricular septal defect or tetralogy of Fallot. and 57 Type of congenital heart disease Atrioventricular septal defect Ventricular septal defect Atrial septal defect Tetralogy of Fallot Double outlet right ventricle Hypoplastic left heart Truncus arteriosus communis Transposition of great arteries Coarctation of the aorta Tricuspid atresia Ebstein’s anomaly Aortic stenosis Pulmonary stenosis/ atresia Chromosomal anomaly (%) 35–47 37–48 1–3 27 12–45 4–10 14–33 0–3 21–30 2–9 0–3 1–15 4–5 Extracardiac anomaly (%) 30–50 30–37 16 25–30 19–20 1 15–21 15–26 12–20 15–34 6 13 20–26 *In most prenatal series the number of chromosomal and non-chromosomal extracardiac anomalies seems to be distinctly overrepresented.59. . Therefore. Essential to the pathomechanism is the development of atrioventricular–valvular dysfunction in combination with a relative restriction of the transatrial flow through the fossa ovalis. 52.and tachyarrhythmias are included as anomalies.55. Fetal echocardiography should be performed in all cases of fetal arrhythmia. chromosome anomalies may result in hydrops by non-cardiac mechanisms. for example in Turner syndrome.

65–68 While initial reports suggested that narrowing of the aortic isthmus with overperfusion of the head and neck is an underlying pathophysiological mechanism. Cardiac abnormalities were found in 16/1000 of those with NT of 2.5) with a combined total of 6982 fetuses. these initial observations were not confirmed by studies published subsequently. to 8. 4.25 There are six series (Table 9.4.4 mm 6/1102 (0.7 for NT between the median and ≤ 95th centile. 200574 Atzei et al.1%) 5/122 (4.5–4. 64. twin pregnancies must be mentioned. other hypotheses include abnormalities of the extracellular matrix. there is no evidence that this is directly related to the structural cardiac abnormality itself. in most cases representing Turner syndrome.3%) 64/1079 (5. hygroma colli is often persistent or even progressive and accompanied by pleural effusion and/or general hydrops. Chromosomal abnormalities are reported in up to 50%. in addition to a cardiac cause. karyotyping has to be discussed.1%) 2/335 (0. tricuspid regurgitation. hydrops.36 Abnormalities of the cardiovascular system are found at increased frequency in the group of fetuses with thickened nuchal translucency detected by screening between 11 and 14 weeks of gestation.9%) 135/2151 (6.5 mm 9/287 (3. 199769 Ghi et al. The prevalence of major cardiac defects increases exponentially with fetal NT thickness.64. and most recently a delay in development of the lymphatic vessels of the neck.5–3. or less commonly.76 Another indication for fetal echocardiography is fetal growth restriction. even at 12–14 weeks. Finally.7 for respective NTs of 3.116 Fetal Cardiology and cardiac defect is higher in first. including a high proportion of euploid fetuses with cardiac defects or genetic syndromes.25.1/1000).5 mm and more. the frequency of an associated abnormal karyotype (20–30%) is far higher in prenatal studies than in liveborn infants.6.1%) 42/597 (7.0%) 11/90 (12. Whereas increased nuchal translucency often represents a transient anomaly.5–3.71 However. The prevalence increased from 4.4 mm.5–5. these pregnancies constitute a new high-risk group and should receive detailed fetal echocardiography at mid-gestation and.75.and early secondtrimester studies. As an isolated heart defect diagnosed in utero might be the only sign of a chromosomal abnormality. In contrast to lesions Table 9. preferentially.36.8%) 76/4831 (1.72 In a recent study by Atzei and colleagues including 6921 fetuses. to 35.55 Therefore. 200525 Total .55.75 Coarctation of the aorta and other obstructions of the left ventricular outflow tract are the most common cardiac malformations associated with Turner syndrome in utero as well as postpartum. As the latter group corresponds to a 2–3 times higher risk for cardiac disease than those with a positive family history.25.5 mm.51. and demonstration of a cardiac defect should be followed by karyotyping. complete cardiac examination should be performed in all fetuses with early growth restriction.64. possibly accompanied by abnormalities of amniotic fluid. of nonchromosomal syndromes.77 A high intrauterine mortality rate is reported in those fetuses with prenatally detected heart disease and associated chromosomal pathology. related to increased NT. mediastinal compression.51.4 mm.22.70 As a variety of cardiac lesions with different hemodynamic flow pattern can be found among fetuses with increased NT.6%) 43/2365 (1. especially trisomy 18 and triploidy. and early growth restriction for detailed cardiac examination. in these settings the incidence of chromosomal anomalies is high. the most common chromosomal disorders associated with CHD are summarized.27. and/or increased pulsatility of the flow velocity waveforms in the ductus venosus seems to be causal for the majority of cases. 200172 Lopes et al.5%) 2/185 (1.5%) 18/722 (2.3%) Hyett et al. major cardiac defects were identified in 132 (19. an increased risk (4–7%) for cardiac anomalies is predominantly seen in monozygous twins. reporting the prevalence of major cardiac defects in fetuses with increased NT.27. Because echocardiographically detectable cardiac anomalies are often present in these disorders.6%) NT > 3. and ≥ 5.63.2. a temporary impairment of cardiac function in early pregnancy.5%) 1/43 (2.78 Another cause for the high frequency of chromosomal aberrations in prenatal series with CHD is the referral bias that primarily selects fetuses with an extracardiac malformation.2 for NT between the 95th and 99th centiles. and 126.64 Nuchal translucency (NT) screening selects about 70–80% of fetuses with chromosomal abnormalities at a 5% false-positive rate.5 Study Prevalence of cardiac defects in fetuses with increased nuchal translucency (NT) n 1389 1319 275 177 378 3444 6982 NT 2.78 In Table 9.69 With regard to etiology. 200373 McAuliffe et al.25.52. 18.9 per 1000 in those with NT below the median. The identification of a lethal chromosomal aberration might have an enormous impact on the further obstetric and neonatal management.4 mm and 64/1000 in those with NT of 3.51 Early intrauterine growth restriction with manifestation before 32 weeks of gestation as well as the gestational age-independent symmetric type might raise suspicion of chromosomal abnormality. 200535 Bahado-Singh et al.2%) 8/55 (14.

with a false-positive rate of 14%. abnormal serum biochemistry. This includes a variety of major malformations as described for trisomy 18. TAC. atrial septal defect.80 An increasing proportion of patients requiring fetal echocardiography are those who want to avoid invasive prenatal diagnosis despite given indications such as advanced maternal age. as approximately 50% of newborns with trisomy 21. By evaluating the heart with regard to ventricular septal defects. Offering second trimester ‘genetic sonography’ to women at increased risk for fetal trisomy 21 is associated with a high utilization rate. ASD. in contrast to hyperreflectoric bowel. AVSD. DORV. Associated heart defects AVSD. DORV. with avoidance of unnecessary amniocentesis-induced fetal loss. truncus arteriosus communis. tetralogy of Fallot. Modified from references 15. transposition of the great arteries. with a falsepositive rate of 6. AS. HLH. PDA Conotruncal defects (VSD. CoA Conotruncal defects (VSD. and coarctation of the aorta (Table 9. Paladini et al reported a 56% incidence of CHD in fetuses with known Down syndrome. totally anomalous pulmonary venous connection. In the second trimester. aortic stenosis.6 and 79 Karyotype Trisomy 21 Trisomy 18 Chromosomal abnormalities associated with CHD. atrioventricular defects. and on center-specific reference images and differences between ethnic groups. VSD. and Ebstein’s anomaly with rarely associated aneuploidies. TAPVC. pericardial effusion.6). 38. AS. IAA. BiPv. interrupted aortic arch. PS. CoA. HLH VSD. VSD. bicuspid aortic valve. 90% of those with trisomy 13. ASD. for instance. Conversely. TAC). 53. ventricular septal defect. ASD. fetal echocardiography is of particular importance. 99% of those with trisomy 18. bicuspid pulmonary valve. By combination of non-cardiac markers with the four-chamber view. One of the main reasons for these variations is the underdetection of congenital heart defects.82–84 However. each chromosomal disorder manifests its own phenotypic expression.4).4%. ToF.86 . AVSD. ventricular septal defects. as is nuchal translucency. PDA VSD. DORV). and 35% of those with Turner syndrome have a cardiac defect (Table 9. ASD VSD. IAA Risk (%) 40–50 99 80–90 40 60 30–40 40–60 30–60 75–85 Trisomy 13 Partial trisomy 22 (cat-eye syndrome) Triploidy Turner syndrome Partial monosomy 4p− (Wolf–Hirschhorn syndrome) 5p− (cri-du-chat syndrome) Microdeletion 22q11 AS. and are independent of observer and scanning conditions. In a targeted echocardiographic study. such as tricuspid atresia. coarctation of the aorta. pulmonary stenosis. VSD.55. atrioventricular septal defect. tetralogy of Fallot. as are nuchal thickness and pyelectasis. ToF. most studies evaluating second trimester screening for trisomy 21 detect fewer than 20% of heart malformations.57. double outlet right ventricle. are not dependent on gestational age-related cut-off levels. persistent ductus arteriosus. In addition to the detailed sonographic examination. the sensitivity was increased from 60 to 75%. trisomy 13. reported sensitivities of second trimester identification of fetal Down syndrome differ widely. 53% of cases with AVSD and normal visceral situs were associated with Down syndrome. and more subtle anomalies associated with trisomy 21. This approach clearly decreased the rate of genetic amniocentesis based on age-related risk and/or abnormal biochemical markers. and triploidy. hypoplastic left heart. and chamber discordance. PS. ToF TAPVC. In the presence of a normal ultrasound examination. PDA. the risk for fetal Down syndrome was reduced by 83% in patients with advanced maternal age and 88% in patients with abnormal triple screen. as are long bone and nasal bone lengths. ToF. ASD. which was an atrioventricular septal defect (AVSD) in 44% of cases. In appropriate clinical settings it was found to be cost-effective.81 In contrast to the majority of sonographic markers for trisomy 21. The targeted echocardiographic examination and the addition of color-coded Doppler imaging for better visualization has been shown to markedly increase the likelihood of detecting trisomy 21 during ‘genetic sonography’. with an increase in the detection rate of trisomy 21. BiAv. HLH.Indications for fetal echocardiography 117 Table 9. ToF. heart defects are usually not transient. 18. ASD. there is an accumulation of cases with chromosomal anomalies in fetuses with atrioventricular septal defect. ASD VSD CoA (mostly tubular). or other risk factors for fetal chromosomal anomalies. CoA. BiPv. BiAv.85 In contrast. tricuspid regurgitation. HLH. the sensitivity of ‘genetic sonography’ was further increased to 91%.79.

are considered severe. 50% of infant deaths are attributed to congenital anomalies. congenital heart defects represent the most common congenital malformations. intrauterine fetal death.3. with a low incidence of false-positive and false-negative results.90 The general use of ultrasound and the screening for malformations is widely accepted because this offers reassurance in most pregnancies and has been considered safe for both mother and fetus. Fetal echocardiography performed by skilled operators is known to be a reliable method for the diagnosis of congenital heart disease.91. the associated major malformations. In particular. There are therapeutic options for most cardiac anomalies. In cases of termination of pregnancy. the prognosis is unfavorable if there are chromosomal and non-chromosomal extracardiac anomalies and if the CHD is associated with fetal congestive heart failure.104 For the most serious CHD. it has to be interpreted correctly. or at cardiac surgery. Screening using the four-chamber view yields a more severe spectrum of CHD than is found in unselected live births. Thus.87–89 2.95. With a birth prevalence of 8/1000.5 times higher incidence than that of chromosomal anomalies and are four times as common as neural tube defects. For other cardiac lesions with predominant palliative treatment.5.55. non-directive counseling of parents can only barely be a reality.99. opinions differ. However. In series of prenatally diagnosed CHD. i. With appropriate training the four-chamber view is the easiest of the echocardiographic views to obtain.7. However. Methods of confirmation and further follow-up observation of the disease should be available. its natural history. 4/1000. the value of any screening is assessed in terms of the following criteria: 1.96 Therefore. Screening programs have been established in many countries. or neonatal death. owing to the increasing detection rate in fetal cardiac scanning for isolated and minor cardiac defects and also owing to significant improvements in interventional cardiology and cardiac surgery during recent decades.98 5.1). autopsy results should be obtained.3. the percentage of parents opting for termination of pregnancy has been continuously decreasing in most centers. in fetuses with increased nuchal translucency ≥ 3. In all fetuses with predicted heart disease. The disease should have a fixed spectrum of symptoms. The common criteria for screening are much more easily fulfilled by fetal echocardiographic screening compared to screening for chromosomal anomalies and neural tube defects. the gestational age at diagnosis. accounting for approximately 20% of perinatal deaths. and choosing aggressive cardiac surgery or deciding against surgical intervention. opinions on treatment options amongst pediatric cardiologists and pediatric cardiosurgeons differ widely.90.95–97 While false-positive diagnoses are rarities in most studies.5. In referring the patient to appropriate centers this criterion can be met. as in the latter the intention is mostly the diagnosis of anomalies with subsequent pregnancy termination. Non-directive counseling with pregnancy termination is one option. Screening should be accurate. with the Norwood operation and heart transplantation being alternatives in hypoplastic left heart syndrome. Counseling involving a multidisciplinary team should provide information on the cardiac anomaly. by cardiac catheterization. as do the accepted treatment opportunities in various countries. Exclusion of cardiac lesions is recommended in all malformations frequently associated with heart defects. The underlying disease should have a sufficient prevalence and severity. A fetal heart abnormality is likely if an abnormal four-chamber view or abnormal cardiac position is present.5 mm. generally with a poorer prognosis than CHD diagnosed after birth. while the other half.94 some have had extremely high falsenegative rates. According to the World Health Organization. continued educational support in fetal echocardiography is a critical issue in improving both sensitivity and specificity. The management options will depend on the type of cardiovascular anomaly. especially in fetuses with atrioventricular septal defect. most of these studies with high detection rates have been carried out in university hospitals with level-three experience.5 6.93 However. cardiac anomalies have at birth a 6.118 Fetal Cardiology Screening examination of low-risk patients There has been increasing demand for the assessment of the fetal heart during the routine screening program for fetal abnormalities. nonimmune fetal hydrops. and fetal growth restriction (Table 9. As long-term prognosis of some operation . warranting detailed fetal echocardiography. Approximately 50% of cardiac malformations are considered easily amenable to surgery. and the further management possibilities.27.7). about 30–40% of parents elected to interrupt the affected pregnancy (Table 9. and complete heart block.e. The disease should be treatable. These account for approximately 20% of neonatal deaths. as personal opinion and past experiences of the counselor enter counseling decisions.80 However.94 Rather disappointing results have been reported in screening a low-risk population outside these referral centers. left atrial isomerism.99 Most parents’ experience conflicts with decisions such as terminating or continuing the pregnancy. The screening methods should be simple and acceptable. with conflicting results. Furthermore.92 4. the diagnoses should be confirmed postpartum by echocardiography. as about 90% of fetuses with congenital heart disease present without any known risk factors. and the ethical considerations of the patient. in the presence of arrhythmia. 3. it is the most lifethreatening condition in the first month of life.

termination of pregnancy. left atrial isomerism with atrioventricular septal defect.7.111. 200252 Brick and Allan. This is of particular importance for anomalies dependent on a patent ductus arteriosus.105. are reported. 2003102 Fesslová et al. neonates with postnatal diagnosis but dying before reaching the hospital or after early discharge. since the results were influenced by many factors. 19973 Fesslová et al. the length of hospitalization.112. associated anomalies.113 However.105 In this context. including those after surgery. 199957 Paladini et al. heart transplantation) is not fully established and there is the expectation of further innovations in the therapy of complex cardiac anomalies. as knowledge of the condition should help to avoid the hazards of neonatal transfer and clinical worsening with the possibility of hypoxia and acidosis causing multiorgan failure and long-term neurological damage.101. These include tricuspid valve dysplasia. which present not only the most severe lesions. all previous studies on the prognosis of prenatally diagnosed cardiac anomalies are not helpful.113 Also.51. Apart from fetuses with hydrops.58.104. 7. techniques (Fontan operation. 1991100 Respondek et al. or other critical anomalies requiring immediate therapeutic and palliative interventions such as prompt infusion of prostaglandin E1.7 Outcome of CHD (studies with at least 100 fetuses included only) Cases of continuing pregnancy (n = 2291) Study Sharland et al.99.9.106 If termination of pregnancy is excluded.112 Retrospective studies on the outcome of all neonates with CHD could not demonstrate marked improvements in the early physiologic condition. which may otherwise result from mismanagement due to postpartum delays. survival advantage as a result of prenatal diagnosis has been difficult to demonstrate conclusively. survival rates of 40–60%. Prenatal diagnosis of a cardiac anomaly is thought to optimize perinatal management of the affected child. Screening should result in an improved outcome and a positive cost–benefit ratio.80 The outcome of several studies of congenital heart disease is summarized in Table 9.3. complete heart block. Ebstein’s anomaly. but also situations rarely observed postnatally. early balloon atrial septostomy. TOP. 2002101 Levi et al.55. Cost–benefit and costeffectiveness analyses from a background of hightechnology medicine seem artificial. including patient selection.107–110 However.11. 199451 Yagel et al. since in these circumstances the possible improvement in mortality and long-term prognosis in a few children justifies the expense of prenatal diagnosis.106. 2003103 Total Number of fetuses 442 100 168 847 400 408 271 670 3306 Abnormal karyotype 42 14 16 149 104 39 50 98 512 (16%) Extracardiac anomaly 71 42 10 162 118 63 135 159 760 (23%) TOP 220 26 25 245 150 98 77 174 1015 (31%) Intrauterine fetal death 57 5 15 72 16 19 43 50 277 (12%) Death postpartum 87 24 25 259 85 73 47 167 767 (34%) Alive* 78 45 103 271 149 185 104 279 1214 (53%) Lost/ ongoing 33 *Variable from 6 days after birth to median follow-up of 3 years.Indications for fetal echocardiography 119 Table 9.9. most postnatal series do not take into account neonates with CHD who die without diagnosis.103 Associated anomalies and chromosome disorders are poor prognostic factors associated with a higher intrauterine and postpartum mortality rate. counseling is even more difficult.57. and cases considered unsuitable for cardiac surgery because of low birth weight or serious extracardiac anomalies. Norwood operation. and cardiac tumors. the possibility of pregnancy termination after prenatal diagnosis may cause the prenatal selection of a group of . and growth restriction. or balloon valvuloplasty. heart failure. screening based on the four-chamber view mostly detects cardiac anomalies with an unfavorable prognosis. or the in-hospital mortality rate by prenatal diagnosis and following delivery and management in a tertiary care center.

detailed cardiac and extracardiac assessment is an important prerequisite for effective counseling of parents and for optimizing perinatal management. but not for neonates with univentricular hearts.115 whereas other authors could not show any reduction of mortality by prenatal diagnosis of hypoplastic left heart syndrome.120 Fetal Cardiology parents predisposed towards more aggressive neonatal management even in the most serious CHD. prematurity.120 It is important to realize that the largest decrease in survival after Norwood palliation occurs in the first month. In 1995.11 Most studies.121. neonatal heart transplantation may be an alternative option. and multiorgan failure. and further improvements in the surgical treatment and extracorporeal bypass technique may avoid significant long-term neurological impairment in infants with hypoplastic left heart syndrome after the Norwood procedure.110. because many affected pregnancies are terminated if the serious CHD is diagnosed early in gestation127 (Table 9. multiorgan failure. or pulmonary hypertension may already be predicted by fetal echocardiography.118 Recently. profound metabolic acidosis.119 in which the first-stage survival rate in full-term neonates with ideal anatomy and without other anomalies is above 80%. A considerably lower incidence of hypoplastic left heart in liveborn children.120 In some of these cases. with avoidance of severe hemodynamic compromise by hypoxemia. However. a profound effect on the prevalence and spectrum of CHD at term would be seen. clearly demonstrated a decreased preoperative morbidity. the cost for each cardiopathy diagnosed correctly (23 cases) was US$14 048.9.10. even if performed as simple four-chamber view screening. This applies especially to pre.112. such as hypoplastic left heart (HLH). ductus arteriosus constriction. in the overall incidence of live births with cardiac anomalies. optimized perinatal care.118 It can be speculated that a good preoperative condition resulting from prenatal diagnosis.112 A lower surgical mortality of first-stage Norwood surgery after prenatal diagnosis has been reported in one study. including two-dimensional echocardiography. there is only a small decrease. but success has been limited so far. intrauterine transthoracic balloon valvuloplasty has been advocated in critical aortic stenosis.7).117 Accordingly.126 If the detection rate of fetal cardiac screening. The cost averted by each severe cardiopathy resulting in pregnancy termination (13 cases) was US$24 854.124 as well as of pulmonary atresia with intact ventricular septum.10.109 especially the appearance of critical hemodynamic compromise with severe hypoxemia. and neurologic events.128 Having spared a family of a child with complex congenital heart disease.10. for first-stage surgical reconstruction is US$126 600 (range US$5600–551 000).114.10. by 2%. but not uniformly to mortality.125 has already been demonstrated as a result of sonographic screening.107 Consequently.109 In the small high-risk group of neonates with dextrotransposition of the great arteries. and is US$126 695 for cardiac transplantation. and delivery and postnatal care at a specialized center could clearly be demonstrated.116 For fetuses with hypoplastic left heart syndrome. is an emotive subject.109.115.109 A lower postoperative mortality has also been observed in one of these series. leading to a small reduction of infant deaths due to congenital cardiac anomalies and cardiosurgical procedures for congenital cardiac anomalies if the detection rate of screening for cardiac anomalies remains low.116 In neonates with dextrotransposition of the great arteries.109 Inadequate interatrial mixing might require urgent balloon atrial septostomy shortly after birth in around 10% of newborns with transposition of the great arteries.123. studies could demonstrate that the preoperative mortality109 and preoperative morbidity. at the time of prenatal diagnosis. with many social. and is caused by various factors such as non-chromosomal and chromosomal extracardiac anomalies. Therefore. the complicated preoperative course as a result of a restrictive foramen ovale. In order to prevent impaired left ventricular growth and to avoid a postnatal univentricular situation.108.114 In specific lesions. In the United States the average hospital cost for newborns with hypoplastic left heart is US$57 400 (range US$0–759 000).122 It is obvious that postnatal treatment in complex CHD is more expensive than the average cost of termination of pregnancy. however. and might contribute to a decrease in perinatal mortality and morbidity.and postoperative morbidity. and emotional consequences. based on US$40 for a 30-minute ultrasound examination of the fetus. could be significantly decreased if transposition of the great arteries was diagnosed before birth.9.9. and additional intracardiac findings such as a restrictive foramen ovale resulting in severe preoperative obstruction to pulmonary venous return or severe tricuspid dysplasia. Rustico and co-workers calculated the total cost of screening for cardiac anomalies in 7024 women (repeated twice in 15% of cases) as being US$323 104.108. a small study reported improved hospital survival in cases with HLH which underwent intrauterine left atrial decompression for restrictive atrial septum. and coarctation of the aorta.12. prenatal diagnosis of coarctation of the aorta has been shown to reduce preoperative hemodynamic instability and improve survival. profound metabolic acidosis. psychological. but has to be taken .9. This option in complex uncorrectable cardiac anomalies and those associated with severe chromosomal and non-chromosomal extracardiac anomalies drastically cuts the cost of neonatal management by avoidance of unnecessary cardiologic and cardiosurgical intervention. improvement of survival after prenatal diagnosis has been shown for neonates with cardiac defects amenable to biventricular repair. were higher. better outcomes resulting from prenatal diagnosis of CHD. transposition of the great arteries.

and sonographic screening for cardiac anomalies in particular. Nevertheless.130 With appropriate educational support and practical training the examiner is sufficiently skilled to obtain optimal images. For this reason.5. the four chambers cannot be seen adequately in a large proportion of fetuses. A potential advantage of prenatal diagnosis of heart disease is to allow planning of the delivery in a unit with all neonatal facilities.98. At this time the four-chamber view can be obtained in 95–98% during the first examination. screening by using the four-chamber view reduces the examination to a small part of the fetal heart. and as some lesions present secondary structural changes resulting from in utero development.133. with sensitivities ranging between 10%96. to avoid emergency transfer and separation from parents in this situation. the atrial and ventricular size as well as wall thickness can be assessed. scar tissue due to previous abdominal surgery. Studies on the efficacy of screening using the fourchamber view for the detection of cardiac malformations have differed widely.26 Some conotruncal malformations. A last profound benefit of prenatal diagnosis in general. more rarely.Indications for fetal echocardiography 121 into consideration. Before 18 weeks of gestation.95. Lesions usually associated with abnormal four-chamber view images are listed in Table 9.8.135 Buskens and co-workers136 described differences and deficiencies of various screening studies for the detection of fetal cardiac anomalies which are similar to those of many other screening studies for fetal anomalies. Some of the major differences are addressed here.6. The influence of gestational age in depicting a correct four-chamber view has been demonstrated by several investigators. often associated with chromosomal anomalies. Table 9. both atrioventricular valves. many conotruncal anomalies are potentially not detectable if screening is limited to the four-chamber view only. However. tricuspid dysplasia At the ventricular–systemic junction • aortic atresia • severe aortic stenosis • pulmonary atresia with intact interventricular septum • severe pulmonary stenosis At the venous–atrial junction • complete anomalous pulmonary venous drainage without pulmonary venous obstruction Others • severe coarctation of the aorta • interrupted aortic arch • double inlet ventricle • large ventricular septal defect • large atrial septal defect Cardiac anomalies with usually normal four-chamber view • transposition of the great arteries • tetralogy of Fallot • pulmonary atresia with ventricular septal defect • double outlet right ventricle • truncus arteriosus communis • mild Ebstein’s anomaly • small and/or outlet ventricular septal defect • atrial septal defect of secundum type • mild/moderate aortic stenosis • mild/moderate pulmonary stenosis • mild/moderate coarctation of the aorta • partial anomalous pulmonary venous drainage Four-chamber view screening Many screening models for cardiac anomalies based on demonstration of the four-chamber view have been incorporated into level 1 screening in many countries. fetal position and oligohydramnios have been reported as the main causes for unsatisfactory imaging in the four-chamber view. which is generally not considered in such cost-effectiveness analyses. generally within 1–2 minutes. is the exclusion of an anomaly for the vast majority of pregnant women.98. which open and close.93. providing reassurance and reducing anxiety.131 Higher-frequency probes will improve the likelihood of recognition of small defects: a ≥ 5-MHz transducer generally seems to be superior to a 3.8 Demonstration of cardiac anomalies by primary and secondary changes in the four-chamber view in the second trimester.97 and 80%. to calculate cost savings in favor of pregnancy termination seems questionable and ethically unacceptable. This includes demonstration of both atria. the interatrial septum with the valve of the foramen ovale beating into the left atrium. . The associated relief facilitates bonding between the mother and the unborn fetus.98. and. and the interventricular septum with the inflow tracts. Modified from references 98 and 104 Cardiac anomalies with abnormal four-chamber view At the atrioventricular junction • mitral atresia • tricuspid atresia • atrioventricular septal defect • severe Ebstein’s anomaly. reducing false-negative and in particular false-positive results.134 Displacing intrathoracic anomalies and abdominal wall defects may also result in an abnormal cardiac position detectable in the fourchamber view.129 It has been suggested that this would result in a detection rate of 2 per 1000 heart defects.132 However. may result in an increased left-sided deviation.5-MHz probe. The normal left-sided deviation of the heart is about 45 ± 20° (2 standard deviations (SD)). Marked hampering of ultrasound transmisson by adipose tissue. the cardiac axis and position should be analyzed. In addition. screening between 18 and 23 weeks (preferentially ≥ 20 weeks) of gestation using the four-chamber view improves the identification considerably. Furthermore.

19957 Tegnander et al. four-chamber view. 199597 Incidence/ 1000 3 5 8 5 8 12 Population risk Mixed Mixed Low Low Low Low Level of experience 1 1 1 1 1 1 Weeks of pregnancy — 18–22 16–24 18–22 18 16–22 Cardiac examination 4 CV Detection (%) Ott.8% for all CHD to 74. 199696 Todros et al. 1997137 Westin. 1995107 Hafner et al. 199627 Yagel et al. 200654 Tegnander et al.9 Data on different screening studies with detection rates of CHD Study Sharland and Allan. . others included a mixed patient population of low-risk as well as high-risk patients.9). incorporating a more homogeneous level of expertise. including the persistent patent ductus arteriosus and the type II atrium septum defect in the fossa ovalis region. Tertiary centers seem to have had a higher detection rate. This holds true also for lesions considered as physiologic in utero. 199294 Stümpflen et al. Although the time between 18 and 24 weeks of gestation is generally accepted as optimal for the assessment of the fetal heart. longer examination time. better equipment.3%. smaller defects. The detection rate must be significantly lower if cases of ventricular septal defects and type II atrial septal defects are included. The studied population was different. whose detection rate rose from 43. 2003140 Achiron et al.5. 19973 12 14 9 13 4 9 7 4 7 8 Low High risk Low Low Low Low Low Low Low Mixed Mixed 1–2 1–2 2 2 2–3 2 2 3 3 3 3 — 16–22 20–22 (16–)22* 18–23 > 18 — 18–24 18–28 13–16/20–22 69 (detectable in 4 CV) 4 CV 81 4 CV 5 4 CV 15 4 CV 15 (major CHD only) Anomaly screening 18 (major CHD only) 4 CV 26 (major CHD only) 4 CV + left outflow tract 14 Extended echo 62 4 CV + outflow tract 15 (57 of major CHD) 4 CV + outflow tract 35 (61 of major CHD) Extended echo ± color 44 4 CV + outflow tract 75 (major CHD only) 4 CV + outflow tract 65 4 CV + outflow tract 21 4 CV + outflow tract 61 4 CV 48 Extended echo 78 4 CV 47 Extended echo + color 88 Extended echo 81 *Some fetuses were examined at 16–18 weeks. as they might be associated with chromosomal and non-chromosomal anomalies. 1998138 Carvalho et al. the majority of examinations were performed from 18 weeks onwards and some even as early as 16 weeks Table 9. if ventricular and atrial septal defects (type II) were excluded. Some studies were designed as retrospective and some as prospective studies. and a mostly preselected study population with a higher incidence of cardiac anomalies. Mild cardiac defects are of equal importance. 200680 Rustico et al. 2.140 4. such as muscular ventricular septal defects or mild valve dysfunction. 3. Depending on the definition of a cardiac anomaly. as demonstrated by Hafner and co-workers. This was partially based on selection factors – some studies reported on screening results in the general population. 1992135 Buskens et al. 2006139 Wong et al. 5. were included in some and excluded in other studies. With an expected incidence of eight cardiac anomalies in 1000 live births it seems extremely unlikely that these differences reflect true incidences of cardiac anomalies. The examination periods varied in studies. but the vast majority at 22 weeks. 4 CV. the informative and important differentiation into severe and mild defects used in some studies should not be applied uncritically to the efficacy assessment of prenatal screening.138 However. 20025 Oggè et al. 199298 Vergani et al.122 Fetal Cardiology 1. The reported incidence of congenital cardiac defects in the study populations was between 3 and 13 per 1000 births (Table 9.

16. a maximum of 40–50% of congenital anomalies can be diagnosed by the four-chamber view.107. respectively. Lesions in the region of the outflow tracts and the large vessels are usually not detectable via the four-chamber view.54.95. Postnatal echocardiography was restricted to selected cases. Others will manifest in the course of pregnancy in lesions of sufficient severity (pulmonary stenosis.142 Without firm integration of the four-chamber view into sonographic screening. naturally.124 If the last prenatally ‘physiologic’ anomalies are excluded. compared to using the four-chamber view alone. 10%. tetralogy of Fallot.8). multiple studies have shown a further increase of 20–30% in the detection rate of cardiac anomalies by incorporating visualization of the outflow tracts and the great arteries into the examination. respectively. These defects can be diagnosed .124 There were different observation periods and follow-up information.2. aortic stenosis. is not nearly approached in most of the general screening settings. Major cardiac anomalies are usually detected in the first week of life. including some severe anomalies which become symptomatic early after birth (transposition of the great arteries.141. however. which has rarely been performed in published studies.90. complete two-dimensional echocardiography. 8. since an improvement of long-term prognosis and mortality rates can be expected from prenatal diagnosis and the optimization of perinatal management. especially marked in fetuses with chromosomal anomalies.94 This means that 2 or 3 in 1000 live births or 1:300 to 1:500. Furthermore.27. Comparison of studies is difficult.97 This is consistent with prospective sonographic screening studies for anomalies in general. double outlet right ventricle and truncus arteriosus communis). 10%.g.Indications for fetal echocardiography 123 of gestation. but the majority of defects – which are usually not critical – are commonly detected after months. coarctation of the aorta). it becomes apparent that some are always detectable in the four-chamber view and some only with certain localization and size (e. atrial septal defects of the secundum type. others included visualization of the left ventricular outflow. pulmonary stenosis and pulmonary atresia with ventricular septal defect. found in the majority of fetuses with trisomies 13 and 18. and 10% of all cardiac anomalies. utilizing the four-chamber view in the second phase between 16 and 22 weeks of gestation. If we look at the different types of cardiac anomaly (Table 9.97 Higher sensitivity of screening with the four-chamber view and detection rates between 48 and 69% in unselected and mixed-risk populations has been reported in other centers.131.94. as reported detection rates are between 5 and 20%.126 but a significant increase in detection rate is seen in areas with a high training level of all participating examiners. and postnatal deaths was incomplete in many studies. double outlet right ventricle. The findings of second-trimester ‘first level’ ultrasound studies in unselected populations by demonstration of the four-chamber view are sobering. 7.54. ventricular septal defects of the outflow tract.129. Optimal follow-up for complete detection of all cardiac anomalies.9). whereas in other studies it was generally performed. have a cardiac anomaly that can be detected by demonstration of the four-chamber view.97 Tegnander and colleagues detected only 10% of all cardiac anomalies and 26% of critical ones. The frequency of postnatal detection of atrial and ventricular septal defects and pulmonary stenoses is profoundly enhanced by the use of color Doppler echocardiography. The postmortem examination of spontaneous abortions.95.27. The latter defects account for 25% of all congenital cardiac anomalies in live births. pulmonary stenoses. and.143 The results of different studies are summarized in Table 9.9. In some studies cardiac views were limited to the fourchamber view. ranging from a 2-day instruction course for demonstration of the four-chamber view to many years of echocardiographic experience. tetralogy of Fallot. ventricular septal defects). and truncus arteriosus communis are common cardiac anomalies. The spontaneous loss rate between the second trimester and term. will result in slightly higher rates and different relative distributions of cardiac anomalies in the second trimester when compared to those in live births. or even color Doppler echocardiography (Table 9. presumably owing to the higher level of training of the examiners. To date. 6. in which a correspondingly small percentage of anomalies was detected. clinical examination was not performed in all newborns.7. Realistic estimates state that currently only 20% of those anomalies detectable in the second trimester fourchamber view are being diagnosed.127 Additional visualization of the ventricular outflow tracts Prenatal detection of the transposition of the great arteries would be of particular importance. would include a pediatric cardiology follow-up examination with colorcoded Doppler echocardiography at the end of the first week of life and a repeated examination at 4–6 months.96. A critical factor that especially applied to cardiac anomalies was the availability of a thorough postnatal evaluation with respect to examination methods as well as the follow-up period. Almost never diagnosable in the four-chamber view are the majority of ventricular septal defects. as the lower sensitivity as well as a higher incidence of cardiac anomalies at earlier gestational ages contributes to different results.96.124 In certain studies. the patent ductus arteriosus which account for 30%. This potential. stillbirths. There were marked differences in the level of experience of the examiners. the detection rate of congenital cardiac defects is even lower.

Rotstein Z et al. The Baltimore–Washington Infant Study 1981–1989. early fetal echocardiography should be offered at 13–14 weeks of gestation followed by an additional echocardiographic examination at 20–22 weeks of gestation at specialized centers.8%) 105 (1.0%) 666 (7.2%) 1442 (16.124 Fetal Cardiology Table 9.and long-term outcome for critical heart defects. the examiner should be trained to visualize the four-chamber view as well as the ventricular outflow tracts with the origin and course of the great arteries.2%) 9035 CHD 23 (2. detailed fetal echocardiography should be performed in all high-risk pregnancies such as those with a positive family history for cardiac defects.6%) 1 (0. 2. Magee CA.4%) 1511 (16. Weissman A. On the other hand. in particular for the dextrotransposition of the great arteries and hypoplastic left heart. Therefore. Loffredo CA. In this context the increased thickness of nuchal translucency at 11–14 weeks of gestation seems to be an important marker for CHD even in chromosomally normal fetuses (Table 9. Mavrides E. Am J Cardiol 1978. especially transposition of the great arteries. 1993. Loffredo CA. Cardiac anatomy screening: what is the best time for screening in pregnancy? Curr Opin Obstet Gynecol 2003. Christianson R. Thilaganathan B. However. In: Ferenz C. Campbell S. Yagel S.5 Conclusions Many reasons stated above make the prenatal diagnosis of as many cardiac anomalies as possible desirable from the viewpoint of the perinatologist and the perinatal cardiologist. Rubin JD. 4: 33–62. 96: 550–5. 4. Perspectives in Pediatric Cardiology. this can be achieved by permanent training and liaison between the screening examiner and the specialized referral center. prenatal diagnosis of congenital heart defects with adequate perinatal management has been shown to improve the short. As nearly 90% of all pregnancies with fetuses identified as having a cardiac anomaly are not initially considered to be at increased risk. Current ultrasonographic technology is sufficient for this purpose.10 Yield of CHD by indication for fetal echocardiography. Epidemiology of Congenital Heart Disease. By achieving detection of serious CHD earlier in fetal life. Carvalho JS. and 145 Indication Family history of CHD Maternal diabetes Teratogen exposure Unsatisfactory/abnormal cardiac views Arrhythmia Extracardiac anomalies including hydrops Intracardiac echogenic focus Aneuploidy Increased nuchal translucency with normal karyotype Total Referrals 2374 (26. NY: Futura. with maternal diabetes. Neill CA.7%) 22 (2. Additional visualization of the outflow tracts and the great arteries increases the rate to 65–70% and includes all critical cardiac defects. Demonstration of the four-chamber view allows the detection of 40–50% of congenital cardiac anomalies in a low-risk group of patients. 42: 641–7. eds. Improving the effectiveness of routine . and as customary prenatal sonograms almost always miss an isolated cardiac defect.7 14 1. termination of an abnormal fetus is more acceptable.8%) 1 (0.4%) 959 (10. Congenital heart disease in a cohort of 19502 births with long-term follow-up.5 472 16 152 4 122 67 88 only by imaging of the outflow tracts of both ventricles.6%) 1391 (15. Hoffman JIE.1%) 42 (5. Rubin JD. The full potential of fetal echocardiographic screening can be attained only with further improvement in the level of expertise of all persons involved in screening. In particular. 3. an improvement of postnatal outcome could be demonstrated for some subgroups of cardiac defects.7%) 230 (28. 15: 143–6. 144.9%) 20 (2. and with already detected cardiovascular and extracardiac fetal anomalies (Table 9. general cardiac screening performed by many different examiners of varying experience results in detection rates that fall far below the potential. Perry LW.7%) 344 (3. References 1. Ferencz C. For all high-risk patients. Based on the combined data of references 5. Therefore.10). Shinebourne EA.1%) 453 (56. Armonk.9%) 799 Prevalence/1000 for indication 9. as this has been shown to improve prenatal detection rates considerably.2%) 7 (0. Circulation 1997. adequate detection of cardiac anomalies can be achieved only by cardiac screening of all fetuses in the second trimester coupled with referral of women with affected fetuses to centers with specialist cardiac experience. 5.7%) 243 (2. Allan LD. 56. Infants with congenital heart disease: the cases. Congenital heart defects: natural course and in utero development.10).

88: 387–91. 84: 581–5. Loffredo CA. 9: 535–47. and models. 91: 494–504. Am J Cardiol 1999. Bald R. Payne R. Redel DA. J Ultrasound Med 2002. Circulation 1995. J Perinat Med 1991. Color flow mapping of fetal heart. Mahle WT. 78: 1–8. Koren G. 30: 191–9. 32. 24: 1060–7. Randall P. Bacha EA. 17: 1283–98. J Thorac Cardiovasc Surg 2007. Campobasso G. eds. Stümpflen A. Gembruch U. Bald R. 172: 1741–9. Dyson DC. Edwards MB. 27. Marino B. 107: 1277–82. Hansmann M. 1993. Brealey S. 20. Hobbins JC. Relationship between nuchal translucency thickness and prevalence of major cardiac defects in fetuses with normal karyotype. Early fetal echocardiography – a reliable prenatal diagnosis tool. Bernaschek G. Allan L. Shalev E. Geipel A et al. 30. Ferencz C. Boughman JA. Nora JJ. Gembruch U. Perspectives in Pediatric Cardiology. The Baltimore–Washington Infant Study 1981–1989. NY: Futura. Loffredo CA. 348: 854–7. Pediatrics 2001. Ultrasound Obstet Gynecol 2006. Hypoplastic left heart syndrome with intact or highly restrictive atrial septum: surgical experience from a single center. 18. Ultrasound Obstet Gynecol 2005. Atzei A. Chita SK. 112: 24–30. Chaoui R. 35. 19. Parsons JM Accuracy of fetal echocardiography in the routine detection of congenital heart disease among unselected low risk populations: a systematic review. J Ultrasound Med 2006. Wimmer M. Familial recurrence of congenital heart disease: an overview and review of the literature. Prenatal diagnosis of congenital heart disease. Am J Obstet Gynecol 1987. 29. 31. 28: 229–36. Cooper M. Fetal transesophageal echocardiography during balloon valvuloplasty for severe aortic valve stenosis at 28+6 weeks of gestation. discussion 586. 4: 123–67. Lorber A. Nora JJ. Prenat Diagn 1997. Wegner RD. Pierpont ME. 12. Breuer J. Nicolaides KH. Kamenir SA. Lancet 1996. Obstet Gynecol 1995. Z Kardiol 1990. Roge CI. Kleinman CS. Circulation 2007. 24. Fetal echocardiography: retrospective review of clinical experience and an evaluation of indications. Morotti R. Am Heart J 1993. Tegnander E. 154: 1121–32. Newburger JW. Nora AH. 166: 111–16. 10. 86: 1649–53. International Society of Ultrasound in Obstetrics and Gynecology. Gembruch U. Novel application of 4D sonography with B-flow imaging and spatiotemporal image correlation (STIC) in the assessment of the anatomy of pulmonary arteries in fetuses with pulmonary atresia and ventricular septal defect. 26: 154–7. Int J Cardiol 1990. Prenat Diagn 2004. 156: 79–85. Schwanitz G et al. Gauvreau K. Fetal twodimensional Doppler echocardiography (colour flow mapping) and its place in prenatal diagnosis. Epidemiology of Congenital Heart Disease.Indications for fetal echocardiography 125 6. 8. Kumar RK. Rubin JD. Ultrasound Obstet Gynecol 2006. 28. Pilu G. 16. Hahn S. 193: 1253–9. Am J Med Genet 1988. 36. Genetic basis for congenital heart defects: current knowledge: a scientific statement from the American Heart Congenital Cardiac Defects Committee. Clancy RR. Magee CA. Congenital heart disease and extracardiac anomalies. Comparison of outcome when hypoplastic left heart syndrome and transposition of the great arteries are diagnosed prenatally versus when diagnosis of these two conditions is made only postnatally. Digilio MC. Sarkozy A. Berg C. Heart 2002. Gembruch U. Johnson M. 27: 613–18. prenatal screening for major congenital heart defects. Cardiac screening examination of the fetus: guidelines for performing the ‘basic’ and ‘extended basic’ cardiac scan. Causes of congenital heart diseases: old and new modes. Huggon IC. 134: 256–7. . 7. Eik-Nes SH. 33. 21: 23–9. 25: 187–96. 125: 1409–19. 13. Stanziano A et al. 15. McGaurn SP et al. Trines J. The accuracy of antenatal fetal echocardiography screening in high. Calcagni G. 28: 8–14. The examiner’s ultrasound experience has a significant impact on the detection rate of congenital heart defects at the second-trimester fetal examination. The genetics of congenital heart disease. 23. Detection rate of early fetal echocardiography and in utero development of congenital heart defects. Enderlein MA. 21. Möglichkeiten und Grenzen eines Screeningprogramms Pränatale Diagnostik fetaler Herzfehler durch Untersuchung von ‘High-risk’ und ‘Low-risk’-Kollektiven. Gynäkologe 1997. Armonk. Update on counseling the family with a first degree relative with a congenital heart defect. 27: 107– 13. associations and indications for fetal echocardiography. Redel DA. Ann Thorac Surg 2007. Detailed screening for fetal anomalies and cardiac defects at the 11–13-week scan. The use of colour Doppler in fetal echocardiography. 14. 25. Copel JA. Dallapiccola B. 79: 96–106. Grant J. 19: 27–32. Prenat Diagn 1989. Allan LD. Heep A et al. In: C Ferencz. 22. Strauss A. Ott WJ. 86: 577–82. The antenatal diagnosis of congenital heart disease using fetal echocardiography: is color flow mapping necessary? Obstet Gynecol 1991. mechanisms. 9. Basson CT. Gajewska K. Chatterjee M. Kohl T. Kleinman CS. Benson W et al. Impact of prenatal diagnosis on survival and early neurologic morbidity in neonates with the hypoplastic left heart syndrome. Towards a molecular understanding of congenital heart disease. BJOG 2005. Hornberger LK. Vida VL. Smrcek JM. Hansmann M. Neill CA. Zerres K.and low-risk patients. 34. 37. 115: 3015–38. Becker R. Fetale Echokardiographie und klinische Genetik—Eine enge Wechselbeziehung. Copel JA. Larrazabal A et al. 17. 28: 40–6. Am J Obstet Gynecol 1986. 26. Nield LE et al. Stümpflen I. Effect of detailed fetal echocardiography as part of routine prenatal ultrasonographic screening on detection of congenital heart disease. Gembruch U. Sharland GK. Antenatal sonography of fetal malformations associated with drugs and chemicals: a guide. Volpe P. Diagnosis of congenital cardiac defects between 11 and 14 weeks’ gestation in highrisk patients. 29: 137–42. Miskin M. Ultrasound Obstet Gynecol 2006. Am J Obstet Gynecol 2005. Council on Cardiovascular Disease in the Young. 11. Weiner Z. Eur J Pediatr 2007. Ultrasound Obstet Gynecol 2006. Khan KS. Am J Obstet Gynecol 1995. McAuliffe FM. Carvalho JS. Fetal heart scanning in the first trimester. Tarnoff H.

Ultrasound Obstet Gynecol 2002. Little J. Recurrence risk in offspring of adults with major heart defects: results from first cohort of British collaborative study. 81: 954–7. Fesslová V. 47. Fetal cardiac abnormalities identified prior to 14 weeks’ gestation. 18: 645–8. Allan LD. Azen C. 24: 921–31. Berg KA. 58: 334–7. Allan LD. Sharland GK et al. Milburn A et al. Prenat Diagn 2002. Lorenz RP et al. Siu S. 50. Pavlova M. Scanning indications and technique. Gill HK. Stewart PA. 81: 225–6. Ultrasound Obstet Gynecol 2001. Rubin JD. Snijders RJM. Smit S et al. Chita CK et al. Sebire N. Fetal. BJOG 2006. Von Kaisenberg CS. Souka AP. Genetic and environmental influences on malformations of the cardiac outflow tract. Spectrum and outcome of atrioventricular septal defect in fetal life. 351: 311–16. J Clin Ultrasound 2004. Pregnancy in women with congenital heart defects: what are the risks? Heart 1999. 352: 343–6. 54. 67: 775–8. 42. 8: 8–10.640 consecutive pregnancies evaluated by detailed fetal echocardiography. Factors affecting the prognosis of Ebstein’s anomaly during fetal life. Increased nuchal translucency with normal karyotype. 43. Rouse B. 67. Ultrasound Obstet Gynecol 1996. Kuehl KS. Patterns of recurrence of congenital heart disease. Buskens E. 22: 545–52. Ultrasound Obstet Gynecol 1997. 58.and third-trimester fetuses with congenital heart disease.126 Fetal Cardiology 38. 192: 89–95. Evolution and long term outcome in cases with fetal diagnosis of congenital heart disease: Italian multicenter study. Am J Obstet Gynecol 2005. 60. 4: 272–8. 62. Increases nuchal translucency at 10–14 weeks of gestation as a marker for major cardiac defects. Indications for fetal echocardiography from a tertiary-care obstetric sonography practice. 53. Respondek ML. Extracardiac anomalies. Teratology 1990. Incidence of congenital heart defects in fetuses of diabetic mothers: a retrospective study of 326 cases. Kozlowski P et al. Familial recurrence of congenital heart disease in a prospective series of mothers referred for fetal echocardiography. . Simpson LL. Souka A. 63. Ghi T. 12: 18–26. Obstet Gynecol 1993. German-Speaking Down Syndrome Screening Group. Koch R et al. Meise C. Papapanagiotou G et al. 113: 675–82. 41. Shields LF. Maternal Phenylketonuria Collaborative Study (MPKUCS) offspring: facial anomalies. Kamil D et al. Villa L. 32: 123–8. Hyett JA et al. 65. Prospective diagnosis of 1006 consecutive cases of congenital heart disease in the fetus. Tercanli S. Huggon IC et al. 22: 345–50. Austria and Switzerland. 26: 839–49. 23: 1452–8. 82: 594–9. Sharland GK. Hyett JA. Gembruch U. Boughman JA. Germer U. In: Drose JA. 39. Sharland GK. 192: 1005–21. Berg C. Nava S. An analysis of 6. Cook AC et al. Saltvedt S. Fesslová V. J Ultrasound Med 2005. 56. Murphy HF et al. 135: 1081–5. Hyett JA. and early neurological sequelae. Chitayat D. Maternal diabetes and cardiovascular malformations: predominance of double outlet right ventricle and truncus arteriosus. 57. Webb G. Abnormalities in the heart and great arteries in chromosomally normal fetuses with increased nuchal translucency thickness at 11–13 weeks of gestation. Nava S et al. Cardiol Young 2002. Paladini D. 66. Sharland GK. Comstock CH. 69. Familial risks of congenital heart defect assessed in a populationbased epidemiologic study. 20: 22–9. Am J Obstet Gynecol 2005. 69: 89–95. Berg C. Villa L. Sullivan IA. 51. Astemborski JA et al. Teodoro A et al. 90: 575–9. malformations. ed. 42: 923–9. 64. Ferencz C. Ultrasound Obstet Gynecol 2001. J Am Coll Cardiol 2003. 68. Fouran JC. Westin M. 52. 44. Philadelphia: WB Saunders. Burn J. Lancet 1998. Meise C. Expert Rev Cardiovasc Ther 2005. Noble P. Russo MG. Simpson JM. 7: 245–50. 49. Drose JA. McCarter RJ. Routine ultrasound examination at 12 or 18 gestational weeks for prenatal detection of major congenital heart malformations? A 55. Efficiency of fetal echocardiography and yield by risk category. Gasiorek-Wiens A. Smith RS. Nicolaides KH. Ultrasound Obstet Gynecol 1994. Nuchal translucency and fetal cardiac defects: a pooled analysis of major fetal echocardiography centers. Fetal Echocardiography. The prognostic values of hemoglobin A1c in predicting fetal heart disease in diabetic pregnancies. Bergman G et al. 1998: 15–57. Ultrasound Obstet Gynecol 2003. Germer U. Screening for trisomy α1 by fetal nuchal translucency and maternal age: a multicenter project in Germany. Huggon IC. 61. Heart 1999. Allan LD. J Am Coll Cardiol 1994. Drblik SP et al. Ultrasound Obstet Gynecol 1996. Maternal diabetes mellitus: which views are essential for fetal echocardiography? Obstet Gynecol 1997.299 fetuses. Venous Doppler ultrasound in 146 fetuses with congenital heart disease. Holloway S et al. Loffredo CA. Cullen S. 10: 242–6. Am Heart J 1998. Prenatal diagnosis of congenital heart disease in the Naples area during the years 1994–1999 – the experience of a joint fetal-pediatric cardiology unit. 87: 423–8. Arch Dis Child 1992. Obstet Gynecol 1996. 40. 48. 41: 319–26. Moscoso G. UK multicentre project on assessment of risk of trisomy 21 by maternal age and fetal nuchal-translucency thickness at 10–14 weeks of gestation. Geipel A. Splitt M. 17: 398–402. Lancet 1998. randomised controlled trial comprising 36. 59. Perdu M. 3: 1125–30. Potential impact of population screening for prenatal diagnosis of congenital heart disease. Meyer-Wittkopf M. Arterial Doppler ultrasound in 115 second. Binotto CN. Am J Med Genet 1987. Am J Med Genet 1997. Sharland GK. The syndrome of left isomerism: sonographic findings and outcome in prenatally diagnosed cases. Makrydimas G. Gan EA. Hess J et al. aneuploidy and growth retardation in 100 consecutive fetal congenital heart defects. Simpson JM. Sotiriadis A. Gembruch U. 45. Cardiology Study Group of the Italian Society of Pediatric Cardiology. Geipel A. Am J Cardiol 1986. Crawford DC. 46.

Viart P. Tegnander E. 82. Vintzileos AM. Stenley FJ. Prenatal diagnosis of Turner syndrome: report on 69 cases. Wapner R. 78. Copel JA. Riley MM. Lee K. Natural history and outcome of prenatally diagnosed cystic hygroma. Prenat Diagn 2004. 84. 77. Brick DH. 29: 619–24. J Am Coll Cardiol 1996. Ultrasound Obstet Gynecol 2003. Halliday JL. Alexander S. Screening for congenital heart disease prenatally. 27: 252–65. Prenatal detection of heart defects at the routine fetal examination at 18 weeks in a non-selected population. Results of a 2 1/2-year study in the South East Thames region. 22: 3–10. 96. 81: 679–82. 101. Landau LJ. The association between congenital heart disease and Down syndrome in prenatal life. Kustermann A. Sharland GK. Tegnander E. Michael CA. DeVore GR. Arch Dis Child 1991. Is genetic ultrasound cost-effective? Semin Perinatol 2003. Outcome of prenatally diagnosed congenital heart disease: an update. Lancet 1993. 5: 372–80. 27: 173–82. Agangi A et al. Manning N. 74. Short-term outcome of isolated and associated congenital heart defects in relation to antenatal ultrasound screening. Georgia. and in-hospital deaths among infants with selected birth defects – United States. New York: Informa Healthcare. A study to determine the incidence of structural congenital heart disease in monochorionic twins. 95. Carnforth. 157: 648–55. Nikolaides KH. Hegesh J. 80. 93. van Vugt JM. 75. Bahado-Singh RO. Yostalo P. DeVore GR. Centers for Disease Control and Prevention (CDC). Obstet Gynecol 2001. Extended fetal echocardiographic examination for detecting cardiac malformations in low risk pregnancies. 10: 391–6. Am J Obstet Gynecol 1987. Lancet 1990. Yeo L. 71. Allan LD. Hess J. Khoury MJ. 66: 284–7. UK: Parthenon Publishing. Elevated first-trimester nuchal translucency increases the risk of congenital heart defects. Buskens E. Paladini D. 15: 104–8. Pediatr Cardiol 2002. 76. Chew C. Fesslová V.Indications for fetal echocardiography 127 70. Fetal echocardiographic screening for congenital heart disease: the importance of the four-chamber view. Knöpfle G. 79. Allan LD. Populationbased study of antenatal detection of congenital heart disease by ultrasound examination. 27: 160–72. Atlanta. Allan LD. The use of genetic sonography to reduce the need for amniocentesis in women at high-risk for Down syndrome. Br J Obstet Gynaecol 1992. Gelernter I. Hum Reprod Update 2003. Saari-Kemppainen A. Semin Perinatol 2003. 21: 532–8. 187: 1226–9. Papp C. 92. Results of chromosomal analysis in fetuses with cardiac anomalies diagnosed by first. 56: 25–9.and early second-trimester echocardiography. Hobbins JC.. Am J Obstet Gynecol 2002. Palmieri S. Yagel S. Infant mortality from congenital malformations in the United States. Ultrasound Obstet Gynecol 2000. Villa L. Prenatal diagnosis of congenital heart disease and fetal karyotyping. 2003. 97. 103. Chaoui R. Ultrasound Obstet Gynecol 2003. Khoshnood B. Semin Perinatol 2003. 98. Wladimiroff JW. 1990–1994. Prenat Diagn 2006. Beke A. Factors influencing the outcome of congenital heart disease detected prenatally. 304: 671–4. Newnham JP. Zosmer N. 26: 1062–4. Pathophysiology of increased nuchal translucency: a review of the literature. 100. Carvalho JS. Lopes LM. 2006: 103–19. Br Med J 1992. 192: 1357–61. Grobbee DE. Ultrasound Obstet Gynecol 2003. 23: 449–53. Gembruch U. 86. Chita SK. 28: 1805–9. 85. 89.? The quest for the etiology of increased nuchal translucency. Brizot ML. Penny DJ. Keinonen OP. 9: 175–84. In: van Vugt JMG. Smulian JC et al. eds. Ultrasound screening and perinatal mortality: controlled trial of systematic one stage screening in pregnancy. Guven M. 72. Johansen OJ et al. Baschat AA. Archer N. The four-chamber view: four reasons why it seems to fail in screening for cardiac abnormalities and suggestions to improve detection rate. Lopes MA et al. Williams W. Cragan JD et al. Grandjean H. Eik-Nes SH. Montana E. Tartaglione A. hospital charges. Routine screening for congenital heart disease: a prospective study in the Netherlands. Pilu G. Achiron R. 102. Ultrasound Obstet Gynecol 1997. 83. Lockhart SM. The fetal heart or the lymphatic system or . 91. Guzman ER. 1970–1997. Incidence of major structural cardiac defects associated with increased nuchal translucency but normal karyotype. Ultrasound Obstet Gynecol 2007. 88. 25: 711–17. Glaser J. Long-term experience with the prenatal diagnosis of cardiac anomalies in . Shulman LP. Ultrasound Obstet Gynecol 2005. 98: 620–7. Opinion.. 99. 25: 215–20. Pilu G. Hospital stays. Prenat Diagn 2004. The Eurofetus Study Group. Sethna F et al. Huggon C. Obstet Gynecol 1993. 90. Structural and functional cardiac abnormalities identified prior to 16 weeks’ gestation in fetuses with increased nuchal translucency. Ultrasound Obstet Gynecol 2001. The role of fetal echocardiography in genetic sonography. eds. Genetic ultrasound scan – second trimester. Geipel A. Ghi T. 73. Haak MC. Ganapathy R. D’Andrea T. Prenatal detection of heart defects in a non-selected population of 30.149 fetuses – detection rates and outcome. 24: 965–8. 94. J Ultrasound Med 2006. Chen L et al. Huggon IC. Ultrasound and the Fetal Heart. Thom E et al. Allan LD. Prenatal Medicine. 1996: 71–80. Effects of frequent ultrasound during pregnancy: a randomised controlled trial. 27: 152–9. In: Wladimiroff JW. Zhang WH. 336: 387–91. Mezei G et al. 342: 887–91. 24: 1136–42. Kleinman CS. Levi S. MMWR Morb Mortal Wkly Rep 2007. Karjarlainen O. Evans SF. Counselling following a diagnosis of congenital heart disease. Paladini D. Green J. Am J Obstet Gynecol 2005. 18: 610–14. 87. Ultrasound Obstet Gynecol 1995. 99: 220–5. Ultrasound Obstet Gynecol 2006. Trends and outcomes of prenatal diagnosis of congenital cardiac malformations by fetal echocardiography in a well defined birth population. 81. Calabro R. 22: 470–8. Vintzileos AM. Sharland GK. Gembruch U. Hansmann M. Down syndrome risk estimation after normal genetic sonography. Johansen OJ et al.

6: 313–19. Jones OD. Ultrasound Obstet Gynecol 1994. Franklin O. 337: 959–61. Br Heart J 1995. 112. Tworetzky W. 118. high-risk pregnancies in a tertiary center. Sharland GK. 126. De Havay P. 4: 855–64. Bull C. 77: P1–3. Eapen RS. 133. Stoll C. Coltri A. Circulation 1999. 117. Abnormal fetal cardiac axis in the detection of intrathoracic anomalies and congenital heart disease. Ghidini A et al. Copel JA. Impact of fetal echocardiography on incidence at birth and postnatal outcome. 1996: 107–15. Ultrasound Obstet Gynecol 2000. Grobbee DE. 17: 380–5. 110: 2125–31. Ultrasound Obstet Gynecol 2001. Sholler GF. D’Alton ME. 108. Incorporating the four-chamber view of the fetal heart into the second-trimester routine fetal examination. Congenital heart disease: the impact of delivery in a tertiary center on SNAP scores (scores for neonatal acute physiology). British Paediatric Cardiac Association. 15: 271–8. Sharland GK. Allan LD. Daubeney PFF. McElhinney DB. Ital Heart J 2003. Cook AC et al. Yates RS. 120. Simpson JM. Pediatrics 1998. Lee W. 109. 130. Bear MB. End-result of routine ultrasound screening for congenital anomalies: The Belgian Multicentric Study 1984–92. 10: 90–3. Chang AC. Burch M. In: Wladimiroff JW. Am J Perinatol 1998. Kirk JS. 113. Maeno YV. Hey E. Linker DT. De Vore GR. Tworetzky W. Brenner JI. 12: 659–63. Heart 2002. 5: 366–71. Yoon GY et al. Harvey-Wilkes K. Kamenir SA. Prevalence and outcome of congenital heart disease in infancy: a 10-year population-based experience. World experience of percutaneous ultrasound-guided balloon valvuloplasty in human fetuses with severe aortic valve obstruction. Manning N et al. Abu-Harb M. Simpson LL. 182: 184–91. Ash K. 110. Ultrasound and the Fetal Heart. Bove FJ. Garne E. and results of successful intervention. 105. Ultrasound Obstet Gynecol 2001. 87: 67–9. Nereo NE et al. Jaeggi ET. eds. Platt LD. Effect of prenatal diagnosis of critical left heart obstruction on perinatal morbidity and mortality. Reddy VM et al. Richmond S. Eik-Nes SH. Sinclair B et al. Ultrasonographic left cardiac axis deviation: a marker for fetal anomalies. 4: 24–8. Apfel HD. Improved surgical outcome after fetal diagnosis of hypoplastic left heart syndrome. 24: 1143–9. 111. Prenatal features of ductus arteriosus constriction and restrictive foramen ovale in d-transposition of the great arteries. Tan ASA. Outcome after prenatal diagnosis of the hypoplastic left heart syndrome. 136. Defoort P. Am J Cardiol 2000. transport. Mariani S. Fink N. and outcome in fetuses with left ventricular outflow tract obstruction. Butera G et al. Jpn J Thorac Cardiovasc Surg 1999. 122. 98: 562–6. 103: 1269–73. Antenatal diagnosis of congenital heart disease and Down’s syndrome: the potential effect on practice of paediatric cardiology. Am J Obstet Gynecol 1992. Circulation 2004. Circulation 1998. Balloon dilation of severe aortic stenosis in the fetus: potential for prevention of hypoplastic left heart syndrome: candidate selection. Jennings RW et al. 128. Prenatal diagnosis of coarctation of the aorta improves survival and reduces morbidity. 79: 371–4. 114. Franklin WH. Kern JH. Kohl T. 74: 192–8. Circulation 2001. UK and Eire Collaborative Study of PAIVS. Benettoni A. Gavard L. 123. 10: 237–41.128 Fetal Cardiology 104. Hypoplastic left heart. J Thorac Cardiovasc Surg 1991. Rowland DG. 129. Fermont L et al. Hypoplastic left heart syndrome: effects of fetal echocardiography on birth prevalence. Pilu G. Heart 1998. Does a prenatal diagnosis of congenital heart disease alter short-term outcome? Ultrasound Obstet Gynecol 1997. Cool A. management and outcome of pre. Diagnosis. Massaro TA. 17: 386–91. Kleinman CS. 85: 187–91. Am J Obstet Gynecol 2000. J Ultrasound Med 1993. Management of hypoplastic left heart in a consortium of university hospitals. 102: 841–8. Wyllie J. Sensitivity and specificity of prenatal features of physiological shunts to predict neonatal clinical status in transposition of the great arteries. Wren C. Hess J. Changing impact of fetal diagnosis of congenital heart disease. Sharland G. Medcaris AL. 115. Lancet 1991.versus postnatally diagnosed major congenital heart disease: a population-based study. Prenat Diagn 2004. Early developmental outcome after the Norwood procedure for hypoplastic left heart syndrome. Lancet 1999. 116. Obstet Gynecol 1995. Detection of transposition of the great arteries in fetuses reduces neonatal morbidity and mortality. Am J Cardiol 1995. Current and potential impact of fetal diagnosis on prevalence and spectrum of serious congenital heart disease at term in the UK. 15: 237– 42. technique. 124. Smith RS. Evaluation of prenatal diagnosis of congenital heart diseases by ultrasound: experience from 20 European registries. 76: 809–11. 85: 1230–3. Levi S. Prenatal diagnosis of congenital heart disease. 60: 5–11. 167: 1000–3. Allan LD. Cooper SG. Clementi M. Wladimiroff JW. 99: 916–18. Comparative analysis of pattern. 110: 1743–6. Tegnander E. 47: 47–56. Desjardins C. Sullivan I. Buskens E. UK: Parthenon Publishing. Horenstein J. Ultrasound Obstet Gynecol 1995. Fetal echocardiography: factors that influence imaging of the fetal heart during the second trimester of pregnancy. Coulon R. D’Ottavio GD et al. Schaaps JP. Surgical treatment for hypoplastic left heart syndrome. Circulation 2004. 125. Crane JM. Bonnet D. Rustico MA. 132. Pulmonary atresia with intact ventricular septum. Fetal heart screening in low-risk pregnancies. Comstock CH. The Euroscan Group. 127. Huhta JC. Circulation 1999. 106. Hinton VJ. Ultrasound Obstet Gynecol 1997. Sharland G. 354: 1242–7. Wilkins-Haug L. 102: 1148–52. 119. Carnforth. Eur J Obstet Gynecol Reprod Biol 1995. 121. 135. Printz BF. Screening for congenital heart disease with the four-chamber view of fetal heart. Allan LD et al. The influence of prenatal diagnosis on postnatal outcome in patients with structural congenital heart disease. 134. 131. Arch Dis Child 1997. 107. prospects and problems. Gutgesell HP. Ultrasound Obstet Gynecol 1995. Jouannic JM. 99: 1209–14. . Vergani P.

Gruppo Piemontese for Prenatal Screening of Congenital Heart Disease. 142. Sanders SP et al. 143. . Ward C. 18: 808–15. Philipp K. Lancet 1996. Hafner E. Ultrasound Obstet Gynecol 2006. and outcome of anomalous fetuses. Accuracy of routine ultrasonography in screening heart diseases prenatally. 25: 197–202. Prenat Diagn 1998. 10: 901–6. 166: 1473–81. Ultrasound Obstet Gynecol 2003. Chiappa E et al. 28: 137–420. Kremer C. Alembik Y. 146. 171: 392–9. 138. Am J Obstet Gynecol 1994. Estroff JA. Am J Obstet Gynecol 1992. Wong SF. 144. the RADIUS Study group. Geipel A et al. Todros T. Crane JP. Winborn RC et al. Friedman A et al. 1985 to 2003. Lee-Tannock A. 140. Gaglioti P. Factors influencing the prenatal detection of structural congenital heart diseases. Ultrasound Obstet Gynecol 2006. Faggiano F. Trends in fetal echocardiography and implications for clinical practice. 348: 1732. Prenat Diagn 1997. Hamar BD. Routine fetal echocardiography and detection of congenital heart disease. A randomized trial of prenatal ultrasonographic screening: impact on the detection. Stoll C. Chan FY. Prenatal screening for congenital heart disease with four-chamber and outflow-tract views: a multicenter study. 21: 19–25. Bromley B. management. Berg C. LeFevre ML. Schuchter K. Scholler J. Silverman NH. 28: 779–84. Ductus venosus blood flow alterations in fetuses with obstructive lesions of the right heart. Maccanti S. Fetal echocardiography: accuracy and limitations in a population at high and low risk for heart defects. 141. Faggiano F. Sterniste W. J Ultrasound Med 2006. Dziura J. 145. Cincotta RB. 24: 781–6. Todros T.Indications for fetal echocardiography 129 137. Friedberg MK. 139. Oggè G. Prenat Diagn 2004. Detection of fetal congenital heart disease in low-risk population. Changing indications for fetal echocardiography in a University Center population.

.

Umbilical venous blood is distributed to the right lobe of the liver through the arcuate connection to the portal vein. In the mammalian fetus.4 We have studied the distribution of umbilical and portal venous and inferior vena cava blood in fetal lambs by injecting radionuclide-labeled microspheres into each vessel and determining the distribution of the spheres in the fetal body. Although a proportion of the substrates enter the hepatic circulation. The left hepatic vein drains into the inferior vena cava in close proximity to the ductus venosus. where it is joined by the portal vein. In the fetal lamb. which receives about 90% of its blood supply from this source. apart from passage of a small amount of coronary venous blood into the left ventricular cavity through Thebesian veins. the remaining 10% is derived from the descending aorta through the hepatic artery. I have proposed that it facilitates preferential streaming of blood flow. The ductus venosus acts as a bypass for umbilical venous blood. oxygen uptake and carbon dioxide removal from the body occur in the lungs. Most portal venous blood flows to the right lobe of the liver. To some extent it reduces the resistance to the flow of umbilical venous blood to the inferior vena cava by diverting it away from the hepatic microcirculation. The right hepatic vein drains separately into the right and posterior aspect of the inferior vena cava and the orifice is also partly covered by a distal valve-like membrane. In the fetus. oxygen uptake and carbon dioxide removal are accomplished in the placenta through the umbilical circulation. as mentioned below. Only a small amount passes through the ductus venosus and none is distributed to the left lobe of the liver. Blood thus flows serially through the circulation and. branches are distributed to the right lobe of the liver. the remaining small amount is contributed by the portal vein. Course of the circulation Postnatally the pulmonary circulation is completely separated from the systemic circulation.5. no mixing of arterial and venous blood occurs.6 Umbilical venous blood is distributed to the left lobe of the liver. the two vessels drain through adjacent orifices separated by a sharp ridge.2. an average of about 50% bypass the liver to enter the general fetal circulation via the ductus venosus the inferior vena cava. In the human. Energy substrates are absorbed in the gastrointestinal tract and delivered to the liver through the portal venous system before entering the general circulation. In the sheep fetus. After this junction. Well-oxygenated arterial blood ejected by the left ventricle flows through the systemic arteries to supply all tissues of the body with oxygen and nutrients. Blood then enters the systemic venous system and returns to the right atrium and ventricle. the ductus venosus originates and passes first posteriorly and then superiorly to connect with . The umbilical vein then arches to the right lobe. Based on angiographic studies in which contrast medium was injected into the umbilical vein. it is ejected into the pulmonary arterial circulation and returns to the left atrium and ventricle through the pulmonary veins. a thin valve-like membrane covers the distal orifice of the ductus venosus and left hepatic vein. Lind et al suggested that umbilical venous blood passing through the ductus venosus flowed largely to the left atrium through the foramen ovale. Energy substrates diffuse or are actively transported from the maternal circulation across the placental membrane and are transferred to the fetal body via the umbilical vein. This blood mixes with systemic venous blood before entering the cardiac ventricles to be ejected to perfuse the fetal body.1.1 Distal to the branches supplying the left lobe. Almost all the blood passing through the ductus venosus (92–95%) is from the umbilical vein. As seen in Figure 10. oxygenated blood returns to the body through the umbilical venous system. the left hepatic vein and ductus venosus drain through a common orifice on the left posterior aspect of the vena cava.10 Circulation in the normal fetus and cardiovascular adaptations to birth Abraham M Rudolph Fetal circulation Postnatally. the umbilical vein enters the porta hepatis and gives rise to several branches that are distributed to the left lobe of the liver.3 Although the function of this membrane has not been defined.

with an average of about 50%.8 Ductus venosus blood is preferentially distributed through the foramen ovale into the left atrium and left ventricle. The crista dividens. also flows preferentially across the foramen ovale.7.9 This will tend to direct blood on the left anterior aspect of the vena cava through the foramen ovale.1). The ductus venosus arises from the umbilical vein.11 The mechanisms responsible for this selective streaming have not been fully defined. Abdominal inferior vena cava blood and the right hepatic venous blood of lower oxygen saturation is preferentially distributed into the right ventricle and pulmonary artery. Similarly. from about 20 to 90%. similarly to the flow of abdominal inferior vena caval blood (Figure 10. This preferential streaming of ductus venosus and left hepatic venous blood through the foramen ovale provides blood of higher oxygen saturation to the left atrium and ventricle and thus into the ascending aorta. the crescentic edge of the superior portion of the atrial septum. It is first diverted in a retrograde direction into the upper portion of the inferior vena cava during atrial systole and then enters the foramen during the rapid inflow phase from the inferior vena cava.9 The ductus venosus stream is directed largely through the foramen ovale.9 . Similar streaming has been observed in the inferior vena cava and atria in fetal sheep by color flow Doppler studies.1 Course of blood flow in the region of the porta hepatis.9–11 The mean velocity of the abdominal inferior vena cava stream is relatively low (about 15 cm/s). In contrast. right hepatic venous blood may be deflected by the valve toward the tricuspid valve. to direct flow through the tricuspid valve.132 Fetal Cardiology Tricuspid valve SVC Foramen ovale LHV RHV Ductus venosus Umbilical vein IVC Portal vein Figure 10. These streaming patterns of abdominal inferior vena cava blood and ductus venosus blood can be identified by direct observation of the thinwalled inferior vena cava in the thorax of the fetal lamb in utero.10. either to the left. much higher (about 55–60 cm/s). Almost all superior vena cava blood passes through the tricuspid valve and is distributed into the right ventricle. which then arches to the right to join the portal vein. the valvelike structures over the entrance of the ductus venosus and left hepatic vein may direct the blood draining from these vessels toward the foramen ovale. Another mechanism that has been proposed is that different velocities of the streams tend to allow separation in the inferior vena cava. Ultrasound examinations of human fetuses have also shown similar flow patterns of blood in the ductus venosus and abdominal inferior vena cava. Ultrasound studies in fetal lambs demonstrated that the small amount of superior vena cava blood that enters the foramen ovale does so indirectly. Normally about 5% or less flows through the foramen ovale into the left atrium. however. both in the lamb and in the human fetus. which enters the inferior vena cava through the same orifice as the ductus venosus. Blood from the left hepatic vein. Umbilical venous blood is distributed to the left lobe of the liver. whereas abdominal inferior vena cava blood flows preferentially through the tricuspid valve to the right ventricle. the Eustachian valve and the lower portion of the atrial septum move in unison. These valves are not well defined in the human fetus. The proportion of umbilical venous blood that passes through the ductus varies greatly. so other mechanisms must account for preferential streaming. to facilitate blood through the foramen ovale. In the sheep fetus. overlies the inferior vena cava. The inferior margin of the atrial septum separates the entrance of the inferior vena cava from the left atrium. Portal venous blood is largely distributed to the right liver lobe and only a small proportion passes through the ductus venosus (reproduced with permission from reference 1). right hepatic venous blood streams preferentially across the tricuspid valve. a poorly oxygenated stream from the abdominal inferior vena cava and right hepatic vein is observed on the posterior right aspect of the vena cava. The ductus venosus mean velocity is. During phases of the cardiac cycle. It is suggested that this high velocity permits the stream to maintain a degree of separation from the abdominal inferior vena cava stream in the thoracic portion of the inferior vena cava and carry the blood through the foramen ovale.The distribution of blood from the left and right hepatic veins was also examined using radionuclidelabeled microspheres in fetal sheep. whereas distal inferior vena cava blood streams across the tricuspid valve. A stream of well-oxygenated blood from the ductus venosus and left hepatic vein can be identified on the left anterior aspect of the vena cava. so that the posterior left portion of the inferior vena cava connects directly through the foramen ovale to the left atrium. or to the right.

pulmonary trunk. there is essentially no mixing of oxygenated pulmonary venous and systemic venous blood. Admixture of oxygenated and systemic venous blood In the adult circulation. About 22% of umbilical venous blood is returned to the placenta without first passing through the systemic microcirculation. in the fetus. Preferential streaming of blood from the ductus venosus and left hepatic vein to some extent separates well-oxygenated and poorly oxygenated blood. and abdominal inferior vena cava all enter the thoracic portion of the inferior vena cava. head and cerebral circulation. as well as the placenta. Systemic venous blood is preferentially directed into the right ventricle. left and right hepatic veins. mean pressure (reproduced with permission from reference 1). feeding. gaseous distension. The figures alongside the chambers and vessels are pressures in mmHg related to amniotic pressure level as zero. Fetal vascular pressures The fetus is surrounded by amniotic fluid in the uterus within the abdomen. Umbilical venous pressure is about 8–10 mmHg near the umbilical ring and 2–3 mmHg lower near the placenta. In the quietly standing ewe. The inefficiency resulting from systemic venous and umbilical venous recirculation constitutes about 33% of the combined ventricular output of the fetal heart. blood delivered to all fetal tissues and to the placenta is a mixture of oxygenated umbilical venous and systemic venous blood. . Normally none of the blood flowing through the ductus arteriosus passes in a retrograde fashion across the aortic isthmus to the ascending aorta and its branches. Blood from the ductus venosus. but a large amount enters the umbilical–placental circulation However. and ductus arteriosus to the descending aorta and its branches to supply the arterial branches of the lower body.2 Course of the circulation in the heart and great vessels in the late-gestation fetal lamb. a small proportion passes into the pulmonary circulation but the majority is directed through the ductus arteriosus to the descending aorta (Figure 10. only a small proportion passes across the aortic isthmus into the descending aorta. A proportion of umbilical venous blood mixes with portal venous blood and enters the central circulation through the right hepatic vein.2). Fetal pressures are thus raised by increases in intra-abdominal pressure as occurs with straining. or uterine contraction. Thus. This arrangement is inefficient because it imposes an additional workload on the heart to supply oxygen to the tissues. has reduced oxygen saturation. which ejects into the ascending aorta. blood returning to the heart from the superior and inferior vena cava that is distributed to the fetal tissues without first being delivered to the placenta for oxygenation contributes to the inefficiency of the fetal circulation. Right ventricular blood is ejected into the pulmonary trunk.Circulation in the normal fetus and cardiovascular adaptations 133 65 70 m55 45 55 70 3 60 70 4 55 65 40 70 m55 45 m3 70 m2 Figure 10. Descending aortic blood is distributed to the abdominal organs and the tissues of the lower trunk and lower extremities. The left atrium receives blood from the foramen ovale and pulmonary veins. it is customary to relate all vascular pressures to amniotic cavity pressure. Most ascending aortic blood is distributed to the coronary circulation. The figures in circles within the chambers and vessels represent per cent oxygen saturation levels. In the sheep fetus under normal conditions. oxygenated umbilical venous and poorly oxygenated systemic venous blood mix at several sites in the fetal circulation before being distributed to the systemic arteries. with no phasic change during atrial or ventricular systole. and upper extremities. All pressures are presented with amniotic pressure as baseline (Figure 10. In the left atrium. intraamniotic pressure is usually about 8–10 mmHg above atmospheric pressure. Normally the pressure shows a continuous flat contour. blood entering the foramen ovale from the inferior vena cava is joined by pulmonary venous blood. about 45% of superior vena cava blood and 53% of inferior vena cava blood are returned to the fetal tissues without having had the opportunity to take up oxygen in the placenta. Similarly. which. m. and then empties into the left ventricle.2). Some umbilical venous blood is returned to the placenta without first being delivered to fetal tissues to permit oxygen uptake.

Systemic arterial blood has a PCO2 of 43–45 mmHg and a pH of about 7. The large oxygen difference between the maternal arterial and fetal umbilical venous PO2 is the result of diffusion limitation across the placental membrane. The P50 (the PO 2 at which hemoglobin is 50% saturated with oxygen) for fetal blood in the sheep is considerably lower (∼27 mmHg) than that of adult blood (∼38 mmHg). Reliable values for blood gases and oxygen saturations for the human fetus in utero are not available. the mean pressure in the superior and inferior vena cava and the right atrium is about 2–3 mmHg and a. Postnatally. Umbilical venous blood PCO2 is about 40 mmHg and the pH is 7.3. Effects of administering oxygen to the mother Administering 100% oxygen to the ewe raises her arterial oxygen saturation to 100% and the PO2 to more than 400 mmHg. at about 145 days’ gestation. from a mean level of 25–30 mmHg at about 60 days’ gestation to 60–70 mmHg close to term. with an oxygen saturation of about 80%. whereas descending aortic blood has a PO2 of 20–23 mmHg and an oxygen saturation of about 55%. Therefore. whereas right hepatic venous blood oxygen saturation is lower. whereas the left atrium shows a dominant v-wave. 63 RHV PS 73 DV 83 LHV 83 PV 35 40 UV 55 DAo IVC P Figure 10. Cardiac output and its distribution Postnatally blood is ejected by the left ventricle into the aorta and distributed to the tissues. Umbilical venous blood PO2 in the fetal lamb increases to 40–50 mmHg and oxygen saturation reaches 95–100%. because the sheep has a syndesmochorial placenta. The left hepatic venous blood oxygen saturation is about 75%. In the fetus. There is a large PO2 gradient across the placenta.38–7. The PO2 and oxygen saturation in superior vena cava blood are similar. it returns through the veins to the right atrium and is ejected by the right . as discussed above. the Po2 of umbilical venous blood is 32–35 mmHg.and v-wave pressures are both about 4–5 mmHg. The PO2 of right ventricular and pulmonary arterial blood is 18–20 mmHg and oxygen saturation is about 50%. This is in contrast to the inferior and superior vena cava.134 Fetal Cardiology This lack of pulsatile pressure extends into the porta hepatis. Arterial PO2 increases to only 30–35 mmHg. Blood gases and oxygen saturation Maternal arterial blood in the pregnant ewe has a PO (partial pressure of oxygen) of 90–100 mmHg and a PcO2 of about 35 mmHg. In the lamb fetus. the left atrial is higher than the right atrial mean pressure. Aortic pressure increases with gestational age in the lamb fetus. and the mean pressure is 1–2 mmHg lower than right atrial pressure. Blood in the abdominal inferior vena cava distal to the entrance of the ductus venosus and hepatic veins has a PO2 of about 12–14 mmHg and an oxygen saturation of 35–40%. Right and left systolic and end-diastolic pressures are similar. Oxygen saturations of blood in various cardiac chambers and great vessels measured in the fetal lamb are shown in Figures 10. the right ventricular and pulmonary arterial systolic pressure tends to exceed the left ventricular and aortic pressure by 5–8 mmHg during late gestation. It is possible that the maternal–fetal oxygen gradient is lower in the human fetus because the placental membrane has fewer layers.2 and 10. when PO2 of umbilical venous blood is 32–35 mmHg. at about 65%. Left atrial pressure has a contour similar to that of the right atrium. where the mean pressure is 5–6 mmHg. Umbilical venous blood has an oxygen saturation of 80–90%. The separation between the maternal and fetal circulations in the sheep is fairly broad. which show variations of pressure with the cardiac cycle.40.3 Oxygen saturations of blood in the vessels in the region of the porta hepatis and in the inferior vena cava and hepatic veins (reproduced with permission from reference 1). This is related to the difference in the blood supplying the left and right lobes of the liver. this is probably the result of mild constriction of the ductus arteriosus. oxygen saturation is about 90%. Left ventricular and ascending aortic blood has a PO2 of about 25–28 mmHg and an oxygen saturation of about 65%.39. the right atrial pressure contour shows a dominant a-wave.

and is largely directed across the tricuspid valve into the right ventricle. The blood entering the heart from the inferior vena cava includes blood from the ductus venosus.Circulation in the normal fetus and cardiovascular adaptations 135 ventricle to the pulmonary circulation and returns to the left atrium and ventricle. this blood is derived predominantly from the ductus venosus. The output of the heart is usually expressed as combined ventricular output. and the right lobe receives ∼30 ml/min per kg fetal body weight. About 115 ml/min per kg or about 25% of CVO passes through the foramen ovale to the left atrium. In chronically instrumented fetal lambs during the later months of gestation (term is about 145 days).4 The percentages of the combined ventricular output that return to the fetal heart that are ejected by each ventricle and that flow through the main vascular channels. this is ∼30% of CVO or about 135 ml/min per kg. the combined ventricular output is about 450 ml/min per kg fetal body weight. systemic and umbilical venous blood mix and the mixed blood is distributed to the various parts of the body and to the placenta. and blood flow to the fetal body is about 250 ml/ min per kg. The right ventricle ejects about 66% of 21 90 31 59 66 34 10 45 265 3 7 300 150 15 35 66 69 21 27 7 90 195 15 115 35 310 3 69 310 Placenta 110 200 Figure 10. left and right hepatic veins.5 The volumes of blood flowing through cardiac chambers and great vessels for the late-gestation fetal lamb (ml/min per kg body weight) (reproduced with permission from reference 1). about 110 ml/min per kg passes through the ductus venosus. In the fetal lamb. or about 315 ml/min per kg (Figures 10.5). Venous return from the superior vena cava is 90–95 ml/min per kg. the volumes of blood ejected by left and right ventricles are different in the fetus. blood to many organs is derived from both ventricles. The right and left lobes of the liver receive a total of about 90 ml/min per kg fetal body weight of blood from the umbilical vein. Figures represent values for late-gestation lambs (reproduced with permission from reference 1). Inferior vena cava blood distal to the entrance of the hepatic veins and ductus venosus (abdominal inferior vena cava) is derived from the lower body organs and the lower extremities as well as the lower portion of the trunk. About 200 ml/min per kg of inferior vena cava blood as well as coronary venous blood enters the right ventricle.12.4 and 10. In this circulation the volume of blood ejected by each ventricle is similar and is termed the cardiac output. In the fetus. . Unlike in the postnatal circulation.13 Umbilical– placental blood flow is about 200 ml/min per kg body weight. and constitutes about 70% of CVO. the sum of the volumes ejected by the two ventricles. as mentioned above. Thus. Although the proportions vary.5). and abdominal inferior vena cava. Figure 10. representing about 21% of CVO. The right ventricle ejects about two-thirds and the left ventricle about one-third of combined ventricular output in the fetal lamb (Figures 10. Umbilical–placental flow of about 200 ml/min per kg represents 40–45% of combined ventricular output (CVO). about 55% of umbilical venous blood passes through the ductus venosus and 45% through the hepatic circulation.4 and 10.

136

Fetal Cardiology

CVO, or about 300 ml/min per kg. Only about 10–15% of the blood ejected by the right ventricle is directed to the pulmonary circulation; this constitutes about 8% of CVO or about 35 ml/min per kg fetal weight. The remaining 58% of CVO, about 265 ml/min per kg, passes through the ductus arteriosus. The left ventricle receives the 115 ml/min per kg of blood that passes through the foramen ovale as well as about 35 ml/min per kg from pulmonary venous return. It ejects about 150 ml/min per kg, representing about 33% of CVO. Less than a third of the blood ejected by the left ventricle passes across the aortic isthmus to the descending aorta. This represents about 10% of CVO or about 45 ml/min per kg. The coronary circulation receives about 3% of CVO; about 20% of CVO, about 90 ml/min per kg, is distributed to the head, brain, upper extremities, and upper portion of the trunk. The magnitude of blood flow through the major arteries is reflected by the relative diameters of these vessels. The pulmonary trunk is very large, and the ascending aorta somewhat narrower; the descending aorta is also very wide, whereas the isthmus of the aorta is much narrower than the ascending or descending aorta and the ductus arteriosus. The combined ventricular output is about 450 ml/min per kg fetal body weight. In the sheep fetus it is fairly constant in relation to fetal body weight from about 90 to 140 days’ gestation (term is 140–145 days). In the last few days of pregnancy there is a modest fall in CVO related to body weight. This could be related to uterine contraction or other unidentified factors. The proportions of CVO distributed to different organs change during gestational development. In the fetal lamb, the placenta receives about 42–44% of CVO at 75–90 days’ gestation, but the proportion falls to 36–38% near term. The percentage distributed to the brain and lungs increases with gestational age. This change during growth is more striking when the changes in blood flow per unit of organ tissue weight are examined. Changes in organ blood flow per 100 g of weight, during development of the fetal lamb, are shown in Figure 10.6. At about 110 days (0.75 gestation), blood flow per 100 g organ weight progressively increases to the brain, gut and lungs. The cause of the increase in flow to these organs is not known; it could be related to an increase in size of the vascular bed due to the growth of new vessels, or to increased metabolic activity with vasodilatation, or a combination of these factors.

300 Heart 250

200 Flow ml/100 g/min Kidneys 150 Brain Lungs 100 Gut 50 Trunk 0 0.5 0.6 0.7 0.8 Gestation 0.9 1.0

Figure 10.6
Changes in blood flow to various organs in relation to tissue weight in the fetal lamb during the latter half of gestation (reproduced with permission from reference 1).

Circulation in the human fetus
Although the general course of the circulation in the human fetus is similar to that in the lamb, the proportions of combined ventricular output distributed to body organs differ, because the relative weights of some organs are very different. Perhaps the most important factor is brain size; in the mature human fetus the brain constitutes 12% of

body weight, as compared with 3% in the lamb. It is reasonable to assume that the blood flow related to weight is similar in the two species. Near term, both human and sheep fetal body weights are about 3.5 kg, and brain weights are about 65 g in the sheep and 350 g in the human. If it is assumed that blood flows to the brain are similar in relation to tissue weight at 120 ml/min per 100 g, total brain flows would be 80 ml/min in the sheep and 420 ml/ min in the human, or 22 and 120 ml/min per kg, respectively. Limited data are available for blood flows in the human fetus, based on Doppler flow studies.14,15 The umbilical blood flow has been reported to be about 180 ml/ min per kg estimated fetal weight16, but recently it has been proposed that umbilical flow in the human fetus is lower, at about 120–140 ml/min per kg fetal weight.17 Although reports of left and right ventricular output vary considerably, generally the combined ventricular output appears to be similar to that in the sheep fetus at about 450 ml/min per kg estimated weight. However, the ratio of right to left ventricular output, which in the sheep is about 2:1, is about 1.2–1.3:1 in the human.14,15 I have attempted to estimate the quantities of blood flowing through the cardiac chambers and great vessels in the human fetus. Unfortunately, limited information is currently available, but one publication provides estimates of pulmonary blood flow.18 Based on this report, I have assumed pulmonary blood flow in the late-gestation fetus

Circulation in the normal fetus and cardiovascular adaptations

137

31

140

10 39 175

45

3 55 45

17 250 200

15

75

31 3

21

28 17

49

140 15 Placenta

95

125 75

220

49

9

40

220

Placenta 40 180

Figure 10.7
The percentages of the combined ventricular output that return to the fetal heart that are ejected by each ventricle and that flow through the main vascular channels for the late-gestation human fetus (reproduced with permission from reference 1).

Blood flows in human fetus (ml/min/kg)

Figure 10.8
The volumes of blood (ml/min per kg) flowing through cardiac chambers and great vessels of the late-gestation human fetus (reproduced with permission from reference 1).

to be about 75 ml/min per kg fetal weight. Using this information, I have estimated the blood flow patterns in the late-gestation human infant. The percentages of CVO are shown in Figure 10.7 and the blood flows through various great vessels and ejected by each ventricle are depicted in Figure 10.8. The pattern of flow in the liver region of the human appears to be similar to that of the lamb fetus. In fetal lambs, Doppler flow studies demonstrated a blood flow velocity of 55–60 cm/s in the ductus venosus, whereas the velocity in the abdominal inferior vena cava was only about 16 cm/s.9 The ductus venosus stream was directed preferentially through the foramen ovale (see above). In the human fetus, blood flow velocity in the ductus venosus has been reported to be 65–75 cm/s, and the ductus venosus stream also preferentially flows through the foramen ovale.10,11

Flow velocity contours
Arterial flow
Patterns of blood flow in various sites of the circulation in the fetus have been studied in fetal lambs by electromagnetic

or ultrasonic means, and in human fetuses by means of Doppler ultrasound. Velocity recordings in the ascending aorta and main pulmonary trunk are similar to those in the adult. However, in the fetal lamb, there are distinct differences between aortic and pulmonary blood flow patterns. The velocity in the main pulmonary trunk rises rapidly after the onset of ejection to reach its peak early in systole. It then falls rapidly initially, but the rapid decrease in velocity is interrupted and a definite incisura in the velocity profile is frequently noted (Figure 10.9). The ascending aortic velocity shows a slower rise at the onset of ejection, and peak velocity is achieved in about midsystole. The peak of aortic velocity is close to the incisura on the downslope of the pulmonary trunk velocity tracing. The mechanisms responsible for these differences in the velocity patterns have not been defined. One possible explanation is that there are marked differences in impedance in the two circulations. The left ventricle ejects into the ascending aorta, which in the fetal lamb carries only about a third of CVO. The aortic isthmus is relatively narrow and it transmits only about 10% of CVO; it imposes some degree of obstruction to flow from the ascending to

138

Fetal Cardiology

Pulmonary trunk

PA press

QDuctus

Ascending aorta

QLPA

Figure 10.9
Blood flow velocities were recorded simultaneously from the ascending aorta and the pulmonary trunk with electromagnetic flow transducers in a late-gestation fetal lamb in utero. The calibration for the two flows is identical. Note that stroke volume, the area under the curve, is almost twice as great from the right as from the left ventricle. The differences in flow contours are discussed in the text.

Figure 10.10
Simultaneous recordings of blood flow velocities in the ductus arteriosus and left pulmonary artery (LPA), using ultrasonic flowmeters and pulmonary arterial pressure (PA Press) in a fetal lamb in utero. See text for discussion (reproduced with permission from reference 1).

the descending aorta. The arterial system into which the left ventricle ejects has a relatively low compliance and this may explain the slow rise and late peaking of velocity. The right ventricle ejects predominantly through the ductus arteriosus to the descending aorta, from which the relatively low-resistance umbilical–placental circulation arises. The ejection into this highly compliant circulation could explain the rapid rise and early peaking of the velocity. The incisura on the downslope of the pulmonary trunk velocity tracing indicates that the fall in velocity is briefly interrupted. A possible explanation for this phenomenon is that when blood ejected by the left ventricle crosses the aortic isthmus to reach the ductus arteriosus, it interferes with flow from the pulmonary trunk through the ductus and thus slows the fall in velocity momentarily. This hypothesis is supported by the fact that the incisura coincides with the peak of the aortic velocity profile. The velocity profiles for the pulmonary trunk and ascending aorta have not been described in human fetuses. They could differ considerably from those of the lamb, because the differences in left and right ventricular stroke volumes are less marked. Also, the volume of blood flowing into the cerebral circulation in the human fetus is much greater than that in the lamb, and the compliance of the arterial system into which the left ventricle ejects is considerably greater than that in the lamb. The velocity pattern of blood flow in the branch pulmonary arteries is distinctive in the fetal lamb.19 In the normal lamb, velocity increases rapidly early in systole, but forward flow ceases about half-way through systole

and a variable amount of retrograde flow occurs through the remainder of systole (Figure 10.10). No antegrade flow is evident during diastole but a small amount of retrograde flow continues through much of diastole. The factors contributing to this pattern of flow have not been fully defined, but it appears to be related to the low compliance of the large pulmonary arteries, the high resistance of the pulmonary circulation and the presence of the ductus arteriosus. During the early phase of right ventricular systole, blood flows into the main and branch pulmonary arteries, as well as through the ductus arteriosus. Owing to the high vascular resistance, flow into the lung ceases, but flow through the ductus continues. The recoil of the large pulmonary arteries results in reversed flow of blood through the ductus into the low-resistance circulation of the lower body and placenta, resulting in retrograde flow at the site of recording of the velocity profile. The interruption of the rapid decline in forward flow in the ductus is evident in the velocity tracing of the ductus, coincident with the retrograde flow in the left pulmonary artery tracing. The velocity profile in the branch pulmonary arteries is markedly altered by changes in pulmonary vascular resistance. Administration of a pulmonary vasodilator such as acetylcholine to the fetal lamb results in an increase in pulmonary blood flow. The duration of the forward flow phase in the branch pulmonary arteries is prolonged, the degree depending on the magnitude of the vasodilatation and the decrease in the duration and amount of retrograde flow (Figure 10.11). An increase in pulmonary

Circulation in the normal fetus and cardiovascular adaptations

139

systole PA pressure mmHg 75 PA pressure mmHg 75

systole

0

0

Phasic flow pulmonary artery

0

Phasic flow pulmonary artery 1 second Control

0

1 second Ach infusion

Figure 10.11
The pattern of blood flow in the left pulmonary artery is shown before (left panel) and during (right panel) infusion of acetylcholine (Ach) into the pulmonary artery. See text for discussion.

vascular resistance resulting from induced hypoxia in the lamb markedly reduces the duration and magnitude of the forward flow phase and increases the duration and degree of retrograde flow. The velocity profile in the branch pulmonary arteries of the human fetus has not been well characterized. It is likely that pulmonary blood flow is relatively higher in the human than in the lamb fetus (see above) and the forward flow phase would be longer and the retrograde flow less significant than in the lamb. Defining the velocity profile in human fetuses might provide useful information about the status of pulmonary vascular resistance.

Tracheal pressure ECG

SVC flow IVC flow SVC pressure 1 second

Venous flow
Patterns of flow in the superior and inferior venae cavae are similar, and in the normal fetus do not differ significantly from postnatal flow patterns (Figure 10.12). The flow contours are related inversely to the pressure tracing. Coincident with atrial systole, there is a short phase of retrograde flow, followed by forward flow during ventricular systole; this forward flow is enhanced at the end of ventricular systole, coincident with the phase of rapid inflow into the ventricle. As is evident in Figure 10.12, venous flow is markedly altered during fetal respiratory movements; flow is greatly enhanced during inspiratory movements and reduced during expiration. The velocity profile is markedly affected by alterations in heart rate and vascular resistance. Bradycardia results in an exaggeration

Figure 10.12
Simultaneous recordings of blood flow velocities in the superior (SVC) and inferior vena cava (IVC) by electromagnetic flow transducers placed around the vessels in the thorax, and vena cava and intratracheal pressures in a fetal lamb in utero. Note the increase in forward flow in both vessels associated with inspiratory effort (decrease in intratracheal pressure). For detailed description see text.

of retrograde flow during atrial systole, as does an increase in peripheral vascular resistance. Flow in the umbilical vein is continuous, and the phasic changes associated with the cardiac cycle are normally not evident, so that the velocity profile is flat. However, induction of hypoxia in the lamb fetus results

140

Fetal Cardiology

in some degree of phasic change in umbilical flow velocity. This is probably related to the fact that a greater proportion of umbilical venous blood traverses the ductus venosus and dilates the vessel, allowing transmission of central venous pressure.

Determinants of cardiac output
Cardiac ventricular output is the product of heart rate and stroke volume. Stroke volume is determined by preload, afterload, and myocardial contractility. Preload determines the degree to which the ventricular muscle is stretched immediately prior to contraction. In the intact heart, ventricular volume at end-diastole determines the length of the cardiac myocytes and thus sarcomere length. The greater is the length of the sarcomere, up to an optimal level, immediately before contraction occurs, the greater is the force generated during contraction. An increase in end-diastolic ventricular volume increases the force of contraction of the muscle and, in the intact heart, increases the stroke volume if other factors are unchanged. Afterload, or load on the heart muscle during development of active force, determines the degree of shortening of the sarcomeres and thus the volume ejected during systole. With the same force of ventricular contraction, the chamber will empty more if the afterload is low, and the ejection volume will be lower if the afterload is increased. In the intact circulation, afterload is influenced by several factors, such as arterial pressure, compliance of the arterial system, and peripheral vascular resistance. Contractility is the intrinsic force of contraction of the muscle; with isolated muscle, increased contractility increases the force developed and, in the intact heart, increases the stroke volume, or developed pressure. In the intact circulation, heart rate, preload, afterload, and contractility are interrelated, and a change in one factor may modify other parameters. It is therefore important to consider possible changes in these other parameters when assessing the effects of alteration of one regulatory factor.

In studies in fetal sheep, spontaneous increases in heart rate above the resting level of about 160 beats/min are associated with increases of ventricular output of up to 15–20%, and spontaneous decreases in heart rate result in a fall in output.20 In these studies, it was not apparent whether the tachycardia was directly responsible for the increase in cardiac output. It is possible that the factors inducing the increase in heart rate also affected loading conditions or contractility. The effects of electrical pacing of the right or left atrium to increase rates to 240–300 beats/min were studied in fetal lambs. Pacing the right atrium resulted in an increase of left ventricular output of up to 15%, with only a small increase or no change in right ventricular output. At rates above 300–320 beats/min, ventricular output fell progressively with increasing rate, presumably because diastolic filling time was greatly reduced. Pacing the left atrium increased right ventricular output modestly, but decreased left ventricular output. Normally in the fetus, the right atrial pressure is slightly higher than that in the left atrium throughout the cardiac cycle. During pacing, the left atrial pressure pulse is altered so that the left atrial pressure exceeds that in the right atrium during some phases of the cycle and interferes with flow through the foramen ovale into the left atrium, reducing the left ventricular filling and output. Vagal stimulation decreased the output of both ventricles by about 15–20%, associated with bradycardia. However, the decrease in output could not be ascribed entirely to the decrease in heart rate, because vagal stimulation increases systemic arterial pressure and elevates afterload; this could contribute to the fall in ventricular output.

Effects of preload and afterload
Preload and afterload are discussed together because there is usually an interaction between them in the intact circulation. If afterload is increased, the volume ejected by the ventricle during systole is reduced and residual ventricular volume increases. If ventricular filling is maintained, preload is greater with the next beat. In utero studies of fetal lambs have been performed to assess the role of preload on cardiac output. In most of these studies, ventricular end-diastolic or atrial pressures have been used as an index of preload. Pressure measurements may not, however, be a reliable indicator of volume, because ventricular compliance determines the volume at any particular pressure. Studies in isolated myocardium and intact hearts have shown that fetal myocardium is less compliant than that of the adult.21 Rapid intravenous infusions of 0.9% NaCl solution raised cardiac output associated with an increase in atrial pressure in newborn lambs.22 Cardiac output increased progressively with elevation of atrial pressure to levels of about 15 mmHg. Several investigators have studied the

Effects of heart rate
In the adult, cardiac output is relatively constant over a wide range of heart rates. Increasing the heart rate to 150 beats/min or decreasing it to 50 beats/min from a resting rate of about 70 beats/min does not alter output. Greater increases in heart rate may decrease cardiac output because the reduction in diastolic filling time does not permit adequate filling to maintain stroke volume. With very slow heart rates, stroke volume is increased to maintain cardiac output, but when maximum diastolic filling has been achieved, further slowing results in a decrease of ventricular output.

Circulation in the normal fetus and cardiovascular adaptations

141

600

1.0 LA-3.5

LA-7 LA-10

0.9 LVSV ml/kg

CVO ml/min/kg

400

0.8 0.7 0.6

200

0.5 30 40 50 60 70 Arterial pressure (mmHg) 80 90

4

8 RA pressure mmHg

12

16

Figure 10.14
When systemic arterial pressure is regulated, an increase in pressure results in a fall of left ventricular stroke volume (LVSV) at fixed left atrial (LA) pressure. At any level of arterial pressure, an increase in left atrial pressure increases left ventricular stroke volume (reproduced with permission from reference 1).

Figure 10.13
Changes in combined ventricular output (CVO) associated with acute reduction of atrial pressure by blood removal and increase of atrial pressure by infusion of electrolyte solution in fetal lambs (reproduced with permission from reference 1).

effects of decreasing or increasing preload in fetal lambs in utero.23–26 Preload was decreased by reducing fetal blood volume by removal of blood, and increased by rapid intravenous infusion of electrolyte solution. A fall in right atrial and right ventricular end-diastolic pressure resulted in a marked decrease in cardiac output. Output rose when atrial pressure increased by 2–4 mmHg above resting levels, but further increases in pressure did not result in greater output by the ventricle (Figure 10.13). This response is distinctly different from that in the postnatal lamb, in which increases of atrial pressure to levels of 15–20 mmHg are associated with a progressive increase in ventricular output. Based on these studies, it was proposed by Gilbert23,24 that the fetal heart is normally operating near the top of its ventricular function curve; the elevation of cardiac output associated with an increase in preload is limited because myocardial performance, or contractility, is relatively poor in the fetus. However, a decrease in atrial pressure reduces preload, resulting in a fall in cardiac output. In these studies, the effects of rapid infusion of electrolytes on arterial pressure were not considered. Associated with the infusion, fetal arterial pressure also increases and thus changes afterload. We examined the effects of changing preload at various constant levels of arterial pressure.27 Arterial pressure elevation dramatically reduced left ventricular stroke volume at all levels of mean atrial pressure (Figure 10.14). At constant arterial pressure levels, progressive elevation of left atrial pressure increased left ventricular stroke volume even with atrial pressures of 10–12 mmHg. This study demonstrated that the fetal heart responds to increases in preload by increasing its output.

It did not, however, resolve whether performances of the fetal and adult myocardium are comparable.

Myocardial performance
Studies of isolated myocardium from fetal and adult sheep have demonstrated that fetal myocardium develops less active tension than adult myocardium at similar muscle lengths.28 Also, the maximum force that can be generated is considerably lower for fetal than for adult myocardium. Several differences in morphological and biochemical parameters of myocardium have been described that could account for the lesser contractility of fetal myocardium. It was suggested that fetal myocardium contains fewer sarcomeres, or contractile units, in each myocyte. Another factor that may be important is development of the sarcoplasmic reticulum, which regulates the movement of calcium ions, essential for myocardial contraction. The fetal myocardial sarcoplasmic reticulum is well developed, but the T-tubular system, representing the extension of the sarcoplasmic reticulum to provide closer relations with the contractile elements, is either poorly developed or absent in the immature myocardium. Not only are there structural differences in the sarcoplasmic reticulum, but, in studies with isolated sarcoplasmic reticulum vesicles, calcium uptake was found to be impaired in fetal myocardium.29 Local release of norepinephrine (noradrenaline) at sympathetic nerve endings is an important mechanism for increasing myocardial contractility. Sympathetic nerve endings are sparse or even absent in fetal myocardium.

142

Fetal Cardiology

The abundance of sympathetic nerve endings varies greatly during development in different species. In the guinea pig, myocardial sympathetic innervation is almost fully developed at birth,30 whereas in the rabbit and the rat there is almost no innervation at birth, but it develops within 14–21 days after birth.31 The sheep fetus has no detectable sympathetic innervation at 75 days (mid-gestation), but innervation begins to appear at 90–100 days, and is abundant but not yet fully developed just before birth.32 In addition to the difference in sympathetic innervation, β-adrenergic receptor concentration is lower in fetal than in adult myocardium.33 Although these differences in sympathetic innervation and β-adrenoreceptor concentration may not be important in the resting fetal heart, they could influence the ability to respond to stress.

vasculature. This functional separation of the aorta at the isthmus has been demonstrated in fetal lambs. A rapid reduction in peripheral vascular resistance in the lower body circulation, induced by a vasodilator, causes a decrease in descending aortic pressure and increase in right ventricular stroke volume for several beats, whereas the ascending aortic pressure and left ventricular output do not change. Similarly, injection of a vasodilator into the ascending aorta causes an evanescent decrease of ascending aortic pressure and increase in left ventricular stroke volume.

Reflex regulation
Chemoreflexes Previous studies on the role of chemoreflexes in control of the fetal circulation were conflicting. Some investigators suggested that aortic and carotid chemoreceptors are relatively inactive in the fetus, but these studies were performed in anesthetized exteriorized fetal lambs.34 Other studies indicated that aortic receptors were important in causing a bradycardic response to hypoxia.35 More recent studies in fetal lambs have shown that they are active, at least in the last third of gestation.36 Responses to carotid chemoreceptor stimulation are much greater than to aortic receptor stimulation.37 The chemoreceptors are stimulated by hypoxemia, and experimentally can be activated by intravascular injection of small doses of sodium cyanide. The cardiovascular response dominates, with bradycardia and immediate hypotension, but respiratory gasps are noted. The bradycardia can be abolished if the lambs are pretreated with atropine, indicating that the bradycardia is induced by vagus nerve stimulation. Confirmation of the fact that the cyanide response is the result of chemoreceptor stimulation was obtained by demonstrating the loss of the cardiovascular and respiratory responses in fetal lambs in which sinoaortic denervation had been accomplished.37 In the adult, chemoreceptor stimulation results in reflex peripheral vasoconstriction. It is likely that the peripheral vasoconstriction induced by fetal hypoxemia is largely mediated by chemoreceptor stimulation. From the studies in fetal lambs it is apparent that their chemoreflex responses are different from those in the adult. The respiratory response in the adult animal dominates, whereas chemoreceptor stimulation in the fetus causes only a minor respiratory response. There is as yet no explanation for this difference in response.

Regulation of the circulation in the fetus
In the adult the systemic and pulmonary circulations are separate. Each ventricle is subjected to a potentially different preload and afterload, and the stroke volume of each ventricle might vary greatly. The Frank–Starling mechanism is useful for adjusting the outputs of the two ventricles so that over a short period the ventricles eject similar volumes. A reduction in venous return to the right atrium reduces the filling pressure and end-diastolic volume of the right ventricle, resulting in a decrease of stroke volume. Pulmonary blood flow and venous return to the left atrium and ventricle is reduced and stroke volume falls. An increase in systemic arterial pressure will restrict left ventricular stroke volume; end-diastolic volume will increase so that, with the next beat, greater force is generated to increase stroke volume. In the fetus the presence of the foramen ovale tends to make right and left atrial pressures equal throughout the cardiac cycle. The ductus arteriosus provides a large communication between the aorta and the pulmonary artery, resulting in almost identical pressures in the two vessels. In view of the similar atrial pressures and similar aortic and pulmonary arterial pressures, differences in stroke volumes of the left and right ventricles in the fetal lamb are probably due to differences in afterload on the ventricles. The aortic isthmus, which in the fetus is narrower than the ascending and descending aorta, to some extent functionally separates the upper and lower body circulations. The left ventricle ejects into the ascending aorta and the vessels of the head and neck, a circulation that in the lamb fetus is poorly compliant and has a relatively high vascular resistance. The right ventricle ejects into the pulmonary trunk and directly through the large ductus arteriosus into the descending aorta and its branches. This circulation has a higher compliance and a lower resistance because it includes the umbilical–placental

Baroreflexes In the adult, arterial pressure is maintained over a fairly narrow range through the control of baroreceptors. Stimulation of aortic and carotid baroreceptors by a rise in arterial pressure induces bradycardia, depression of

Circulation in the normal fetus and cardiovascular adaptations

143

myocardial contractility, and peripheral vasodilatation, all of which tend to decrease arterial pressure. Ablation of aortic and carotid baroreceptors, by bilateral section of the aortic and carotid afferent nerves, results in an initial increase in resting heart rate and arterial pressure, but within 1–2 days these parameters return to average levels that were present during the pre-denervation period. Wide swings of arterial pressure and heart rate occur around the average pressure and rate, in association with stimuli that produce only small changes in the normal animal.38 Arterial baroreceptors are functional in the fetus relatively early in gestation, but their importance in regulating fetal arterial pressure has been questioned. In fetal lambs, baroreflex sensitivity increases with gestational age from about 80 days’ gestation; near term gestation, the bradycardia induced by increased arterial pressure is equal to that noted postnatally. In fetal lambs, sinoaortic denervation results in the same wide variation in heart rate and blood pressure as observed in adult animals.39 It is thus apparent that baroreflexes are important in stabilizing arterial blood pressure in the fetus as well.

Birth-associated changes in the circulation
Delivery of the fetus from the uterus disrupts the umbilical–placental circulation. The functions of oxygen uptake and carbon dioxide removal are transferred to the lungs. Pulmonary ventilation has to be established to provide gas exchange. During fetal life, pulmonary blood flow is relatively low, and has to increase to allow oxygen uptake adequate for postnatal survival. The well-oxygenated blood from the umbilical veins and poorly oxygenated blood from the vena cava mix partially and venous blood is diverted away from the lungs through the foramen ovale and the ductus arteriosus. In the adult, blood circulates in series. All venous blood is returned to the right atrium and ventricle, ejected into the lungs where it is oxygenated, and then passes to the left atrium and ventricle to be ejected into the systemic arterial circulation. Apart from the return of minor amounts of venous blood into the left ventricle via Thebesian veins, there is no mixing of arterial and venous blood. The foramen ovale and ductus arteriosus have to be closed functionally or anatomically to establish the adult circulation. As the fetus is delivered, several events occur in a short time period. Fluid in the fetal airways is removed either by expulsion through the mouth as a result of chest compression, or by absorption into the pulmonary circulation with the onset of breathing. Regular ventilation is established, and the umbilical–placental circulation is terminated by disruption or clamping of the umbilical

cord. Ventilation by room air is associated with an increase in the alveolar oxygen concentration and also with the rhythmic physical expansion of the lung and removal of alveolar fluid. It has been difficult to assess the role of each of these factors in contributing to the circulatory changes associated with birth, because they occur almost simultaneously. We developed a fetal lamb preparation to examine the individual role of birth events in these changes.40 Catheters were implanted in various fetal vessels, a tube was inserted into the trachea, and an inflatable balloon occluder was placed around the umbilical cord. All the catheters were exteriorized to the maternal flank and the ewe and fetus allowed to recover from surgery. Fetal vascular pressures and blood gases were monitored and blood flows were measured repeatedly by the radionuclidelabeled microsphere technique. The effect of rhythmic expansion of the lung was assessed by ventilating the fetus with a gas mixture of 5% carbon dioxide, 3% oxygen, and 92% nitrogen. This did not significantly change fetal blood gas levels of PO2 21 mmHg and Pco2 40 mmHg in descending aortic blood. The fetus was then ventilated with 100% oxygen; this raised fetal descending aortic PO2 to about 50 mmHg and oxygen saturation to above 90%. With the fetus well oxygenated, the effect of occluding the umbilical cord was then assessed. The proportions of CVO ejected by each ventricle and distributed to the major vessels are shown in Figure 10.7. Rhythmic ventilation without altering fetal blood gases produced a considerable increase in pulmonary blood flow (Figure 10.15) and decrease in pulmonary vascular resistance (Figure 10.16).41 The proportion of CVO passing to the lungs increased from 9 to 31%. Interestingly, pulmonary arterial pressure did not fall, suggesting that

2500 Pulmonary blood flow (ml/100 g/min)

2000

1500

1000

500

0 Control Lung expansion Oxygenation O2 + cord occlusion

Figure 10.15
Changes in pulmonary blood flow resulting from physical expansion of the lung, ventilation with oxygen, and umbilical cord occlusion in fetal lambs (reproduced with permission from reference 1).

144

Fetal Cardiology

1.2 1.0 PVR mmHg/ml/100 g/min

18

27 45

0.8 24 0.6 0.4 0.2 18 Control Lung expansion Oxygenation O2 + cord occlusion 3 48 52 17 31 52 48 3 3 31

48

Figure 10.16
Changes in pulmonary vascular resistance (PVR) resulting from physical expansion of the lung, ventilation with oxygen, and umbilical cord occlusion in fetal lambs (reproduced with permission from reference 1).

Figure 10.17
Proportions of combined ventricular output ejected by the right and left ventricles and flowing through great vessels in fetal lambs during ventilation without changing fetal blood gases (reproduced with permission from reference 40).

the ductus arteriosus was still widely patent and that aortic pressure was transmitted to the pulmonary artery. Associated with the increased pulmonary blood flow, pulmonary venous return to the left atrium increased and the proportion of CVO passing through the foramen ovale decreased (Figure 10.17). Although CVO did not change, right ventricular output constituted about 52% of CVO, as compared with 65% in the unventilated fetus. Also, only 24% of CVO passed through the ductus arteriosus to the descending aorta, compared with 57% in the control state. Left ventricular output increased from 34 to 48% of CVO, so that the outputs of the two ventricles were now similar. Ventilation with oxygen resulted in a further decline in pulmonary vascular resistance and rise in pulmonary blood flow (Figure 10.18). Only a minor proportion of blood ejected by the right ventricle passed through the ductus arteriosus to the descending aorta, almost all being distributed to the pulmonary circulation. The large venous return to the left atrium elevated left atrial pressure above that in the systemic veins and right atrium; this resulted in closure of the foramen ovale, with only insignificant flow from the right to the left atrium. Total CVO did not change significantly, but left ventricular output exceeded right ventricular output, with 55% of CVO contributed by the left and 45% by the right ventricle. This higher output by the left ventricle was the result of the development of a shunt from the aorta to the pulmonary artery through the still open ductus arteriosus, constituting about 10% of CVO. Pulmonary arterial pressure gradually and progressively decreased below aortic levels. This reflected the separation of systemic and pulmonary arteries by constriction of the ductus arteriosus.

10

43 53 4 11 45 55 45 10 2 36 3 52 2 52

36

Figure 10.18
Proportions of combined ventricular output ejected by the right and left ventricles and flowing through great vessels in fetal lambs during ventilation with oxygen (reproduced with permission from reference 40).

Occlusion of the umbilical cord completely eliminated umbilical blood flow. It resulted in a modest increase in systemic arterial pressure and a small increase in the shunt through the ductus arteriosus from the aorta to the pulmonary artery. However, no other additional changes

Circulation in the normal fetus and cardiovascular adaptations

145

14

43 57 2 16 41 59 41 14 2 29 4 55 2 55

29

Figure 10.19
Proportions of combined ventricular output ejected by the right and left ventricles and flowing through great vessels in fetal lambs after occlusion of the umbilical cord during ventilation with oxygen (reproduced with permission from reference 40).

occurred and CVO was still similar to that in the control fetal state (Figure 10.19). From these studies it is apparent that the dramatic decrease in pulmonary vascular resistance resulting from ventilation of the lungs is the dominant factor contributing to the circulatory changes during the perinatal period. The elevation in left atrial pressure resulting from increased pulmonary venous return to the left atrium closes the foramen ovale. The cessation of umbilical venous return may also contribute to closure of the foramen. Constriction of the ductus arteriosus (vide infra) completes the separation between the left and right sides of the heart and the major arteries, resulting in the series circulation characteristic of the adult.

Perinatal changes in the pulmonary circulation
Rhythmic physical expansion of the lungs and an increase in oxygen levels in the ventilating gas mixture have independent but complementary roles in pulmonary vasodilatation. The mechanisms by which these processes reduce pulmonary vascular resistance have been investigated, but are not yet fully resolved. There is considerable evidence suggesting that rhythmic expansion of the lungs results in the production of the prostaglandin, prostacyclin (PGI2), probably from endothelial cells.42 PGI2 is a pulmonary vasodilator and could be responsible for the effect of ventilation. Although inhibition of prostaglandin production

in the sheep fetus limits the degree of pulmonary vasodilatation with ventilation, it does not completely prevent it, suggesting that other factors may be involved. The possibility should be entertained that the removal of fluid present in fetal airways and the replacement with gas could contribute to a decrease in pulmonary vascular resistance based on physical phenomena alone. During fetal life, the alveoli contain fluid, and the positive pressure from the amniotic cavity is transmitted through the thorax to the lungs. This would tend to compress the pulmonary vessels alongside the alveoli and small bronchi. During delivery of the fetus, airway fluid is removed; spontaneous breathing results in the development of a negative intrapleural pressure, with a gradient from the airways outward to the pleura, and this tends to dilate pulmonary vessels. However, as shown experimentally in fetal lambs, positive-pressure ventilation with no change in blood gas concentrations also reduces pulmonary vascular resistance. Physical factors that could possibly contribute to this are changes in surface forces on the alveoli. During fetal life, there is a fluid–fluid interface on the alveolar surface with no significant surface tension forces. When the alveoli are filled with gas, a strong surface tension at the gas–fluid interface tends to collapse the alveoli. This would result in a force tending to dilate pulmonary vessels associated with the alveoli, and thus decrease pulmonary vascular resistance. During fetal life, the pulmonary vessels are exposed to the PO2 of blood in the pulmonary arteries, which, in fetal lambs, is about 18 mmHg. Fetal pulmonary vessels are markedly constricted when PO2 is reduced below and dilated by an increase of PO2 above control levels. Ventilation with air increases the PO2 in precapillary pulmonary vessels, because oxygen diffuses into these vessels from surrounding alveoli, resulting in vasodilatation. It has been suggested that the response to changes in PO2 may be due to a direct effect on smooth muscle cells, and could be related to transmembrane movement of potassium via oxygen-sensitive potassium channels. A reduction in PO2 blocks potassium channels and results in constriction, whereas a rise in PO2 opens the channels, causing vasodilatation. The role of potassium channels in the response is supported by studies in fetal lambs showing that potassium channel blockers cause pulmonary vasoconstriction.43 The vasodilator effect of oxygen on pulmonary arterioles has also been shown to be associated with nitric oxide (NO) mechanisms. NO production by endothelial cells, which is associated with an increase in PO2, induces relaxation of vascular smooth muscle cells, resulting in a fall in pulmonary vascular resistance. N-nitro-L-arginine inhibits NO production; in studies in fetal lambs, it markedly limited the reduction in pulmonary vascular resistance associated with oxygenation.44 In the lamb, the rapid increase in pulmonary blood flow with ventilation is associated with a drop in pulmonary

146

Fetal Cardiology

vascular resistance from the fetal level of 1.6 to 0.3 mmHg/ ml/min/kg. Functional responses of the pulmonary circulation are considerably greater in the fetus than in the adult, due to differences in morphology of the small pulmonary arteries. The pre-acinar arteries in the fetal lung have a thick wall with a prominent smooth muscle layer. The intra-acinar arteries, associated with bronchioli, are partly muscular or non-muscular, and more distal vessels do not have smooth muscle cells. The muscular medial layer does not change significantly during the latter half of gestation, but the number of intra-acinar and alveolar duct arteries increases with lung growth during fetal development. After birth, the pulmonary arterioles have a thinner muscular media as a result of an increase in lumen size. Subsequently the smooth muscle layer gradually regresses, so that pre-acinar arteries develop the morphological features of adult vessels with the thin wall and large lumen/wall ratio. These changes occur over several weeks. The normal development of the pulmonary circulation before and after birth may be influenced by alteration of the PO2 and the intraluminal pressure to which the arteries are exposed. Interference with normal oxygenation after birth may delay the normal regression of smooth muscle. Individuals born and continuing to live at high altitude are exposed to a lower partial pressure of oxygen in inspired air; they retain a greater amount of smooth muscle in pulmonary vessels and have higher pulmonary arterial pressures than individuals at sea level.45 An increase of pulmonary arterial pressure in the fetus, as may occur with constriction of the ductus arteriosus, results in increased development of the pulmonary arteriolar smooth muscle layer.46 The greater amount of smooth muscle may interfere with postnatal adaptation, and the fall in pulmonary arterial pressure may be slower than normal as a result of the higher pulmonary vascular resistance. After birth, pulmonary arterial pressure may not fall normally if there is a congenital cardiovascular malformation with a large communication between the left and right ventricles or the aorta and pulmonary artery. The large communication results in a tendency for pressures on the left and right sides of the heart to equalize, and thus pulmonary arterial pressure does not fall as in the normal infant. Persistence of the high pulmonary arterial pressure delays normal maturation of the pulmonary vessels and the smooth muscle component persists. The role of the pulmonary circulation in the hemodynamic and clinical manifestations of congenital cardiovascular malformations is discussed in Chapter 39.

postnatal levels, and pulmonary blood flow adequate for oxygen needs may not be established. This phenomenon has been named persistent pulmonary hypertension of the newborn’, and may result from several conditions. An inability to establish normal respiration after birth will interfere with the expansion of alveoli with air, and thus the PO2 will not be increased normally and pulmonary vasoconstriction will persist. This may occur in babies who are depressed as a result of sedative drugs given to the mother. It may also result from an obstruction to airways, as with meconium aspiration. In these infants, development of the lung and pulmonary vasculature may be normal, but pulmonary vasoconstriction persists because alveolar oxygen concentration does not increase. Relief of airway obstruction or stimulation of breathing results in a rapid fall of pulmonary vascular resistance. Failure of the normal postnatal decrease in pulmonary vascular resistance may be the result of abnormal prenatal development of the pulmonary resistance arteries. As mentioned above, an increase of pulmonary arterial pressure in the fetus, as may result from constriction of the ductus arteriosus, induces increased development of the pulmonary vascular smooth muscle.47 This may interfere with the achievement of normal postnatal pulmonary vascular resistance. These vessels are exquisitely sensitive to changes in PO2, and even mild degrees of hypoxemia may result in marked pulmonary vasoconstriction. The muscle does regress slowly postnatally, but normal pulmonary arterial pressure and flow after birth may not be achieved for several weeks. Persistent pulmonary hypertension of the newborn may occur in infants when the fetus has been exposed to indomethacin in utero, because prostaglandin inhibition causes ductus arteriosus constriction.46 Experimental studies have suggested that prolonged fetal hypoxia may also induce an increase in pulmonary vascular smooth muscle. Persistent pulmonary hypertension of the newborn may be associated with inadequate cross-sectional area of the pulmonary vascular bed. This is most commonly the result of an interference with lung development resulting from encroachment of a space-occupying lesion in the thorax, such as a large lung cyst, or by intestine herniating through a diaphragmatic hernia. It may also be the result of agenesis of a lung. Since pulmonary vascular development parallels the growth of alveolar units, lack of lung development will be associated with a reduced size of the pulmonary vascular bed. Adequate development of the pulmonary circulation after birth will occur only slowly, as new alveolar units are added with growth of the lung.

Persistent pulmonary hypertension of the newborn
If pulmonary vascular resistance does not fall normally after birth, pulmonary arterial pressure will not drop to normal

Ductus arteriosus closure after birth
The ductus arteriosus connects the pulmonary trunk, before origin of the left and right pulmonary arteries, to

Circulation in the normal fetus and cardiovascular adaptations

147

the descending aorta, just beyond the origin of the left subclavian artery, in the human fetus. The wall of the ductus is morphologically quite different from that of the aorta and pulmonary artery. Whereas the walls of these arteries are largely composed of elastic tissue, the predominant tissue in media of the ductus is smooth muscle. During fetal life, the ductus arteriosus diverts a major proportion of right ventricular output away from the lungs to the descending aorta. In the sheep fetus, right ventricular output is about 66% of combined ventricular output and almost 90% of right ventricular blood passes through the ductus arteriosus. Estimates of flow through the ductus arteriosus, measured by ultrasound in human fetuses, show considerable variation. The proportion of right ventricular output distributed to the lungs is greater in the human, and it is estimated that about 60% of the blood ejected by the right ventricle passes though the ductus. The ductus remains widely patent throughout gestation, and no pressure gradient between the pulmonary trunk and the descending aorta can be detected. However, during the latter weeks of gestation, mild constriction may occur, as evidenced by a 5–8-mmHg drop in systolic pressure across the ductus. After birth, the ductus constricts rapidly; the rate of constriction appears to vary in different species. In the rat, rabbit, and guinea pig it is essentially closed within minutes. In the sheep the process is somewhat slower, closure usually being achieved within an hour. In full-term human infants, functional closure usually occurs within 12–15 hours. During the first few hours, a bidirectional shunt may be detected by ultrasound, but after about 6 hours, only a small left to right shunt may occur for up to about 15 hours. Prior to complete closure, the ductus responds to a decrease of PO2 by dilatation and, with the increase of pulmonary arterial pressure resulting from pulmonary vasoconstriction, some degree of shunting from the pulmonary artery to the aorta may again occur. The PO2 to which the ductus arteriosus is subjected is an important determinant of the degree of constriction. In the fetus, right ventricular blood with a Po2 of about 18 mmHg in the lamb passes through the ductus to the descending aorta. The ductus arteriosus is constricted by an increase in oxygen levels. This has been observed in ductus rings in a tissue bath48 or in isolated perfused ductus preparations,49 as well as in fetuses in utero. In a tissue bath, ductus preparations are relaxed at Po2 of 25–30 mmHg; they show a progressive increase in the degree of constriction from PO2 of about 40 to 100 mmHg. With air breathing, the Po2 of arterial blood increases to 90–100 mmHg, but the mechanism by which oxygen constricts the ductus has not yet been resolved. The magnitude of response to oxygen is dependent on gestational age. The more immature is the fetus, the less is the constrictor response, and the level of PO2 required to initiate constriction is greater. The ductus arteriosus has also been shown to be very sensitive to prostagandins. The ductus is relaxed by

prostaglandin E2 and prostaglandin I2 (prostacyclin); both are produced by the ductus wall, but they are also produced elsewhere in the fetus, and circulating levels of PGE2 are elevated as compared with postnatally. Although large amounts of PGI2 are produced by the ductus, PGE2 is probably more important in regulating its tone, because the ductus muscle is much more sensitive to PGE2. Prostaglandins are synthesized from arachidonic acid by the cyclo-oxygenase enzymes. Inhibition of prostaglandin synthesis by agents such as aspirin or indomethacin in the lamb fetus results in constriction of the ductus, confirming the fact that prostaglandins are important in maintaining its patency prenatally. Blood PGE2 concentrations fall rapidly after birth, and the relative contributions of the removal of PGE2 and the increase in blood to ductus constriction after birth have yet to be resolved. Studies by Clyman et al indicate that the rise in PO2 is more important than the fall in PGE2 levels in late gestation, whereas in the second and early part of the third trimester, the ductus is more sensitive to the removal of prostaglandin than to the increase in PO2.50 Studies in fetal lambs indicate that the sensitivity of the ductus to oxygen and PGE2 can be matured by administering corticosteroids to the immature fetus. Recently, the role of nitric oxide in influencing the ductus has been demonstrated. Nitric oxide relaxes the ductus muscle. In prematurely delivered baboons, a cyclooxygenase inhibitor did not induce complete closure of the ductus, but combination with an inhibitor of nitric oxide production caused complete closure.51 The role of nitric oxide in normal regulation of the ductus is yet to be determined. The mechanisms by which permanent closure of the ductus arteriosus is achieved have not yet been fully resolved. Constriction results in thickening of the intima and the development of intimal mounds that encroach on the lumen. Disruption of the internal elastic lamina results in migration of endothelial and smooth muscle cells. Clyman et al have suggested that the middle layers of the ductus receive oxygen supply from blood in the lumen. Constriction results in thickening of the wall and severe hypoxia of the middle portion of the wall.52 This results in cell damage, with replacement by fibrous tissue. Permanent closure of the ductus is usually complete within a week, but may not occur for up to 3 weeks in some infants. While the ductus is still patent, a small shunt from the aorta to the pulmonary artery (left to right shunt) may be detected by ultrasound.

The ductus arteriosus in the preterm infant
Postnatal closure of the ductus arteriosus is frequently delayed in infants born prematurely. The younger is the gestational age at birth, the more likely it is that the ductus

148

Fetal Cardiology

will not close soon after birth. In preterm infants with birth weights under 750 g, more than 80% have persistent patency of the ductus beyond the third day after birth. In infants with birth weights of 1000–1500 g, the incidence of persistent patency of the ductus is 40–50%, whereas in infants weighing 1500–1750 g at birth the incidence is about 20%. Most of the more immature infants also have severe respiratory distress syndrome. Several possible explanations for delayed closure of the ductus in preterm infants have been proposed. It was first suggested that the preterm infant did not achieve an adequate elevation of arterial PO2 due to poor ventilation, but this possibility has been excluded because, with assisted ventilation, PO2 may be raised to levels normal for mature infants and ductus patency persists. As mentioned above, the ductus arteriosus of the immature animal is more sensitive to the effect of removal of prostaglandins than to an increase of PO2, and even the high PO2 achieved may not induce complete closure. The possibility that PGE2 levels do not fall postnatally as rapidly as in the mature infant has also been examined. Results of these studies indicate that there is a delay in the fall of blood PGE2 concentrations after birth in preterm infants, and levels may be elevated for as long as 2–4 weeks postnatally. However, although PGE2 concentrations have been elevated in many infants with respiratory distress syndrome, only some had ductus arteriosus patency. The hemodynamic and clinical consequences of patency of the ductus in preterm infants are discussed in Chapter 39.

Postnatal changes in cardiac output
The combined output (CVO) of the left and right ventricles is about 400–450 ml/min per kg body weight in term fetal lambs in utero. Measurements in newborn lambs showed cardiac output levels of 400–450 ml/min per kg. Since left and right ventricular outputs are the same postnatally, the output of each ventricle is about 400–450 ml/ min per kg and combined output is 800–900 ml/min per kg, about twice that in the fetus. In the fetus, right ventricular output is about 66% of CVO, or about 300 ml/min per kg; it increases by about 50% after birth. The left ventricle, which in the fetus ejects about 150 ml/min per kg, increases its output almost three-fold to 400–450 ml/ min per kg after birth. The factors contributing to this rise of cardiac output postnatally have not been fully assessed. The studies mentioned above in fetal lambs in utero, in which we sequentially induced physical expansion of the lungs, oxygenation and umbilical cord occlusion, indicated that none of these events resulted in a rise of CVO. The proportion of CVO ejected by the left ventricle increased, and that by the right ventricle decreased. The possibility was considered that delivery from the in utero

environment with a temperature of about 39°C into a room air environment with a temperature of about 25°C could contribute to the increase in CVO. In studies in which we measured CVO in non-breathing fetal lambs delivered into a water bath but with no alteration of umbilical–placental flow, changing the bath temperature from 37°C to 25°C did not result in a significant change of CVO.53 In these experimental studies, the fetuses did not breathe spontaneously and were not exposed to the stress of delivery. It has been proposed that fetal cardiac output is regulated by a high pericardial pressure transmitted from the intrauterine cavity across the thorax and fluidfilled lungs; this high pericardial pressure restricts filling of the ventricles and limits stroke volume. Spontaneous ventilation after birth results in the development of a negative intrapleural (relative to atmospheric) pressure, also creating a negative intrapericardial pressure, which facilitates filling of the ventricles. This greater diastolic filling of the ventricles would, if myocardial function is adequate, result in a higher stroke volume and greater ventricular output. Several hormones are recognized as having important roles in the perinatal circulatory adjustments. Plasma catecholamine concentrations increase during delivery and could, by their effect of increasing myocardial contractility, facilitate the increase in cardiac output occurring after natural vaginal delivery. In the experimental studies the fetuses were not exposed to the stress of delivery, and thus may not have experienced catecholamine stimulation of the myocardium. In sheep, plasma cortisol concentrations increase slowly from about 120 days’ gestation, and 2–3 days prior to delivery (at about 150 days) they rise sharply, several-fold. Cortisol plays an important role in maturing the myocardium in the perinatal period; it has been shown to reduce nuclear proliferation and to increase protein concentration in fetal myocytes; this could be a factor in providing an increase in cardiac output after birth.54 Thyroid hormone has long been known to affect the myocardium in adults. Deficiency of the thyroid results in a depression of myocardial performance. The number of β-adrenoreceptors in the myocardium is greatly reduced in adult animals with thyroid deficiency. Although decreased response to catecholamine stimulation resulting from reduced β-adrenoreceptors could well contribute to the decreased function, thyroid hormone may have additional effects by modifying heavy chain myosin expression. The role of thyroid hormone on cardiac performance in the perinatal period has been demonstrated in sheep. In fetal lambs, plasma tri-iodothyronine (T3) concentration is about 1.0 ng/ml. After vaginal delivery it rises to about 4.0 ng/ml within 30–60 min. This is unlikely to be a factor in the rapid increase in cardiac output, because, in adults, the effect of thyroid hormone is noted within days rather than minutes. In studies of fetal lambs, we observed

Circulation in the normal fetus and cardiovascular adaptations

149

that complete thyroidectomy performed just prior to delivery resulted in no increase in T3 concentrations, but the lambs showed the expected increase in cardiac output.55 However, if thyroidectomy was done about 10 days before delivery, fetal T3 levels were undetectable, and after delivery, the lambs showed a limitation of cardiac output as well as a blunted response to catecholamine infusion. This suggested that T3 was important prenatally for normal myocardial development, and we showed that β-adrenoreceptor numbers in ventricular muscle were significantly reduced.56

Morphological changes in the myocardium after birth
Histological studies of adult and fetal myocardium show dramatic differences. The adult myocyte in the sheep heart has a diameter of about 15–20 μm, whereas in the fetal lamb heart myocytes are much smaller, with a diameter of 5–7 μm. The nucleus in adult myocytes is relatively small, and polyploidy is very common. In the fetal myocyte the nucleus is relatively larger, and most cells have a single nucleus. Observations in fetal lambs over the latter half of gestation show no significant change in myocyte diameter.1 Since the weight of the heart increases greatly, the increase in muscle mass is almost exclusively by an increase in cell numbers, or by hyperplasia. Postnatally, myocyte size increases dramatically and almost all growth is the result of hypertrophy; minimal mitosis occurs postnatally. Measurements of the DNA and protein content of the myocardium during fetal and postnatal life confirm this difference in growth patterns. DNA concentration reflects the number of nuclei in tissues; protein concentration reflects the total tissue mass. A high DNA/protein ratio suggests a relatively large number of nuclei, indicating that cells are small. The reverse suggests that the cells are large relative to nuclear numbers. During fetal life, the lamb heart shows a relatively high DNA/protein ratio, but after birth the DNA/protein ratio falls, reflecting the cessation of mitosis and the increase in myocyte size. Although all the factors responsible for the dramatic change in the pattern of myocardial growth after birth are not determined, cortisol at least appears to be important. In studies in fetal lambs, in which we infused cortisol into the left coronary artery for up to 96 hours in utero, the DNA/protein ratio of the myocardium fell, in a similar manner to that noted normally after birth.56

the liver and 50% passes through the ductus venosus.57 Portal venous flow in the fetal lamb is quite low, and almost all is distributed to the right lobe of the liver, with less than 10% passing through the ductus venosus. The liver receives a very large blood supply of almost 450 ml/min per 100 g liver weight. After birth, umbilical blood flow ceases, and apart from a small amount of flow from the hepatic artery, all hepatic flow is derived from the portal vein. After birth the hepatic blood flow is about 100 ml/min per 100 g liver weight; the flow increases to 140 ml/min per 100 g and then increases rapidly after feeding to about 300 ml/min per kg liver weight.58 The ductus venosus had been thought to react passively to the intraluminal pressure, but it was shown that prostaglandin is partly responsible for maintaining patency of the ductus venosus. Postnatally, removal of prostaglandins and cessation of flow from the umbilical vein contribute to closure of the ductus venosus. Soon after birth, considerable proportions of portal venous blood may pass through the ductus venosus, but by 3–4 days this becomes negligible, and by about 6–10 days after birth the ductus closes.

References
1. Rudolph AM. Congenital Diseases of the Heart. Armonk, NY: Futura, 2001. 2. Bristow J, Rudolph AM, Itskovitz J. A preparation for studying liver blood flow, oxygen consumption, and metabolism in the fetal lamb in utero. J Dev Physiol 1981; 3: 255–66. 3. Rudolph AM. Hepatic and ductus venosus blood flows during fetal life. Hepatology 1983; 3: 254–8. 4. Lind J, Stern L, Wegelius C. Human Foetal and Neonatal Circulation. Springfield, IL: CC Thomas, 1964. 5. Reuss ML, Rudolph AM, Heymann MA. Selective distribution of microspheres injected into the umbilical veins and inferior venae cavae of fetal sheep. Am J Obstet Gynecol 1981; 141: 427–32. 6. Edelstone DI, Rudolph AM. Preferential streaming of ductus venosus blood to the brain and heart in fetal lambs. Am J Physiol 1979; 237: H724–9. 7. Edelstone DI, Rudolph AM, Heymann MA. Liver and ductus venosus blood flows in fetal lambs in utero. Circ Res 1978; 42: 426–33. 8. Rudolph AM, Heymann MA, Teramo KAW, Barrett CT, Raiha NCR. Studies on the circulation of the previable human fetus. Pediatr Res 1971; 5: 452–65. 9. Schmidt KG, Silverman NH, Rudolph AM. Assessment of flow events at the ductus venosus–inferior vena cava junction and at the foramen ovale in fetal sheep by use of multimodal ultrasound. Circulation 1996; 93: 826–33. 10. Kiserud T. Hemodynamics of the ductus venosus. Eur J Obstet Gynecol Reprod Biol 1999; 84: 139–47.

Postnatal changes in hepatic and ductus venosus blood flow after birth
Prenatally, umbilical venous blood enters the porta hepatis; about 50% is distributed to the left and right lobes of

150

Fetal Cardiology

11. Kiserud T. Fetal venous circulation—an update on hemodynamics. J Perinat Med 2000; 28: 90–6. 12. Rudolph AM, Heymann MA. The circulation of the fetus in utero. Methods for studying distribution of blood flow, cardiac output and organ blood flow. Circ Res 1967; 21: 163–8. 13. Rudolph AM, Heymann MA. Circulatory changes during growth in the fetal lamb. Circ Res 1970; 26: 289–99. 14. Sutton MG, Plappert T, Doubilet P. Relationship between placental blood flow and combined ventricular output with gestational age in normal human fetus. Cardiovasc Res 1991; 25: 603–8. 15. Meijboom EJ, Horowitz S, Valdes-Cruz LM et al. A Doppler echocardiographic method for calculating volume flow across the tricuspid valve: correlative laboratory and clinical studies. Circulation 1985; 71: 551–6. 16. Kiserud T, Rasmussen S, Skulstad S. Blood flow and the degree of shunting through the ductus venosus in the human fetus. Am J Obstet Gynecol 2000; 182: 147–53. 17. Kohl T, Silverman NH. Evaluation of umbilical venous blood flow by Doppler color flow mapping and conventional ultrasonographic methods. J Ultrasound Med 1996; 15: 465–73. 18. Rasanen J, Wood DC, Debbs RH et al. Reactivity of the human fetal pulmonary circulation to maternal hyperoxygenation increases during the second half of pregnancy: a randomized study. Circulation 1998; 97: 257–62. 19. Lewis AB, Heymann MA, Rudolph AM. Gestational changes in pulmonary vascular responses in fetal lambs in utero. Circ Res 1976; 39: 536–41. 20. Rudolph AM, Heymann MA. Cardiac output in the fetal lamb: the effects of spontaneous and induced changes of heart rate on right and left ventricular output. Am J Obstet Gynecol 1976; 124: 183–92. 21. Romero T, Covell J, Friedman WF. A comparison of pressure–volume relations of the fetal, newborn and adult heart. Am J Physiol 1972; 222: 1285–90. 22. Klopfenstein HS, Rudolph AM. Postnatal changes in the circulation and response to volume loading in the sheep. Circ Res 1978; 42: 839–45. 23. Gilbert RD. Control of fetal cardiac output during changes in blood volume. Am J Physiol 1980; 238: H80–6. 24. Gilbert RD. Effects of afterload and baroreceptors on cardiac function in fetal sheep. J Dev Physiol 1982; 4: 299–309. 25. Thornburg KL, Morton MJ. Filling and arterial pressures as determinants of RV stroke volume in the sheep fetus. Am J Physiol 1983; 244: H656–63. 26. Thornburg KL, Morton MJ. Filling and arterial pressures as determinants of left ventricular stroke volume in fetal lambs. Am J Physiol 1986; 251: H961–8. 27. Hawkins J, Van Hare GF, Schmidt KG, Rudolph AM. Effects of increasing afterload on left ventricular output in fetal lambs. Circ Res 1989; 65: 127–34. 28. Friedman WF. The intrinsic physiologic properties of the developing heart. Prog Cardiovasc Dis 1972; 15: 87–111. 29. Mahony L, Jones LR. Developmental changes in cardiac sarcoplasmic reticulum in sheep. J Biol Chem 1986; 261: 15257–65.

30. Lipp JA, Rudolph AM. Sympathetic nerve development in the rat and guinea-pig heart. Biol Neonate 1972; 21: 76–82. 31. Friedman WF, Pool PE, Jacobowitz D et al. Sympathetic innervation of the developing rabbit heart. Biochemical and histochemical comparisons of fetal, neonatal, and adult myocardium. Circ Res 1968; 23: 25–32. 32. Lebowitz EA, Novick JS, Rudolph AM. Development of myocardial sympathetic innervation in the fetal lamb. Pediatr Res 1972; 6: 887–93. 33. Birk E, Tyndall LC, Rudolph AM, Roberts JM. Effects of thyroid hormone on myocardial adrenergic beta-receptor responsiveness and function during late gestation. Pediatr Res 1992; 31: 468–73. 34. Jansen AH, Chernick V. Respiratory response to cyanide in fetal sheep after peripheral chemodenervation. J Appl Physiol 1974; 36: 1–5. 35. Dawes GS. Foetal and Neonatal Physiology. Chicago, IL: YearBook, 1968. 36. Itskovitz J, Rudolph AM. Cardiorespiratory response to cyanide of chemoreceptors in fetal lambs. Am J Physiol 1987; 252: H916–22. 37. Bartelds B, van Bel F, Teitel DF, Rudolph AM. Carotid, not aortic, chemoreceptors mediate the fetal cardiovascular response to acute hypoxemia in lambs. Pediatr Res 1993; 34: 51–5. 38. Cowley AW Jr, Liard JF, Guyton AC. Role of the baroreflex in daily control of arterial blood pressure and other variables in dogs. Circ Res 1973; 32: 564–76. 39. Itskovitz J, LaGamma EF, Rudolph AM. Baroreflex control of the circulation in chronically instrumented fetal lambs. Circ Res 1983; 52: 589–96. 40. Teitel DF, Iwamoto HS, Rudolph AM. Effects of birthrelated events on central blood flow patterns. Pediatr Res 1987; 22: 557–66. 41. Teitel DF, Iwamoto HS, Rudolph AM. Changes in the pulmonary circulation during birth-related events. Pediatr Res 1990; 27: 372–8. 42. Leffler CW, Hessler JR, Green RS. The onset of breathing at birth stimulates pulmonary prostacyclin synthesis. Pediatr Res 1984; 18: 938–42. 43. Mital S, Konduri GG. Vascular potassium channels mediate oxygen-induced pulmonary vasodilation in fetal lambs. Biol Neonate 2000; 77: 58–68. 44. Moore P, Velvis H, Fineman JR et al. EDRF inhibition attenuates the increase in pulmonary blood flow due to oxygen ventilation in fetal lambs. J Appl Physiol 1992; 73: 2151–7. 45. Sime F, Banchero N, Penaloza D et al. Pulmonary hypertension in children born and living at high altitudes. Am J Cardiol 1963; 11: 143–9. 46. Heymann MA, Rudolph AM. Effects of acetylsalicylic acid on the ductus arteriosus and circulation in fetal lambs in utero. Circ Res 1976; 38: 418–22. 47. Levin DL, Hyman AI, Heymann MA, Rudolph AM. Fetal hypertension and the development of increased pulmonary vascular smooth muscle: a possible mechanism for persistent pulmonary hypertension of the newborn infant. J Pediatr 1978; 92: 265–9. 48. Oberhansli-Weiss I, Heymann MA, Rudolph AM, Melmon KL. The pattern and mechanisms of response of the ductus

Circulation in the normal fetus and cardiovascular adaptations

151

49.

50.

51.

52.

arteriosus and umbilical artery to oxygen. Pediatr Res 1972; 6: 693–700. McMurphy DM, Heymann MA, Rudolph AM, Melmon KL. Developmental changes in constriction of the ductus arteriosus: responses to oxygen and vasoactive substances in the isolated ductus arteriosus of the fetal lamb. Pediatr Res 1972; 6: 231–8. Clyman RI, Mauray F, Heymann MA. Ductus arteriosus: developmental response to oxygen and indomethacin. Prostaglandins 1978; 15: 993–8. Keller RL, Tacy TA, Fields S et al. Combined treatment with a nonselective nitric oxide synthase inhibitor (l-NMMA) and indomethacin increases ductus constriction in extremely premature newborns. Pediatr Res 2005; 58: 1216–21. Kajino H, Goldbarg S, Roman C et al. Vasa vasorum hypoperfusion is responsible for medial hypoxia and anatomic remodeling in the newborn lamb ductus arteriosus. Pediatr Res 2002; 51: 228–35.

53. van Bel F, Roman C, Iwamoto HS, Rudolph AM. Sympathoadrenal, metabolic, and regional blood flow responses to cold in fetal sheep. Pediatr Res 1993; 34: 47–50. 54. Rudolph AM, Roman C, Gournay VA. Perinatal myocardial DNA and protein changes in the lamb: effect of cortisol in the fetus. Pediatr Res 1999; 46: 141–6. 55. Breall JA, Rudolph AM, Heymann MA. Role of thyroid hormone in postnatal circulatory and metabolic adjustments. J Clin Invest 1984; 73: 1418–24. 56. Birk E, Tyndall MR, Erickson LC et al. Effects of thyroid hormone on myocardial adrenergic β-receptor responsiveness and function during late gestation. Pediatr Res 1992; 31: 468–73. 57. Smolich JJ, Walker AM, Campbell GR, Adamson TM. Left and right ventricular morphometry in fetal, neonatal, and adult sheep. Am J Physiol 1989; 257: H1–9. 58. Townsend SF, Rudolph CD, Rudolph AM. Changes in ovine hepatic circulation and oxygen consumption at birth. Pediatr Res 1989; 25: 300–4.

.

which is responsible for the antegrade delivery of blood.11 Development of fetal cardiac and extracardiac Doppler flows in early gestation Ahmet A Baschat and Ulrich Gembruch Introduction The onset of rhythmic contractions of the primitive embryonic heart between 21 and 24 days after conception initiates an important sequence in the functional development of the embryonic cardiovascular system. The normal development of the embryonic cardiovascular system and the fetoplacental unit are necessary to ensure adequate blood flow. when the atrioventricular (AV) valves are open and the ventricles receive blood. The principal events that constitute the cardiac cycle are ventricular diastole.3 Previous observations have indicated that certain cardiac defects may produce a unique hemodynamic or structural impact in early gestation which ultimately contributes to their poor prognosis. the normal maturation of these two circulatory systems is important for adequate fetal growth and development. and the level of cardiac function determines the efficiency with which adequate blood flow can be provided to the body under physiological and pathological conditions. and ventricular systole. and gas and nutrient exchange at an organ and cellular level.5 Examination of DV blood flow is often abnormal under such circumstances.2 This example illustrates that the understanding of embryonic and fetal cardiovascular dynamics is likely to gain increasing importance in the future.4. discrete phases of the cardiac cycle have been identified which further subdivide the two principal events. when the aortic and pulmonary valves are open and blood is ejected into the circulation. Thereafter. have been investigated. Cardiovascular control mechanisms The heart undergoes a repetitive orderly sequence of cardiac contraction and valvular action. Since its introduction by FitzGerald and Drumm1 Doppler sonography has evolved as an important tool for the non-invasive examination of the fetal cardiovascular system in uncomplicated pregnancies and fetal disease. Greater understanding of normal early vascular development allows Doppler flow studies to be integrated into prenatal diagnosis.3 Preliminary results suggest that the application of this information and incorporation of DV Doppler screening into the nuchal translucency scan may result in a major reduction in the need for invasive testing. The cardiovascular system has to match these needs of the growing and developing embryo. These events are summarized in the cardiac cycle. A detailed knowledge of the physiology underlying the cardiac cycle is helpful in the application and interpretation of Doppler waveforms obtained from the fetal circulation. and application in pregnancies at high risk for chromosomal defects. As the placenta is the major respiratory organ in utero. Studies investigating the normal and abnormal development of the human fetal cardiovascular system indicate that there are important differences from other mammalian species. The first trimester is a period of rapid development in many organ systems coupled with exponential embryonic growth. This chapter outlines functional cardiovascular changes in the fetus with particular emphasis on the first and second trimesters. The study of these cardiovascular developmental changes has been of importance for several reasons. In addition.2. This explains why abnormal DV blood flow in chromosomally normal fetuses with increased nuchal translucency may be beneficial in identifying fetuses with underlying major cardiac defects. oxygen delivery. It is therefore not surprising that there are important changes in fetal cardiac function which take place in the first and second trimesters. In this context the integration of ductus venosus (DV) Doppler studies into first-trimester anomaly and aneuploidy screening. fetal growth and development continue towards term in a more steady fashion. data that have been gathered in sheep and primate experiments may have to be applied with caution to the human fetus. Evaluation of fetal cardiac . Therefore. The wide application of Doppler sonography has greatly enhanced our knowledge of the maturation of the embryonic cardiovascular system.

Initial passive filling is rapid and reaches a plateau (diastasis) which is followed by atrial contraction triggered by the electrical discharge at the sinoatrial node. Electrical activity precedes atrial and ventricular contractions. Cardiac cycle – diastole Ventricular diastolic filling in the adult is subdivided into a passive phase and an active phase. However.154 Fetal Cardiology function requires knowledge about the characteristics of these phases and their relationship to flow and pressure events. The rapid rise in atrial pressure is transmitted into the venous system. . B B C SV SV A A D Ventricular volume Ventricular pressure Precordial venous flow velocity waveform S-wave D-wave a-wave Electrocardiogram Diastole IC Systole IR Diastole IC Systole IR Figure 11. important differences between fetal and adult cardiac function have been identified using the Doppler technique.6 Atrial contribution to ventricular filling therefore varies inversely with duration of ventricular diastole and directly with atrial contractility. Since Doppler waveforms display only time and velocity information. isovolumetric contraction) until it exceeds pressure in the great vessels. if the heart rate is very high (e. and semilunar valves open (B).1). When ventricular pressure falls below diastolic pressures in the major vessels. The two phases of ventricular inflow produce a biphasic flow velocity waveform across the AV valve consisting of an early peak (E-wave) and a second shorter peak during atrial contraction (A-wave). This causes the valves to float upward (pre-position) before complete closure during the beginning of systole. Nonetheless. In the fetus a larger proportion of ventricular filling occurs by atrial contraction in late diastole. any deductions about cardiovascular pressures are made from our understanding of the phases underlying the cardiac cycle. allowing increased precordial venous forward flow (S-wave). causing a pressure gradient reversal across the AV valves. Ventricular shortening during ejection of the stroke volume (SV) causes rapid descent of the atrioventricular valve ring. The initial rapid inflow of blood into the ventricles causes a proportionate fall in atrial pressure. In the adult heart the majority of ventricular filling (90%) occurs during the E-wave while the A-wave contributes to the remainder when a person is at rest. initiating active filling. When intraventricular pressure falls below atrial pressure. With onset of ventricular contraction atrioventricular valves close (A) and intraventricular pressure rapidly rises without ventricular shortening (IC. which is reflected in a fall in the venous pulse (Y-descent).and A-waves may reverse. the semilunar valves close (C). In such situations the ‘atrial kick’ contributes to a larger proportion of ventricular filling and therefore the relationship of the E. Subsequent contraction of the atria results in a rapid rise in the atrial pressure. the atrial pressure begins to fall. At this time. the ventricular volumes are maximal and this ventricular end-diastolic volume constitutes the ventricular preload (Figure 11. decreasing antegrade flow (a-wave).1 Diagrammatic representation of the cardiac cycle. during exercise). which is reflected in a rise in the venous pulse (a-wave). Shown is the relationship between intraventricular pressure and volume and precordial venous flow velocity waveforms during the cardiac cycle. After atrial contraction is complete.g. the atrioventricular valves open (D) at the end of isovolumetric relaxation (IR). the atrial contraction may account for up to 40% of ventricular filling. The rapid inflow into the ventricle is reflected by increased precordial venous flow (D-wave).

This causes ventricular active tension to decrease and the rate of ejection and ventricular emptying to fall. The rapid ventricular shortening results in a descent of the AV-ring which causes atrial pressures to fall below venous pressures (X-wave). so other indexes such as ventricular end-diastolic volume or pressure are substituted. At this point muscle fibers have shortened. The blood pressure generated by the combined effects of ventricular contraction force. a decrease in compliance. the intraventricular pressures fall below the diastolic pressures in the major vessels. and can no longer contract forcefully. preload is primarily determined by venous return and blood volume. changes of downstream resistance (afterload). and particularly end-diastolic volume. end-diastolic pressure. Changes in the filling state of the heart (preload). When the ventricular ejection falls to zero. Neither of these measures is ideal. which is the cardiac pacemaker with the highest rate of intrinsic automaticity. The difference between the end-diastolic volume and the end-systolic volume is the stroke volume (Figure 11. Degree and velocity of myocardial relaxation are the major determinants of ventricular compliance and therefore resistance of the ventricles to diastolic filling. The heart rate is predetermined by the sinoatrial node. vessel wall resistance and downstream vascular resistance determines fetoplacental perfusion. are used as clinical indices of preload. The overall control of vascular tone and blood pressure is integrated at the level of the vasomotor center. For any organ the perfusion is passively regulated by the pressure drop across the arterial and venous ends of its vascular supply. Following rapid ejection the rate of outflow from the ventricle decreases and the ventricular and aortic pressures start to decrease. In the adult heart an increase in preload or a decrease in afterload within a physiological range will result in an increase of stroke volume. atrial filling begins at this time. If preload is viewed as enddiastolic volume. and is related to the sarcomere length. by affecting filling time. Nevertheless. Ongoing ventricular relaxation ensures decreasing intraventricular pressures until these fall below atrial pressure and the atrioventricular valves open at the end of isovolumetric relaxation. This period corresponds to the S-wave in the venous flow velocity waveform (see below). the pulmonary and aortic valves open. The combined cardiac output is simply the sum of right ventricular and left ventricular outputs. then preload is determined ultimately by the end-diastolic pressure and the compliance of the ventricle. The product of the cardiac output and peripheral resistance determines the blood pressure generated. The volume of blood that remains in the ventricle prior to the opening of the AV valves is called the end-systolic volume. it is important to realize the presence of many interacting factors. Once the intraventricular pressures exceed those of the great arteries. Heart rate. Changes in cardiac output may be due to variations in heart rate and/or stroke volume. Following closure of the atrioventricular valves initially there is a period of rapid rise in intraventricular pressures (isovolumetric contraction). There is superimposed modulation of heart rate and intracardiac conduction through the AV node by the autonomic nervous system. left ventricular blood is ejected into the ascending aorta and right ventricular blood is ejected into the main pulmonary artery during this rapid ejection phase of ventricular systole. A low ventricular compliance is associated with an exaggerated increase in . and myocardial muscle mass. Other factors influencing cardiac function are valvular competence. The parallel arrangement of the fetal circulation results in the unique feature that the relative contribution of the individual chambers to the combined cardiac output can change under physiological and pathological conditions. can have a pronounced inverse effect on preload.8 Under physiological circumstances.7 The downstream resistance experienced by each ventricle is determined by the sum of blood flow resistance in individual downstream vascular beds. Sarcomere length cannot be determined in the intact heart.1). resulting in closure of the aortic and pulmonary valves. In addition. Indices of cardiac function Individual ventricular blood volumes ejected in 1 minute define the ventricular outputs representing the product of individual ventricular stroke volumes and heartbeats per minute. Therefore. Although the mechanisms involved in the cardiac cycle appear relatively straightforward at first. Elevation of intraventricular pressures above the atrial pressures results in closure of the tricuspid and mitral valves. will lead to a reduction in preload unless there is a corresponding increase in end-diastolic pressure. resulting in a temporary increase in atrial pressures transmitted to the venous pressure waveform. During this phase of isovolumetric contraction there may be bulging of the AV valves. The efficiency of the Frank–Starling mechanism is strongly influenced by diastolic and systolic myocardial properties. The Frank–Starling mechanism describes the ability of the myocardium to increase stroke volume in response to increases in preload. are repolarizing. because they do not accurately reflect sarcomere length. as occurs with ventricular hypertrophy. Preload can be defined as the initial stretching of the cardiac myocytes prior to contraction.Development of fetal cardiac and extracardiac Doppler flows 155 Cardiac cycle – systole When the electrical impulses traverse the annulus fibrosus and reach the ventricular conduction system. blood viscosity and inertia of the blood. ventricular myocardial contraction is initiated. and variations in myocardial contractility can achieve alterations of the stroke volume. many organs have the potential to optimize perfusion by local changes in resistance vessel diameter by a process called autoregulation.

when only 33% of blood flow reaches the placenta and 67% is distributed to the lower body9. two-thirds reaches the placenta. The functionality of the myocardium is one of the factors responsible for the efficiency with which cardiovascular control mechanisms can modulate cardiac output. The high resistance in the fetal peripheral circulation and the constantly falling blood flow resistance in the placental bed ensure that a considerable proportion of aortic blood flow is diverted to the placenta via the umbilical arteries for oxygenation. and the descending aorta with its organ branches. Vascular resistance in the ascending aorta and the brachiocephalic circulation predominantly determines left ventricular afterload. which grow with the embryo. in contrast to the other veins. The relative separation of the venous ventricular inflows also has the effect that right ventricular preload comprises mainly the superior and inferior venae cavae. Contractility and contraction velocity are the main determinants of stroke volume in circumstances of increased afterload. while 22% supplies the upper part of the body. displacing the effective period of filling to late diastole.11 Blood with lower oxygen content enters the right atrium via the inferior and superior caval veins and the coronary sinus.9. At mid-gestation 59% of the blood flow reaches the placenta while 41% reaches the lower half of the body.9 The DV develops at approximately 7 weeks of gestation and. and 3% the myocardium. However. the ductus arteriosus. Right ventricular afterload is predominantly determined by vascular resistance in the main pulmonary artery. Slower systolic contraction velocity and strength and low diastolic compliance have a significant impact on cardiac function and the efficiency of cardiovascular reflex mechanisms.13 One of these bloodstreams originates in the umbilical sinus. myocardial contractility has to be of sufficient force to oppose the effects of afterload. The characteristics of the fetal circulation have been extensively studied in the fetal lamb.14–16 (Figure 11. The right ventricle therefore has the role to deliver oxygen-poor blood to the placenta for oxygenation while the left ventricle delivers well-oxygenated blood to the brain and heart. The relative separation of right atrial inflow and little admixture from the pulmonary veins ensures that in the fetal lamb the left ventricle receives approximately 65% of the well-oxygenated blood. Differential directionality of right atrial venous inflow and the crista dividens of the foramen ovale results in an interatrial right-to-left passage of well-oxygenated blood that was also confirmed for the human fetus. Left ventricular output is predominantly distributed to the coronary and brachiocephalic circulations. A second stream originating in the abdominal portion of the inferior vena cava is joined by right hepatic venous blood entering the right atrium via the right upper portion of the inferior vena cava.14. This stream meets with blood from the superior vena cava and coronary sinus and enters the right ventricle through the tricuspid valve. The differential downstream distribution of blood volume passing through the descending aorta changes with gestation. Of the 29% of left ventricular output reaching the descending aorta.10 The narrow diameter of the ductus is responsible for a marked acceleration of the entering blood. while left ventricular preload comprises the pulmonary veins and the left hepatic vein and DV. which has a 15–20% higher oxygen content than right ventricular blood. 8% the lungs. Under such circumstances ventricular filling becomes increasingly dependent on atrial contraction. At this time the brachiocephalic circulation receives approximately 35% of the common cardiac output while the placenta and the remaining body receive 30% each.15 A change in distribution of the cardiac output with advancing gestation has been documented for the fetal lamb. approximately 50% of oxygenated blood of placental origin reaches the fetal heart via the umbilical vein through the DV after bypassing the hepatic circulation.156 Fetal Cardiology intraventricular pressure when preload is increased. According to data from sheep studies. middle and left hepatic venous blood joins this stream. shows little increase in size.14.8:1. The proportion of cardiac output reaching the brachiocephalic circulation increases from approximately 20% at midgestation to term. reaching the left atrium via the left upper portion of the inferior vena cava after considerable flow acceleration in the DV. Because of the parallel arrangement of the fetal circulation the afterload acts separately on each ventricle. In the fetal sheep the ratio of right-to-left ventricular output is 1.9.8. The fetal circulation The unique arrangement of the fetal circulation impacts significantly on cardiac function and distribution of cardiac output.2). In addition.15 Approximately two-thirds of this blood reaches the placental vascular bed for oxygenation via the umbilical arteries. Because of the parallel arrangement there is a differential distribution of right and left ventricular outputs with a predominant contribution of the right ventricle. only 13% of the right ventricular output is distributed to the lungs and the remaining 87% reaches the descending aorta. This proportion changes toward term. Pulmonary venous return reaches the left ventricle via the left atrium.12. and the combined resistance of the fetoplacental circulations. In the fetal lamb the majority (41%) of the combined cardiac output is delivered to the placenta. The non-invasive Doppler examination of the human fetal circulation seems to confirm the developmental changes observed in other mammalian species. Owing to the high pulmonary vascular resistance and the orientation of the ductus arteriosus. there are some important differences that have been . The decreased ventricular filling capacity is prohibitive for an increase in the stroke volume.10.

SVC. Shown is the distribution of right and left ventricular output (RVO. DV. at 1. suggesting a higher priority of the fetal liver than was previously realized.24. It was long presumed that the human fetus was unable to increase cardiac output by the Frank–Starling mechanism. IVC. Descending aorta (63% CCO) 10% LVO (4% CCO) 29% RVO (18% CCO) Lower body 22% CCO 19% LVO (6% CCO) 58% RVO (35% CCO) Placenta 41% CCO identified in the human fetus. the ability to increase preload by alteration of venous return is limited. ascending aorta. the demonstration of post-extrasystolic potentiation in cases of fetal arrhythmia suggests that the Frank– Starling mechanism is operational in the human fetus from 20 weeks onwards. The largest proportion of the CCO is distributed to the placenta for oxygenation.24.17–19 From approximately 20 weeks’ gestation onward. The majority of the fetal blood volume is contained in the placental circulation situated between the arterial and venous limbs of the circulation. Since this large proportion of the vascular volume is extracorporeal. provided arterial pressure does not increase. LVO) and the combined cardiac output (CCO) in the circulation of the fetal lamb.20. The difference from the sheep fetus is due to the fact that the human fetus has a relatively larger brain mass. as well as slower contraction and relaxation rates than the adult or neonatal myocardium.2:1.22. Structural and functional maturation of the fetal myocardium The immature fetal myocardium has decreased contractility and compliance. very sensitive to changes in afterload. but working at the upper limit. fetal cardiac function is described as being relatively afterload-sensitive. However. the ratio between the right and left ventricular output remains relatively constant. superior vena cava. For these reasons. inferior vena cava. the presence of central shunting through the foramen ovale . Compared to the 50% shunting of umbilical blood through the DV found in animal experiments. the degree of shunting in the human fetus under physiological conditions is considerably less.17 Another important difference from the fetal lamb is that in the human fetus a higher proportion of umbilical blood is directed to the liver and less is shunted through the DV. ductus venosus. Through preferential streaming.2 Diagrammatic representation of the fetal circulation. AAO.27 Finally.25 Cardiac output can be increased with volume loading and β-adrenoceptor stimulation. Further studies are necessary to delineate these differences in greater detail to elucidate fetal adaptation to intrauterine life. cardiac output experiences only a slight increase if afterload falls.15. The dominance of the right ventricle contributing approximately 53% of the combined cardiac output has also been demonstrated for the human fetus. well-oxygenated left ventricular blood supplies the brain and heart while right ventricular blood with lower oxygen content is predominantly distributed to the placenta.Development of fetal cardiac and extracardiac Doppler flows 157 Left ventricle ~35% of venous return: DV / pulmonary veins Right ventricle ~65% of venous return IVC & SVC / coronary sinus Coronary artery 8% LVO Heart 3% CCO AAO 63% LVO Brain 26% CCO Pulmonary artery 13% RVO Aortic isthmus 29% LVO (10% CCO) Ductus arteriosus 87% RVO (53% CCO) Lungs 8% CCO Figure 11.23 The fetal myocardial response to increases in afterload is another limitation when compared to neonatal and adult myocardium.26 This indicates that alteration of preload is an important determinant of cardiac output in utero. Investigations in the fetal lamb have demonstrated that resting fetal cardiac function is normally near maximum.25 There is a sharp drop of cardiac output with minimal increases in afterload.21 These findings indicate that there are important differences in regional blood flows in different mammalian species that may have important consequences in pathological states. The Frank–Starling mechanism is necessary for optimal cardiac output at various filling stages of the heart. and largely unaffected by baroreceptors.25. In addition.

The ventricular conduction system develops concomitantly with ventricular septation and first becomes detectable in the myocardium surrounding the interventricular foramen.36 A number of ultrastructural changes have a significant impact on fetal myocardial function with advancing gestational age.1. myosin. Development of the fetal cardiac conduction system and autonomic innervation Fetal cardiac activity is initiated by contracting ventricular myocytes on the caudal region of the straight heart tube between days 21 and 23 after conception. Development of the cardiac conduction system is ongoing during the folding and looping process of the embryonic heart and its maturation continues until after delivery. The density of β-adrenoceptors. Both ventricles are structurally identical at this time. The ratios between right and left ventricles and pulmonary trunk and aorta remain relatively constant at 1:1 and 1:1.44 In addition to the changes in cardiac dimensions.32–34 Fetal troponin isoforms display a lower affinity and sensitivity for calcium ion binding than their adult counterparts. number of sarcomeres. After β-adrenergic stimulation. therefore having a more profound effect on cardiac functional development.41–43 This increase may promote the development of cardiac adrenoceptors as well as switching to adult forms of myosin heavy chains. the sequence and timing of these changes has not been fully delineated. These calcium ions bind to troponin at the actin-to-myosin binding site.15 Many mammalian experimental studies have documented structural and functional maturation of the myocardium in fetal and neonatal life. which correlates with fetal growth and gestational age.39. An increase in myocardial adrenergic receptor density and maturation of the sarcoplasmic reticulum ultimately allow more effective stimulation of the sarcomere. These changes presumably also occur in the human fetus. and tropomyosin constitute the contractile unit of the cardiac muscle. preload of both ventricles is equally affected. The cardiac weight increases 200-fold between 8 and 20 weeks of gestation. Cardiac contractions are initiated at the time of development when truncoventricular fusion commences.158 Fetal Cardiology also results in equalization of pressures between the two ventricles (interdependence). The sarcomere and its protein components actin. the AV node provides the only conductive pathway between these chambers. which contributes to an increase in contractility and maximal generated contractile force. The sinoatrial and AV nodes persist as remnants of the primary myocardium of the sinus venosus and AV junction. development of the sarcoplasmic reticulum and the T-tubule system. the contraction velocity is determined by the reaction rate of the myosin ATPase.40 However.40 The marked rise in thyroxine concentrations prior to term may contribute to improved cardiac function by an upregulation of myocardial β-adrenergic receptors. This is followed by a rise to approximately 170 beats/minute at . Owing to this effect. Activation of troponin by the binding of calcium ions is permissive for this electromechanical coupling. Fetal myocardium has approximately 30% less contractile elements per gram of muscle fiber28. troponin.30. Conversion of adenosine triphosphate (ATP) to adenosine diphosphate (ADP) by a myosin-bound enzyme (ATPase) releases energy that is necessary for the relative motion between actin and myosin.39. Therefore.15 Between 13 and 17 weeks of gestation there is a significant linear increase of all cardiac dimensions. involving the atria and the sinus venosus.30. and a steady further increase toward term. and the number of calcium binding sites on the troponin influence the strength of contractility of the muscle fiber. the primitive heart pumps blood before rotation and septation.30–37 There is a switch in the expression of isoforms of the contractile proteins to more mature adult forms. a significant change in embryonic heart rates with advancing gestation has been documented in response to peripheral and central maturation of autonomic cardiac control.38 The combination of enhanced calcium delivery in response to stimulation and changes in contractile elements results in a markedly increased efficiency of electromechanical coupling. As the atria and ventricles become electrically isolated by growth of the endocardial cushions. structure of contractile proteins. There is an up to 40-fold increase of thyroid hormones in the embryonic tissues from the 9th to the 12th week of gestation.28.31–35 The fetal sarcoplasmic reticulum and the transverse tubular system are less well developed and have a lower capacity for calcium ions than in the adult or neonate.15 An important modulator of this maturational process is the concentration of thyroxine. which results in shortening of the sarcomere and ultimately contraction of the muscle fiber.29 Contraction velocity and maximal contractile force of isolated fetal cardiac muscle fibers are significantly lower than in the neonate or adult. In contrast.29 as well as a lower concentration of contractile proteins. respectively. electrical impulses travel across the sarcoplasmic reticulum resulting in calcium ion release from the T-tubule system.35.31 The myosin ATPase bound to the heavy chain myosin isoforms has a slower reaction rate than that bound to adult isoforms. Increased capacity and distribution of calcium ions within the sarcoplasmic reticulum enhance delivery of this important contractile substrate to troponin binding sites. impairing the ability for selective regulation of ventricular output by this mechanism. In normal pregnancies the mean heart rate is in the region of 110 beats/minute at 5 weeks of gestation.

Between 10 and 20 weeks of gestation.14.44–47 The initial changes in heart rate are primarily attributed to the changes in cardiac morphology.52 The sequential changes of embryonic heart rate are relatively constant. with the velocities on the ordinate and the time on the abscissa (Figure 11.55–58 Doppler investigation of the cardiovascular system The Doppler equation can be used to describe the relationship between the Doppler shift frequency and the absolute blood flow velocity.5 Marked episodes of sinus bradycardia with excellent prognosis may be observed until this period in gestation.48. The TAV can be used to calculate volume flow if the vessel diameter is known. development of the conduction system is generally completed at 12 weeks. The systolic/diastolic ratio. followed by sympathetic nerves between 22 and 24 weeks. The indexes used in fetal medicine utilize the systolic. and pulsatility index predominantly reflect downstream flow resistance and filling pressure of the arterial bed.59. With formation of the AV node and the junctional connection to the ventricles. increasing ventricular muscle mass51 and improved atrioventricular valvular function. the velocity can be calculated from the Doppler shift frequency if the velocity and angle of insonation are known. This allows the establishment of the sinoatrial node as the primary cardiac pacemaker with the highest intrinsic spontaneous rate of rhythmicity. Evidence of maturation of the autonomic innervation of the fetal heart has been gathered using Doppler ultrasound. since this allows ongoing waveform analysis in cases of absent or reversed end-diastolic blood . In the human fetus the parasympathetic inhibition of the sinoatrial node appears to be established between 12 and 17 weeks.57.49 The subsequent decrease in heart rate is thought to reflect increasing parasympathetic modulation. Quantitative analysis is predominantly utilized for intracardiac flow velocity waveforms.50 improved ventricular contractility. After resolution of the equation.54 It therefore appears that the human embryonic heart goes through an ultrastructural maturation sequence. Using these indexes avoids the sources of error inherent in the use of quantitative waveform analysis and the calculation of absolute volume flow.60 The flow velocity waveform can be analyzed by quantitative or qualitative waveform analysis. measurement of absolute volume flow by Doppler requires stringent adherence to correct anatomic planes using an insonation angle close to 0°. time averaged maximum velocity (TAMX). which is coupled with the growth of the cardiac structures and the development of receptor and neural connections. The association between abnormal structural cardiac development and the presence of transient thickening of the nuchal fold suggests that intravascular volume handling in the human embryo and early fetus may be particularly vulnerable to cardiac dysfunction.62. end-diastolic. The integral under the waveform corresponds to the maximal velocity in the time interval examined. resistance index.60. whereas peak systolic velocity variability reflects the activation of a hemodynamic feedback mechanism. It has been realized that bradycardia is associated with a higher risk for chromosomal anomalies and/or spontaneous miscarriage. The increasing heart rate in the early first trimester is predominantly due to the fusion of the sinus venous with the atria and ventricles and the development of the conductive tissue. resulting in persistence of abnormal conduction pathways. and time to peak velocity. Since inaccuracies in vessel diameter are raised to the square. Quantitative analysis utilizes formulas to calculate absolute velocities and volume flow.Development of fetal cardiac and extracardiac Doppler flows 159 9–10 weeks. as shown in Figure 11. These allow establishment of aberrant reentry conduction in some cases. In such an image the flow velocity waveform is displayed.63 Many authors prefer the use of the pulsatility index.53 More recently it was demonstrated that an increase in the embryonic heart rate was associated with a higher risk for aneuploidies. increasing variability of fetal heart rate and time-averaged velocity has been documented in the abdominal aorta. The primary parameters utilized include the absolute diastolic and systolic velocities. The complete formation of the annulus fibrosus may be disturbed or delayed. and the flow across the foramen ovale.59–61 The intracardiac flows that have been predominantly investigated include the diastolic filling across the atrioventricular valves. umbilical artery peak systolic velocity variability and fetal heart rate variability also increase in this period. Extracardiac Doppler ultrasound predominantly uses qualitative waveform analysis based on angle-independent indexes. The processing of incoming signals via fast Fourier transformation allows the graphic display of the calculated velocities in a two-dimensional image.58 The documented inverse relationship between umbilical artery flow velocity and fetal heart rate during this time has been considered as evidence that the Frank–Starling mechanism regulates cardiovascular control as early as the late first and early second trimesters of pregnancy. with a subsequent fall to 150 beats/minute at 14–15 weeks and a continuing gradual further drop to term. Completion of the annulus fibrosus eventually leads to cessation of aberrant conduction. and TAMX velocities.55 In addition.15 The successive decline of baseline rate and increasing heart rate variability toward term reflect a relative increase in parasympathetic cardiac innervation and central maturation of autonomic control. time averaged velocity (TAV).4. the systolic output across the aortic and pulmonary valves.56 The heart rate variability is considered as evidence of maturation of the parasympathetic nervous system.3).

In the adult.3 Arterial flow velocity waveform. The degree of reverse flow during atrial contraction varies considerably in individual veins. Such situations occur with increased atrial and ultimately increased central venous pressure. These studies have allowed greater insight into the effects of many conditions on cardiac preload. Since the beginning of the 1990s examination of the venous system has been increasingly incorporated into the assessment of the human fetus. systole. this context are the inferior vena cava. hepatic veins. atrial contraction) reflects volume flow changes during the cardiac cycle (see text).160 Fetal Cardiology (a) Systolic peak velocity Shaded = time averaged maximum velocity (TAMX) End-diastolic peak velocity Enddiastole Time Acceleration time / time to peak velocity (b) Systolic peak velocity Shaded = time averaged maximum velocity (TAMX) S D Diastolic peak velocity Peak velocity during atrial systole a Time Figure 11. while those of the left hepatic vein and the DV more closely represent left ventricular function. Except for the preload index.68 and the preload index.58 An increase of downstream vascular resistance results in a relative decrease of end-diastolic velocities and a subsequent increase of these three indexes (Figure 11. The ratio between the E. while the D-wave results from passive ventricular filling during early diastole.3b).and A-wave peak velocities is a generally accepted angle-independent index for quantification of the waveform across the AV valves. (a) Diagrammatic representation of an arterial flow velocity waveform. and pre. relaxation velocity and compliance.5). The characteristic waveform has two peaks representing early diastolic ventricular filling (E-wave) and ventricular filling during atrial systole (A-wave) (Figure 11. pulsatility index for veins. DV. The sudden increase in right atrial pressure with atrial contraction in late diastole causes a variable amount of reverse flow. respectively) are each followed by a trough. The umbilical venous flow velocity waveform has a constant pattern in the second and third trimesters. The marked velocities are utilized for the calculation of indexes. producing a second trough after the D-wave (the A-wave) (Figure 11. diastole. The index is thought to reflect changes in ventricular diastolic function. The pulsations correspond to the maximal reverse flow component of atrial contraction.64–66 The venous flow velocity waveform consists of two peaks and troughs.69 Venous indexes are believed primarily to reflect cardiac function and to a lesser extent cardiac afterload64–72 (Figure 11. (b) Diagrammatic representation of a venous flow velocity waveform. while the acceleration time is used as an absolute measurement.73 Development of intracardiac flow velocity waveforms Diastolic ventricular filling is examined by positioning the Doppler gate immediately distal to the respective AV valves. The vessels most commonly utilized in .4). normal blood flow in the DV is antegrade also during atrial systole. The trough that follows the S-wave is produced by ascent of the AV-ring as the ventricle relaxes.4). The S-wave is produced by descent of the AV-ring during ventricular systole. and the free umbilical vein. D.and afterload of the respective ventricle. The flow velocity waveform of the inferior vena cava is influenced more by right ventricular function. all indexes mentioned increase when there is a decrease in antegrade flow. such as that associated with increased reverse flow during atrial systole. There may be reverse flow during atrial contraction in the inferior vena cava and hepatic veins. The triphasic waveform (S. The systolic and diastolic peaks (the S-wave and D-wave. a. Pulsations in the umbilical vein are considered evidence of retrograde transmission of elevated central venous pressures.67 peak velocity index for veins. The most widely applied indexes for the assessment of venous Doppler-derived flow are the percentage of reverse flow. flow velocities. In contrast.

17 With advancing gestation there is in an increase of both transtricuspid and transmitral peak E. diastolic filling occurs predominantly in the early passive phase of diastole resulting in an E/A ratio of > 1. Therefore. Velocity . a lower E/A ratio is thought to reflect these properties of the ventricle.76. resulting in a substantial increase in the E/A ratio from approximately 0.74–76 An increase in cardiac size and atrioventricular valve area contributes to an increase in atrioventricular blood flow volume and thus ventricular preload.4 Formulas used for quantitative and qualitative waveform analysis. requiring atrial contraction to contribute to the major part of ventricular filling. Low ventricular compliance is associated with impaired filling during the passive early phase of diastole.77–82 This is thought to reflect the combined effects of decreasing afterload and improved diastolic ventricular function and both myocardial relaxation and compliance. After birth the relationship between velocities during early diastole (E) and A-wave reverses (see text). the A-wave is generally predominant in utero.83 This suggests the absence of a significant change in ventricular filling pressure and a constant compliance.77–82 During the second and third trimesters of pregnancy a lack of correlation between the Doppler echocardiographically assessed ventricular filling pressures. favoring passive filling during early diastole. In the fetus a greater proportion of diastolic filling occurs during atrial systole. diastolic filling relies mainly on atrial contraction. and gestational age was shown. Doppler measurements of the transatrioventricular flow velocity waveforms showed that monophasic AV flow velocity waveforms become biphasic as early as 8 weeks. Subsequently. Shown is a diagrammatic representation of the flow velocity waveform across atrioventricular valves. reproducible results have been obtained. which showed a linear relationship with the crown–rump length. If the atrioventricular flow profiles are used for calculation of the ventricular output.5 at 13 weeks of gestation to 0. resulting in a higher peak velocity during this part of the cardiac cycle (A). which were determined by measuring the ratio between flow velocity (E) and annular velocity (EA) in early diastole.75 A significant gestational age-dependent increase was observed for all AV waveform parameters.74.77 This is in keeping with the fact that the right ventricle contributes to a greater proportion of the combined cardiac output in utero.Development of fetal cardiac and extracardiac Doppler flows 161 Cardiac indexes E/A ratio = E A Venous indexes TVI (S+D) TVI a S a S–a S S–a D S–a TAMX % Reverse flow Volume flow (Q) Acceleration time Preload index Arterial indexes Pulsatility index = S–D TAV S–D S S D = TAV x πr2 S/A ratio = = = Peak velocity index = Resistance index = Pulsatility index = Figure 11. In both the first and the second trimesters the transtricuspid blood flow volume exceeds that across the mitral valve.9 at 36–38 weeks’ gestation. respectively. showing right ventricular dominance. Systolic diastolic ratio = A-wave dominance A E E E-wave dominance A E/A waves in utero E/A waves after birth Time Figure 11. during fetal life. This A-wave dominance in utero is presumed to be due to the lower compliance and relaxation time of the fetal myocardium when compared to the adult. In contrast. but little change in peak A.8–0.5 E/A diagram in the adult and fetus. resulting in an E/A ratio of > 1. which are likely to result in a decrease of ventricular diastolic pressures.

83 Flow velocity waveforms in the outflow tracts Peak velocities in the ascending aorta and the main pulmonary artery are predominantly determined by cardiac contractility and afterload and.89 Chaoui et al were able to document that the TPV and the ratio between TPV and the ejection time increased significantly between 18 weeks and term.92 . are accepted parameters of ventricular systolic performance. This assumption. has also been confirmed in the human fetus.87 van Splunder et al74 have confirmed these findings for the first-trimester fetus.7 ml at 20 weeks to 7.15 at 38 weeks’ gestation. and there is still controversy about the utility and significance of TPV investigation in the human fetus. the peak blood flow velocity in the ascending aorta averages 60 cm/s and in the main pulmonary artery 54 cm/s.25 at 18 weeks’ gestation to 1.61 The time to peak velocity (TPV) or acceleration time of the arterial flow velocity waveform is primarily determined by mean arterial pressure and to a lesser degree by ventricular contractility. the relationship between cardiac output and fetal body weight remains relatively constant at 450 ml/kg/per minute throughout pregnancy.85 In concordance with these data. therefore. while Sutton et al documented a consistently higher TPV in the aorta.90 These changes throughout gestation are in accordance with animal experiments and suggest that the mean arterial pressure in the main pulmonary artery is higher than in the ascending aorta. Rizzo et al88 found no major differences between these two vessels.90 The majority of authors agree that the difference in TPV between the aorta and the pulmonary artery becomes less marked and may be negligible close to term.68 With increasing gestational age there is a relative shift in cardiac output towards the left ventricle. A positive relationship between stroke volume and gestational age. From approximately 26 weeks of gestation.77. which was previously documented in an animal experiment.6 ml at 40 weeks for the right and from 0. These findings confirm animal experimental data suggesting the intrauterine dominance of the right ventricle.66. There is an inverse relationship between the TPV and the mean arterial pressure.2 ml at 40 weeks for the left ventricle. With continuing gradual increase toward term these velocities increase to 60–120 cm/s in the ascending aorta and 50–110 cm/s in the main pulmonary artery.85 Based on these measurements it has been calculated that the right ventricular stroke volume exceeds the left ventricular stroke volume by approximately 28% through gestation.84. Measurement of outflow tract peak blood flow velocities has been performed as early as 10 weeks of gestation. the changes in ejection times and vascular resistances indicate an increase in perfusion as well as a decrease in the vascular resistance and pressure in the pulmonary circulation.7 ml at 20 weeks to 5.7 ms). It originates near the origin of the left pulmonary artery and ends in the descending aorta immediately distal to the left subclavian artery.86 Machado et al found that between 16 and 30 weeks the TPV was significantly shorter in the pulmonary artery than in the aorta (32. combined cardiac outputs have been found to increase from approximately 100 ml/minute at 18 weeks’ gestation to approximately 1000 ml/minute at 38 weeks’ gestation. In contrast.162 Fetal Cardiology Therefore. which increases exponentially from 0.12. In these examinations peak blood flow velocities in the ascending aorta were found to exceed those in the main pulmonary artery.44 Since the pulmonary artery has a slightly larger diameter than the aorta (approximately 1. progressive enhancement of active relaxation and/or changes in loading conditions are more likely to explain variations in fetal AV blood flow than changes in ventricular compliance. with both vessels showing a linear increase with gestational age.91 Most authors agree that right ventricular pressures are predominantly determined by the blood flow resistance in the ductus arteriosus.85 Calculation of the valvular area allows the estimation of the cardiac output when the TAV and heart rate are known or if the velocity integral and the heart rate are known.3:1) it is expected that the right ventricle contributes to a larger proportion of the cardiac output. At 13 weeks.1 ms compared to 43. outflow tract velocities average 30 cm/s. In the human fetus it has been confirmed that the pulmonary arterial diameter consistently exceeds the aortic diameter.78. the ductus arteriosus exhibits increasing sensitivity toward prostaglandin. In addition. The ratio between right and left ventricular output decreases from 1.84. suggesting that in the mid-trimester fetus the mean pulmonary arterial pressure is higher than in the aorta. Several studies have come to different conclusions regarding the TPV.84 By 20 weeks. Flow velocity waveforms in the ductus arteriosus The ductus arteriosus provides a conduit between the main pulmonary artery and the descending aorta. was found. because a flat velocity profile can be suggested for the blood flow in the ascending aorta and the pulmonary valve directly behind the semilunar valves. However.82 The improvement in systolic cardiac function between 13 and 20 weeks results in an increase in output from both ventricles. which is also associated with increasing peak blood flow velocities in both outflow tracts. Systolic peak blood flow velocities in this vessel increase from 50 cm/s at 15 weeks’ gestation to 130–160 cm/s near term.

103. high-resistance vascular bed.81. A longer diastolic filling time has been found on the left side of the heart (197 ms) compared with the right (174 ms).103 The high blood flow resistance in the pulmonary circulation is predominantly determined by the overall small cross-sectional area of the pulmonary vascular bed. fetoplacental circulation).52) and remains relatively constant throughout pregnancy.74.7% of the common cardiac output reaches the pulmonary circulation in mid-gestation. but findings regarding the development of the pulsatility index from 15 weeks to term differ amongst investigators. the ductus arteriosus. After 16 weeks there is increasing arborization of the pulmonary vascular tree. internal carotid. and a low diastolic forward flow.81.98 It is of note that in intracerebral vessels end-diastolic velocities are present between 11 and 13 weeks. Right ventricular afterload is determined by the blood flow velocities in the pulmonary circulation.105.Development of fetal cardiac and extracardiac Doppler flows 163 End-diastolic flow can first be detected between 13 and 14 weeks’ gestation and is generally established by 17 weeks’ gestation. The blood flow resistance in these vessels develops similarly. since it is readily accessible at close to 0° insonation.91–94 Measurement of the timing of cardiac contractions has allowed a comparison to be made between the duration of individual cardiac cycle phases for individual ventricles.95. a sharp mid-systolic deceleration. changes in vascular resistance in the common carotid arteries reflect predominantly cerebral vascular blood flow resistance (external carotid.96 The significant decrease in heart rate between 10 and 20 weeks’ gestation results in a lengthening of the cardiac cycle from 373 to 406 ms.9 in mild cases and as low as 1.106 The high pulmonary vascular resistance compared to the fetal systemic circulation and the preferential blood flow through the patent ductus arteriosus into the descending aorta may explain this blood flow pattern.105. The pulmonary circulation The vascular connections between intrapulmonary arterioles and the main pulmonary vessels are established between 35 and 50 days after conception.99.15 Several studies have been undertaken to investigate vascular development in the branch pulmonary arteries. In utero the pulmonary circulation is a low-flow. and cerebral arteries). The distal branches. which is more marked in diastole. a decrease in the pulsatility index is observed in all cerebral vessels toward term.104 Therefore. Most authors report a modest rise in the pulsatility index until approximately 26 weeks. This results in a fall of the pulsatility index below 1. Concurrently. with a subsequent fall toward term. and the vascular beds distal to the ductus (descending aorta with its organ branches.98 Thereafter.97. improved ventricular contractility. a deep notch or reverse flow in early diastole. This early in pregnancy it . The decrease of the pulsatility index with advancing gestation suggests a decrease in cerebral blood flow resistance. In contrast. increasing compliance.105. The pulmonary blood flow pattern at the proximal and middle branch level is characterized by a rapid early systolic flow acceleration. fetoplacental blood flow resistance is a more important determinant of the flow velocity in the main pulmonary artery.93 The pulsatility index in the ductus arteriosus is relatively high (2.101. and therefore many authors prefer the middle cerebral artery. and have yielded inconsistent results.94 In constriction of the ductus arteriosus there is an increase of arterial peak blood velocities. however.100 Since the pulsatility index is also influenced by filling pressure. and decreasing cardiac afterload.46 ± 0. The cerebral circulation Intracerebral flow velocity waveforms have been visualized as early as 10 weeks of gestation.106 The peak velocity and pulsatility of blood flow significantly decrease if the Doppler interrogation moves from the proximal to the distal arterial branches.105–109 Laudy et al found Extracardiac arterial Doppler indexes The blood flow resistance in the ascending aorta and therefore left ventricular afterload is predominantly determined by the brachiocephalic circulation (predominantly the common carotid arteries and to a lesser extent the left subclavian artery). The pulsatility index of the intracerebral vessels and the internal carotid artery is relatively constant until 20 weeks of gestation. This proportion increases to 7% at term. there is a significant increase of the diastolic filling time associated with a linear decrease in isovolumetric relaxation and ejection time.81 The pulsatility index in the middle cerebral artery is higher than in all other cerebral vessels.97 Accordingly. In the fetal lamb only 3. which continues toward term.0 with associated tricuspid insufficiency in severe constriction.102 The intrapulmonary arteries possess a relatively thick muscular coat and therefore have a small lumen. the ejection time is shorter for the left ventricle (174 ms) than on the right side of the heart (189 ms).106 A high pulsatility index has been documented by several investigators. an increase in cerebral blood flow volume may be an explanation for this observation. This development documents several maturation processes in fetal cardiac function including elevation of myocardial relaxation velocity.76 is often not possible to distinguish between the individual cerebral arteries. shows a monophasic forward flow pattern with lower pulsatility and acceleration and deceleration velocities.

Between 16 and 20 weeks there is a rapid fall of the pulsatility index. representing between 17 and 31% of combined ventricular output. and was a mean of 22% of combined ventricular output.113. End-diastolic blood flow is established at a similar time in gestation to that in the umbilical artery. determines the observed increase of pulmonary blood flow. At 38 weeks. the vascular beds behave differently with respect to the development of resistance indexes. Since the pulsatility index in the descending aorta remains relatively constant. Doppler sonography has been performed in many vascular territories originating from the descending aorta. .106 Further research into these relationships is necessary. and iliac arteries.114. indicating the high blood flow resistance in these vessels. end-diastolic blood flow velocities become detectable between 13 and 15 weeks of gestation. suggesting that the increase of vessel diameter. These include splenic. adrenal.111 However. It appears safe to assume that there is a progressive increase in pulmonary blood flow with advancing gestation in the human fetus. it is presumed that the sum of all these divergent changes in flow resistance overall cause little change in aortic blood flow resistance.13 Foramen ovale blood flow increased exponentially three-fold. while in the third trimester an increase in lung volume and overall vascular cross-sectional area may be the major contributor to this development. and. mesenteric. the proportion of pulmonary blood flow increased from 13 to 25% and maintained a constant pattern thereafter. further distally.115 while the mesenteric and hepatic arteries represent high-impedance vascular beds.90. rather than a rise of flow velocities.164 Fetal Cardiology no significant change in the pulsatility index in the proximal pulmonary branch artery.106 The lack of change in branch pulmonary artery pulsatility index in later gestation could suggest that during this time lung growth and increase in vascular cross-sectional area result in an in overall increase in blood flow in this area.105–110 Indeed. renal. the femoral arteries113–119 (Figure 11.105–109 Recently. indicating that there is relatively little change in blood flow resistance in these vessels after 20 weeks of gestation.118 The blood flow pattern in the internal iliac artery is predominantly determined by placental and umbilical cord blood flow resistance and therefore shows a different development during gestation.119 The splenic artery shows a temporary rise in the pulsatility index in mid-gestation with a subsequent decline toward term. This has been attributed to divergent development of blood flow resistance in distal vascular beds. The proportions of these flows remained unchanged through the second and third trimesters of pregnancy.109 Laudy et al demonstrated a similar decrease in the pulsatility index between 20 and 30 weeks followed by a significant increase from 31 weeks until term.114 Importantly. despite a linear increase in blood flow volume.114.111 Despite this increase in total pulmonary flow during the second half of gestation. This decrease was linear until approximately 31 weeks and then remained relatively unchanged. Thereafter there is a continuous decrease in pulsatility index paralleling the development in the placental circulation. Rasanen et al found a decreasing pulsatility index with gestational age in the proximal and distal branch pulmonary arteries. Sivon et al found a slight constant increase of pulsatility index in the proximal.6). After this time the pulsatility index in the descending aorta remains relatively constant toward term. middle. In the second trimester this may be due to a decrease in pulmonary vascular impedance to blood flow. therefore later than in the cerebral circulation. middle.105.114.112 This is analogous to the development in the ductus arteriosus.116. they documented that the proportion of foramen ovale blood flow decreased from 34 to 18% with little change in the proportions after 30 weeks.112.118 Throughout gestation there is a diastolic flow reversal in this vessel and the external iliac artery. the right ventricular output was 60% of the combined cardiac output. and distal segments of the branch pulmonary artery.105–111 Descending aorta and fetal visceral blood flow In the descending aorta. the different blood flow resistance is likely to have an impact on the regional distribution of blood flow in the vascular beds distal to the descending aorta. The examination of the external iliac arteries and femoral arteries reflects development of vascular resistance in the lower body. In the femoral artery there is a linear increase in the pulsatility index with advancing gestation. hepatic. but without statistical significance.117 Distal to the bifurcation into the iliac arteries. Rasanen et al established that. the TAMX and peak systolic velocity at the proximal. a linear decrease in the pulsatility index has been observed from 20 to 38 weeks. It is likely that there is a fall in the pulmonary vascular impedance with gestational age. In the renal and adrenal arteries. indicating that the development of the human fetal pulmonary circulation has an important role in the distribution of cardiac output.13 Similarly. from 20 to 30 weeks of gestation.108 Their findings suggest that analysis of the unique waveform of the branch pulmonary arteries using the pulsatility index may not accurately reflect downstream vascular resistance.105 They concluded that fetal branch pulmonary arterial vascular impedance decreases significantly between 20 and 30 weeks and suggested an increase of pulmonary vascular resistance during the last 8–10 weeks of pregnancy. Sutton et al found that blood flow through the lungs increased exponentially with gestational age by almost four-fold between 18 and 37 weeks. and distal pulmonary branch level did not reveal significant gestational age-related change. but a significant gestational age-related drop in the systolic/diastolic (S/D) ratio.

Increasing differentiation of the villous vascular tree into tertiary villi results in an overall decrease in placental blood flow resistance.132–134 In contrast to the adult133 there is continuous antegrade flow toward the left atrium in the human fetus. This may be due to the divergent development of blood flow resistance in downstream vascular beds. In the first trimester the umbilical vein has a pulsatile flow pattern. which is apparent for the precordial veins. the volume flow per unit body weight changes little with gestational age. With increasing gestational age there is a continuous decrease in venous Doppler indexes.120. especially between 11 and 13 weeks of gestation. Concurrent with the decrease in venous Doppler indexes the proportion of fetuses with pulsatile flow in the free umbilical vein decreases. 20 25 30 Gestational age (weeks) 35 40 Uteroplacental blood flow The development of the flow velocity waveform in the maternal and fetal compartments of the placenta is suggestive of a progressive fall in vascular resistance with advancing gestational age. End-diastolic flow is first appreciable at 13–15 weeks’ gestation. and is inversely related to the left atrial pressure pulsations. This decrease in indexes is most marked between 13 and 20 weeks’ gestation.124 These changes in flow velocity waveforms show the increased development of the villous vascular tree and the development of tertiary villi. RV. and there is a continuous rise in end-diastolic velocities contributing to the linear fall in umbilical artery pulsatility index toward term.121–123 Examination of the umbilical flow velocity waveform using Laplace transform techniques suggests that vessel wall tone also decreases at the beginning of the second trimester.132–135 This continuous forward flow even during atrial systole suggests .6 Development of ventricular afterload in gestation.128–130 This is predominantly due to an increased antegrade flow component during atrial systole. and is likely to represent the combined effects of improved cardiac diastolic function and decreasing afterload. In contrast.126 It is of note that. ductus venosus blood flow is antegrade throughout the cardiac cycle. The diastolic notch in the spiral artery flow velocity waveform disappears by 13 weeks’ gestation.125. Extracardiac venous Doppler indexes The blood flow volume in the umbilical vein approximates 140–180 ml/kg per minute in the second trimester and decreases slightly to 110–170 ml/kg per minute toward term. although umbilical venous volume flow increases linearly with fetal weight. In contrast. the highest venous blood flow velocities are found in the ductus venosus.70. The inferior vena cava and hepatic veins have a flow velocity waveform with reverse flow during atrial contraction. intervillous flow fully develops by 14 weeks’ gestation.121 To evaluate the placental blood flow resistance in the fetal compartment the umbilical arteries were examined.127–131 The flow velocity waveform in the pulmonary veins has a biphasic profile with a systolic peak and a second peak during the early part of ventricular diastole.Development of fetal cardiac and extracardiac Doppler flows 165 Afterload PI femoral artery PI external iliac artery PI hepatic artery LV PI DA PI DAO RV PI splenic artery PI renal artery PI umbilical artery PI carotid arteries PI cerebral arteries Figure 11. and a constant Doppler flow pattern is generally established at approximately 13 weeks’ gestation. Therefore. resulting in an overall decrease in intraventricular end-diastolic pressures. right ventricle. These range between 65 and 75 cm/s during systole in the first trimester and increase linearly toward term. there is little change in the pulsatility index in the ductus arteriosus (DA) and descending aorta (DAO).123. With advancing gestation there is a steady decrease in the pulsatility index (PI) in the cerebral circulation and therefore left ventricular (LV) afterload. followed by the arcuate arteries within a further 2 weeks.127 There are differences in flow velocity waveforms in precordial venous vessels.73. This is partly due to the fact that there is marked antegrade acceleration of blood flow as umbilical venous blood enters the ductus venosus.

umbilical artery. Therefore. contributing to a significant decrease in the S/D ratio. During the second and third trimesters relative distribution of the right and left ventricular outputs is influenced by the changes in afterload of the individual ventricles.138 End-diastolic velocities are discernible in the cerebral circulation before the fetoplacental circulation.132 This development is likely to be the result of several factors. UV. which is followed by elevation of the E/A ratio. The latter changes occur in association with a decrease of pulsatility in precordial venous Doppler flow velocity waveform and disappearance of venous pulsation in the free umbilical portion of the umbilical vein.136.166 Fetal Cardiology that the fetal extraparenchymal pulmonary venous system is not spacious and compliant. In addition. Steady decrease in baseline heart rate Heart rate Steady increase in cardiac compliance and contractility 8 13 15 17 20 24 28 32 36 40 Gestational age (weeks) . while isovolumic relaxation occupies approximately 16%. Illustrated is the approximate temporal relationship of arterial and venous flow velocity changes and cardiac function with advancing gestational age. the decreasing S/D ratio in the pulmonary venous vessels documents improved left ventricular diastolic function and increased preload. These changes indicate an enhanced antegrade venous flow toward the atrial chambers. Enhanced diastolic function is manifested by changes in intracardiac and extracardiac flow velocity waveforms – there is change from a monophasic to a biphasic flow velocity profile during ventricular inflow by 8–10 weeks. a significant decrease in the atrial contribution to ventricular filling has been demonstrated between 10 and 14 weeks of gestation. ejection time and isovolumic relaxation time show a significant decrease. end-diastolic flow. UA. Indeed. At 6 weeks of gestation approximately 20% of the cardiac cycle is occupied by the isovolumic contraction. Despite this lengthening of the cardiac cycle.134 Sequential changes of flow velocity waveforms The immature fetal heart has limited diastolic and systolic function in the first trimester. limiting its ability to accommodate changes in cardiac preload and afterload. since the ‘afterload sensitivity’ of the fetal ventricular myocardium limits the ability to regulate cardiac output by other mechanisms. Between 10 and 20 weeks of gestation there is a significant increase in cardiac cycle length due to a decrease in fetal heart rate. With ongoing development there is a Decreasing pulmonary venous indexes Decreasing precordial venous indexes UV pulsations Constant UV flow Increasing impedance lower body EDF UA EDF MCA Decreasing placental impedance Decreasing cerebral impedance Steadily rising cardiac output Steadily rising E/A ratio Exponentially rising cardiac output Exponentially rising E/A ratio Figure 11.75 The persistent decrease in right and left ventricular afterload in the first and early second trimesters associated with the increase in preload and growth of the fetal heart result in an exponential increase in combined cardiac output.136 The concomitant improvement of diastolic function is necessary for an efficient accommodation of this increased preload.132 presumably due to improved diastolic ventricular filling facilitating atrial emptying during diastole. there may be a decrease in the pressure gradient toward the left ventricle. indicating improved diastolic function and the increase in cardiac output necessary to maintain adequate growth.135 From approximately 20 weeks onward there is a marked increase in antegrade peak blood flow velocities.7 Development of cardiac and extracardiac functionality with gestation. suggesting effective heart compliance by this time.75 The isovolumic contraction time shortens progressively and is not measurable after 12 weeks. There may be a significant increase in pulmonary blood flow volume during this time. umbilical vein. which ultimately results in an increase in preload.133. MCA. middle cerebral artery. EDF. extraparenchymal pulmonary venous pressure may be higher than the left atrial pressure even at atrial systole. Cardiac development during the following weeks results in improved cardiac function throughout the cardiac cycle. Therefore.137 Changes in ventricular afterload are of special significance in regulating organ flow in the fetus.

Philadelphia: WB Saunders. Plappert T. 10. De Smedt MC. Tynan MJ. 9. resulting in relative placental insufficiency at term. there is a steady decrease in the umbilical artery and therefore the internal iliac pulsatility index. 1992: 609–19. Br Heart J 1994. 14: 607–12. Baschat AA. Blaas HG. The increased contractility and preload and decrease in afterload are associated with increased peak blood flow velocities and an exponential increase of cardiac output during gestation. Tynan M. Doppler echocardiographic evaluation of the normal human fetal heart. Fetal cardiac arrhythmias and their effect on volume blood flow in descending aorta of human fetus. 14: 307–10. After 18 weeks’ gestation the relationship between cardiac output and fetal body weight remains relatively constant at 450 ml/kg per minute. Non-invasive measurement of human fetal circulation using ultrasound: a new method. Matias A. eds. Circulation 1998. Allan LD. Rao PS.126 Improved resolution of modern ultrasound machines allows the study of small-caliber vessels. Sharland GK. Fetal cardiac output estimated by Doppler echocardiography during mid. 20. 22. 70: 1–6. Distribution and regulation of blood flow in the fetal and neonatal lamb. 12: 380–4. Sharland G. 18.117 With delivery of the fetus. Functional anatomic development in the fetal heart. J Clin Ultrasound 1986.113. Reed KL. Long WA. 60: 338–42. Owing to this development. 25: 341–3.139–142 In addition. Hellevik LR. Henry GW. Sutton MS. 5. Blood flow and the degree of shunting through the ductus venosus in the human fetus. Matias A. 21. Nicolaides KH. Rudolph AM. 115: 151–73. Br Med J 1977. the liver and spleen appear to be preferentially perfused organ systems. Effects of heart rate on ventricular size. Aeschlimann S. A new application of tissue doppler imaging. Montenegro N. Chacko AW. 15. Drumm JE. The natural history of the hypoplastic left heart syndrome. Allan L. 13. Sutton MG. Kiserud T. Carrier O. Studies of human fetuses with growth restriction suggest that the cerebral. and output in the normal human fetus: a prospective Doppler echocardiographic study. Embryonic development of the sphincter of the ductus venosus. Philadelphia: WB Saunders. J Am Soc Echocardiogr 1990. Int J Cardiol 1989.Development of fetal cardiac and extracardiac Doppler flows 167 progressive drop in the pulsatility indexes and presumably also blood flow resistance in the brachiocephalic circulation. Al-Ghazali W. Fetal and Neonatal Physiology. 12. Nicolaides KH. 1992: 629–45. Gomes C. 76: 52–8. 71: 232–7. Skulstad S. Circulation 1987. Guyton AC. Philadelphia: WB Saunders. Normal fetal foramen flap and transatrial Doppler velocity pattern. Fetal and Neonatal Physiology. In: Polin RA. Circ Res 1978. Although placental blood flow increases from 115 ml/minute from 20 weeks to 415 ml/minute at term. Groves A. Obstet Gynecol 1987. Flack N. Cardiac defects in chromosomally normal fetuses with abnormal ductus venosus blood flow at 10–14 weeks. Allan LD. Shenker L. Heymann MA. In: Polin RA. Screening for chromosomal abnormalities at 10–14 weeks: the role of ductus venosus blood flow. 17. Sahn DJ. The combined cardiac output increases steadily toward term. Gill T. 23. 1992: 598–609. 2: 1450–1. Anderson CF. Ultrasound Obstet Gynecol 1999. Reynolds SR. Am J Cardiol 1987. 42: 426–33. it decreases in relation to fetal body weight. 16. Br Heart J 1987. Nagueh SF. the major proportion of descending aorta blood flow is distributed to the placental vascular bed for oxygenation. the fetal circulation gathers its postnatal functionality by closure of the central shunts143 (Figure 11. 15(Suppl I): 60–9. Autonomic and central neuroregulation of fetal cardiovascular function. Knöpfle G. References 1. Wilson AD. Lancet 1991. stroke volume. Crawford DC. 8. MacNeill A. Liver and ductus venosus blood flows in fetal lambs in utero. Rudolph AM. and coronary circulations are capable of autoregulation in the second trimester. 4. 57: 811–21. Kopelen HA et al. Ultrasonographic velocimetry of the fetal ductus venosus. FitzGerald DE. Visser GH. Areias JC. 11. Kenny J. Doubilet P. Chita SK. Anat Rec 1953. eds. Circ Res 1985. 7. Ultrasound Obstet Gynecol 1999. 182: 147–53. Sutton MS. 14: 311–14. Rasmussen S. 3. Noordam MJ. Ultrasound Obstet Gynecol 1998. Kiserud T. adrenal. Huggon I.115. Ross JH. Plappert T. 2. 28: 90–6. Firsttrimester fetal heart block: a marker for cardiac anomaly. Mikati I. Edelstone DI. 98: 1644–50. eds. 3: 491–4. Stewart PA. Assessment of changes in blood flow through the lungs and foramen ovale in the normal human fetus with gestational age: a prospective Doppler echocardiographic study. Montenegro N. Fox WW. Tonge HM. 6. This divergent development of ventricular afterload results in physiological redistribution of the cardiac output in favor of the left ventricle and therefore the upper body. Fetal and Neonatal Physiology. Wladimiroff JW. Circ Res 1964. Am J Obstet Gynecol 2000. Salzman D. from 20 weeks onwards pulsatility indexes in the ductus arteriosus and the descending aorta remain relatively constant. 338: 1412–14. While vascular resistance in the limbs and therefore the external iliac arteries increases. development of Doppler indexes is also divergent in pregnancy. Marx GR. Fetal venous circulation – an update on hemodynamics. Teitel DF. J Perinat Med 2000. Kiserud T. Evidence for tissue demand as the major factor causing autoregulation. Hansmann M. Meijboom EJ.7). Physiologic development of the cardiovascular system of the fetus. 14. 57: 528–33. Cardiac Doppler flows during fetal arrhythmias: physiologic consequences. Fox WW.and late gestation. Fox WW. Gembruch U. At the same time. In: Polin RA. Eik-Nes SH. 19. . In the vascular beds distal to the descending aorta. Doppler estimation of left ventricular filling pressure in sinus tachycardia.

In: Maulik D. Gilbert RD. Seagren SC. 163: 776–84. Early Hum Dev 1994. 1995. Molecular basis of cardiac performance. Circ Res 1968. 44. Struijk PC. Whitsett JA. Barany M. Dirschedl P. 242: H834–43. Hop WC et al. Fetal heart rate in trisomy 21 and other chromosomal abnormalities at 10–14 weeks of gestation. Littler WA. Circ Res 1987. 39. Comline RS. Anderson C. 60: 804–14. Ramos C. 54. 15: 20–31. Ingwall JS. de Ridder MA. Am J Cardiol 1970. Moise KJ Jr. 53. 25: 213–26. In vitro validation of Doppler waveforms indices. Doppler Ultrasound – Principles and Instruments. 34. 250: H407–13. 95: 539–45. Yarlaggada P. Kirby ML. Keller BB. Watanabe AM. Mahdavi V. McNellis D. Thyroid hormone regulation of beta-adrenergic receptor number. Circ Res 1987. Matz S. Jones LR. Nathanielsz PW. Doppler Ultrasound Measurement of Normal Fetal Hemodynamics. Montenegro N. Sweeney LJ. Ursem NT. Bahl JJ. 9: 711–16. Circ Res 1989. Cummins P. Smrcek JM. Pollinger J. 14: 397–401. Izumo S. Ventricular function and morphology in chick embryo from stages 18 to 29. 28. Embryonic heart activity: appearance and development in early human pregnancy. Am J Physiol 1982. 61: 465–83. 56. 33. Assessment of fetal heart rate variability and velocity variability by Doppler velocimetry of the descending aorta at 10–20 weeks of gestation. Wladimiroff JW. Schwartz K. 51. Cheng SY. Shi C. 48. 60. ATPase activity of myosin correlated with speed of muscle shortening. Ultrasound Obstet Gynecol 1998. Philadelphia: WB Saunders. Fetal and Neonatal Physiology. Mari G. 38. Circ Res 1987. 47. 14: 559–68. 84: 1693–700. Schmidt KG. Cheng SY. Br J Obstet Gynaecol 1991. Maturation of energy metabolism in the lamb: changes in myosin ATPase and creatine kinase activities. 27. Effects of increasing afterload on left ventricular output in fetal lambs. 37: 601–5. 11: 274–6. Reed K. 4: 299–309. Dunlop W. Thyroid hormone influences beta myosin heavy chain (beta MHC) expression. Nadal-Ginard B. Indik JH. Wladimiroff JW. Wisser J. 29. 31. J Clin Invest 1989. Developmental changes in the ultrastructure and sarcomere shortening of the isolated rabbit ventricular myocyte. Philadelphia: WB Saunders. 26. Am J Obstet Gynecol 1990. Wladimiroff JW. Jacobowitz D. Morkin E. Snijders RJ. Edwards JG. An estimate of fetal autonomic state by spectral analysis of human umbilical artery flow velocity waveforms. Hop WC. Embryonic heart rate in dated human embryos. Pediatr Res 1987. Reproducibility of ultrasonic measurement of fetal cardiac haemodynamics. 55. 58: 683–4. Clark EB. Besch HR Jr. Lecarpentier Y. 7: 239–44. Fetal Diagn Ther 2000. Price KM. 1992: 589–97. Acute fetal hemodynamic alterations after intrauterine transfusion for treatment of severe red blood cell alloimmunization. Hyett JA. Nakanishi T. Ultrasound Obstet Gynecol 1996. Ursem NT. Nassar R. Kramer MF. Hunter S. Flink IL. Friedman WF. Silver M. Developmental and hormonal regulation of sarcomeric myosin heavy chain gene family. J Endocrinol Invest 1991. Woodman D. 43. Shenker L. Am J Obstet Gynecol 1990. Barros H. Ursem NT. In: Polin RA. Excitation-contraction coupling in neonatal and adult myocardium of cat. J Ultrasound Med 1990. Rempen A. Pediatr Res 1982. 45. 199: 1482–8. Nadal-Ginard B. Development of myocardial contractile system in the fetal rabbit. Diagnosis of viability in early pregnancy with vaginal sonography.168 Fetal Cardiology 24. Pool PE. Clark EB. J Biol Chem 1986. 65: 127–34. Biochem Biophys Res Commun 1994. Manasek FJ. Montenegro N. Yamauchi A. Matias A. Myosin isoenzymic distribution correlates with speed of myocardial contraction. 23: 25–32. Variation of embryonic/fetal heart rate at 6–13 weeks’ gestation. Kamata K et al. J Biol Chem 1977. Clin Sci 1998. 35. Hawkins J. Anderson PA. Plasma triiodothyronine concentration in the foetal and newborn lamb. Okuda H. 49. Heart rate and flow velocity variability as determined from umbilical Doppler velocimetry at 10–20 weeks of gestation. 252: 2787–9. 36. Martin J et al. 42. Arisio R. Hathaway DR. 37. Foetal myosin light chain in human ventricle. Developmental changes in cardiac sarcoplasmic reticulum in sheep. Variation and correlation in human fetal umbilical Doppler velocities with fetal breathing: evidence of the cardiac–placental connection. 97: 989–94. Clark EB. Fox WW. Cardiac conduction system: fetal and postnatal development. Astle C. 13: 1071–5. Reed KL. 31: 333–5. Morkin E. Effects of afterload and baroreceptors on cardiac function in fetal sheep. Braunwald E. Noble PL. 57. neonatal. Mahony L. 61. Van Hare GF. Friedman WF. Maylie JG. Cardiovasc Res 1998. Kempski MH. Biometry of the fetal heart between 10 and 17 weeks of gestation. Fisher DJ et al. Gembruch U. 50(Suppl): 197–218. eds. Am J Physiol 1986. Mahdavi V. 52. Hu N. J Gen Physiol 1967. Thyroid hormones in tissues from human embryos and fetuses. 30. and adult myocardium. 124: 562–87. 163: 1792–6. Beeby AR. Z Anat Entwicklungsgesch 1965. Electron microscopic observations on the development of S-A and A-V nodal tissues in the human embryonic heart. Nicolaides KH. New York: Perinatology Press. Ultrasound Obstet Gynecol 1999. Jansen CA. Sympathetic innervation of the developing rabbit heart. 50. James TN. 32. 41. Maulik D. J Muscle Res Cell Motil 1980. 25. Costa A. 1: 357–66. 58. 37: 107–15. 15: 1128–33. Williams LT. 61: 287–95. 40. Embryonic heart rates before the seventh week of pregnancy. Transitions in cardiac isomyosin expression during differentiation of the embryonic chick heart. Lefkowitz RJ. J Mol Cell Cardiol 1981. Biochemical and histochemical comparisons of fetal. 22: 201–7. Schats R. Reedy MC. 257–82. 46. J Endocrinol 1973. 16: 463–9. Development of the fetal heart. Heads A. Benedetto C. Zak R. Rudolph AM. J Reprod Med 1986. . Kremkau F. Pediatr Res 1981. 59. Plasticity of the myocardium generated through protein isoform switches. 261: 152–7. Br J Obstet Gynaecol 1990. eds. J Dev Physiol 1982. 98: 807–14. Dummett JL et al. Beta-adrenergic receptors and catecholamine sensitive adenylate cyclase in developing rat ventricular myocardium: effect of thyroid status.

Romanini C. 89: 953–6. J Am Coll Cardiol 1991. Arduini D. Umbilical venous Doppler velocity pulsations and inferior vena cava pressure elevations in fetal lambs. eds. 182: 835–8. J Matern Fetal Invest 1991. Br Heart J 1988. Chaffin DG. 93. Cardiac functional changes in the human fetus in the late first and early second trimesters. Anderson CF. Br Heart J 1991. 66: 285–9. Stijnen T. Saltzman DH. Plappert T. Assessment of right and left ventricular function in terms of force development with gestational age in the normal human fetus. Gill T. Wladimiroff JW. Normal flow velocity waveforms in the fetal ductus arteriosus in the first half of pregancy. Chaoui R. 114: 836–51. Doppler echocardiographic assessment of time to peak velocity in the aorta and pulmonary artery of small for gestational age fetuses. Ultrasound Obstet Gynecol 1991. 84. Wladimiroff JW. Obstet Gynecol 1997. 15: 8–19. 66. 80. Doubilet P. 164: 331–5. Wladimiroff JW. 88. Anderson CF. 68. Binotto CN et al. Reed K. 3: 30–4. Reed KL. Reed KL. 5: 168–74. Arduini D. Taddei F. Am Heart J 1987. Sharkey AM et al. Snijders R. 2: 434–45. Wood DC. 18: 532–6. Arduini D. Romanini C. 78. Chiba Y. 85. 59: 639–40. 64. Circulation 1986. Pearson AC. The value of fetal arterial. Silverman NH. 167: 736–9. Severi FM. Mancuso S. 74: 1208–16. Paladini D. 95: 65–9. Reed KL. Plappert T. Newman AT. Kanzaki T. Am J Obstet Gynecol 1991. Sahn DJ et al. 14: 169–74. Campbell S. Doppler blood velocities in the first trimester of pregnancy. 91: 129–38. 79. Brezinka C. Huhta JC. Acceleration time in the aorta and pulmonary artery measured by Doppler echocardiography in the midtrimester normal human fetus. 73: 41–6. Chita SC. Echocardiographic assessment of ventricular filling pressure during the second and third trimesters of gestation. Velocimetrie Ultrasonore Doppler 1974. Russo MG et al. 75. Taylor MG. Am J Obstet Gynecol 2000. New York: Raven Press. Doppler echocardiography of fetal ductus arteriosus constriction versus increased right ventricular output. Stijnen T. van Splunder IP. Leiva MC. Barendregt LG. intracardiac. 69. Rizzo G. 11: 173–9. Wladimiroff JW. Capponi A. Meijboom EJ. Doppler echocardiographic assessment of atrioventricular velocity waveforms in normal and small-for-gestational-age fetuses. 1994: 325–32. Ultrasound Obstet Gynecol 1996. cardiac and venous flows in predicting pH and blood gases measured in umbilical blood at cordocentesis in growth retarded fetuses. Premature constriction of the ductus arteriosus. Wladimiroff JW. Arduini D. Stijnen T. Romanini C. van Splunder P. Ultrasound Obstet Gynecol 1992. Romanini C. Assessment of fetal compromise by Doppler ultrasound investigation of the fetal circulation. Tulzer G. Br J Obstet Gynaecol 1990. 165: 668–74. Tolosa JE. Pourcelot L. van der. 87: 617–20. In: Copel J. Hecher K. Sharkey A. Romanini C. Changes in umbilical venous velocities with physiologic perturbations. Applications cliniques de l’examen Doppler transcutane. Doppler Ultrasound in Obstetrics and Gynecology. Gudmundsson S. 71. Anderson CF. Rizzo G. Huhta JC. Fetal Diagn Ther 2000. . Hecher K. Fetal Diagn Ther 1990. Burghouwt MT. Fetal atrioventricular flow-velocity waveforms and their relation to arterial and venous flow-velocity waveforms at 8 to 20 weeks of gestation. and venous blood flow velocity studies. Influence of alteration in preload on the pattern of left ventricular diastolic filling as assessed by Doppler echocardiography in humans. Changes in intracardiac blood flow velocities and right and left ventricular stroke volumes with gestational age in the normal human fetus: a prospective Doppler echocardiographic study. 65. Huhta J. 70. Doppler echocardiographic assessment of fetal cardiac function. Doubilet P et al. Br J Obstet Gynaecol 1988. Reed KL. Chaffin DG. Obstet Gynecol 1996. 73. 81. Arduini D. Rizzo G et al. 67. Anderson CF. Br Heart J 1987. Rizzo G. Circulation 1989. 83. 1: 73–8. Pacileo G. Evaluation of left ventricular diastolic function: clinical relevance and recent Doppler echocardiographic insights. Mooren K. Intrauterine growth retardation and fetal cardiac function. 7: 411–15. 1: 175–9.Development of fetal cardiac and extracardiac Doppler flows 169 62. J Am Coll Cardiol 1986. Reed KL. Rizzo G. 72. Rizzo G. 97: 603–7. Fetal atrioventricular and outflow tract flow velocity waveforms during normal second half of pregnancy. Doppler echocardiography of the main stems of the pulmonary arteries in the normal human fetus. 63. 86. 8: 391–5. Circulation 1996. 89. Nicolaides K. Interpretation of pulsatility index in feeder arteries to low impedance vascular beds. 58: 15–18. Arterial. Gosling RG. Scagnelli S. Cohen AW. Rizzo G. Stewart PA. Shenker L. 77. Circulation 1986. Kern MJ et al. Doppler echocardiographic studies of diastolic function in the human fetal heart: changes during gestation. Am J Obstet Gynecol 1991. Sutton MS. Ultrasound Obstet Gynecol 1994. Nicolaides K. Machado MV. Labovitz AJ. Ultrasound Obstet Gynecol 1992. Bocchi C et al. 79: 1226–36. 82. Br J Obstet Gynaecol 1995. 91. Harrington K. Allan LD. 76. Fetal cardiac development and hemodynamics in the first trimester. 2: 397–401. Fetal cardiac and extracardiac circulation in early gestation. Ultrasound Obstet Gynecol 1998. Ultrasound Obstet Gynecol 2000. Ultrasound Obstet Gynecol 1999. Kenny JF. Am J Obstet Gynecol 1992. Evaluation of the preload condition of the fetus by inferior vena caval blood flow pattern. 95. J Am Soc Echocardiogr 1990. 16: 128–32. Normal fetal cardiac flow velocity waveforms between 11 and 16 weeks of gestation. 87. Pearson AC. Huisman TWA. The fetal ductus arteriosus. Campbell S. Sahn DJ. 102: 963–9. Lo PTS. 34: 625–7. Doyle P. Acceleration time in the aorta and pulmonary artery measured by Doppler echocardiography in the midtrimester normal human fetus. Tulzer G. 94: 1372–8. 94. Reference ranges for fetal venous and atrioventricular blood flow parameters. Stoddard MF. Umbilical venous velocity pulsations are related to atrial contraction pressure waveforms in fetal lambs. 90. Cardiac Doppler flow velocities in human fetuses. Circulation 1995. 4: 381–90. Mancuso S. 92. 74.

125. Lees C. Cartier MS. Ultrasound Obstet Gynecol 1996. 10: S19. 103. Doppler Ultrasound in Obstetrics and Gynecology. Huhta JC. Romanini C. Rizzo G. van den Wijngaard JA. J Anat 1972. 41: 897–901. Circ Res 1970. 53: 144–51. 127. 97. Br J Obstet Gynaecol 1990. Wladimiroff JW. Copel JA. 1987. 79: 605–10. Superior mesenteric artery flow velocity waveforms in small for gestational age fetuses. 96: 845–9. Weiner S. Fetal and umbilical flow velocity waveforms between 10–16 weeks’ gestation: a preliminary study. Doppler ultrasound assessment of cerebral blood flow in the human fetus. Emerson DS. 126. Is the liver of the fetus the 4th preferential organ for arterial blood supply besides brain. 78: 812–14. J Ultrasound Med 1991. 16: 453–6. Normal fetal Doppler inferior vena cava. 116. 167: 675–7. Norwalk: Appleton-Century-Crofts. Ludomirski A. Struijk PC. Flow velocity waveforms in the ductus venosus. Pediatr Res 1997. De Neubourg F. Wladimiroff JW. Stewart PA. Stewart PA. Wladimiroff JW. Stijnen T. and umbilical artery flow velocity waveforms between 11 and 16 weeks’ gestation. Vyas S. Doppler measurements of diastolic filling and systolic ejection times in the human fetus. Adrenal artery velocity waveforms in the appropriate and small- 114. Pipkin FB. Warren PS. ed. Clin Cardiol 1984. de Ridder MA. Rizzo G. 124. Fetal and Neonatal Cardiology. 105. Br J Obstet Gynaecol 1989. Ultrasound Obstet Gynecol 1996. Long W. Anderson CF. Br J Obstet Gynaecol 1986. Evans AT. Vetter K. Heymann MA. 14: 250–5. Developmental pulmonary circulatory physiology. Sevillia J. Hop WC. Ramsay MM. Reed K. Blood flow velocity waveforms from peripheral pulmonary arteries in normally grown and growth-retarded fetuses. Capponi A. Changes in placental blood flow in the normal human fetus with gestational age. 107. Circulation 1976. Kilavuz O. Assessment of umbilical arterial and venous flow using color Doppler. Cohen-Overbeek TE. Rotstein Z. 1990: 76–96. 8: 87–92. Ultrasound Obstet Gynecol 1996. Rizzo G. In: Copel JA. Tonge HM. Mari G. heart. Bhatia SJ et al. Rhee E. eds. Lipitz S. 99. Br J Obstet Gynaecol 1998. Human fetal pulmonary velocimetry. Abuhamad AZ. 8: 82–6. Groenenberg IA. Obstet Gynecol 1992. Longitudinal evaluation of uteroplacental and umbilical blood flow changes in normal early pregnancy. den Ouden M. Changes of pulsatility index from fetal vessels preceding the onset of late decelerations in growth-retarded fetuses. Role of the pulmonary circulation in the distribution of human fetal cardiac output during the second half of pregnancy. Stijnen T. Middle cerebral artery flow velocity waveforms in fetal hypoxaemia. Philadelphia: WB Saunders. 28: 383–7. Am J Obstet Gynecol 1989. New Orleans. 102. Mari G. Huisman TW. Huisman TW. 111. Am J Obstet Gynecol 1991. Rubin PC. 130. 1985: 87–97. Vyas S. Rasanen J. Nicolaides KH. Theard MA. Copel A. 123. Sivan E. 104. Wladimiroff JW. J Matern Fetal Med 1998. Brackley KJ. Romanini C. Wladimiroff JW. Doppler velocimetry in branch pulmonary arteries of normal human fetuses during the second half of gestation. Wladimiroff JW. Hislop A. umbilical . Buytaert P. Inferior vena cava flow velocity waveforms in appropriate. 7: 120–3. Nicolaides KH. 7: 323–9. Wladimiroff JW. Circulation 1996. Am J Obstet Gynecol 1992. Doppler measurements of umbilical blood flow. Huhta JC. Sutton MS. 93: 471–5. 172: 820–5. Rizzo G. Laudy JA. Fetal branch pulmonary arterial vascular impedance during the second half of pregnancy. 94: 1068–73. Repeatability and normal values with emphasis on middle and distal pulmonary vessels. 120. Stewart PA. Levin DL. Segmentary fetal branch pulmonary artery blood flow velocimetry: in utero Doppler study. 113. Rudolph AM. Coppens M. 100. Blood flow measurements in the fetal descending aorta: technique and clinics. New York: Raven Press. de Ridder MAJ. Gill RW. 128. Arduini D. In: Long W. Bogdan D. Laudy JAM. Presented at AIUM Annual Meeting. Huisman TW. Kollen M. 1994: 125–32. 161: 168–72. Wood DC. Tonge HM. Ludomirski A. Weiner S. Distal pulmonary artery branches in the fetus: new observations with color flow and pulsed Doppler. Albaiges G. 113: 35–48. Stewart PA. 166: 921–4. DeVore G et al. Chaoui R et al. Campbell S. Ultrasound Obstet Gynecol 1999. Ultrasound Obstet Gynecol 2000. James AE. 122. Shenker L. 106. 165: 143–51. Umbilical vein pulsations: a physiologic finding in early gestation. Wladimiroff JW. Am J Obstet Gynecol 1992. Reed KL. Rasanen J. 15: 479–86. Renal artery flowvelocity waveforms in normal and hypoxemic fetuses. Peripheral fetal circulation. Parra M. Heymann MA. Campbell S. Deane C. Am J Obstet Gynecol 1996.and small-forgestational-age fetuses. 118. Abuhamad AZ. 26: 289–99. 101. 119. Romanini C. 174: 1441–9. Achiron R. Wood DC. 115. 98. Loquet P. 117. 7: 114–21. A longitudinal study of umbilical artery Doppler waveforms in normal pregnancy: analysis using Laplace transform techniques. 121. 97: 716–19. 129. and adrenal glands? J Perinat Med 1999. for-gestational-age fetus. Detti L. Obstet Gynecol 1991. Arterial blood flow velocity waveforms of the pelvis and lower extremities in normal and growth-retarded fetuses. eds. Circulatory changes during growth in the fetal lamb. Pediatr Res 1990. 105: 78–82. transtricuspid. Uterine and fetal umbilical artery flow velocity waveforms in normal first trimester pregnancies.170 Fetal Cardiology 96. Cerebral Doppler ultrasound of the human fetus. Ultrasound Obstet Gynecol 2000. 112. Mari G. Morphological development of the pulmonary vascular bed in fetal lambs. Bower S. Doppler flow velocimetry of the splenic artery in the human fetus: is it a marker of chronic hypoxia? Am J Obstet Gynecol 1995. In: Sanders JC. 108. Nicolaides KH. Mari G. 166: 1271–80. Intra-pulmonary arterial development during fetal life – branching pattern and structure. Rudolph AM. Phibbs RH. 97: 797–803. Arduini D. Am J Obstet Gynecol 1992. Arduini D. Wladimiroff JW. Mari G. Uerpairojkit B. Br J Obstet Gynaecol 1990. The Principles and Practice of Ultrasonography in Obstetrics and Gynecology. 110. Reid L. 27: 103–6. 109.

Morton MJ. 137. Hong Y. Fetal and neonatal pulmonary circulation. Choi J.Development of fetal cardiac and extracardiac Doppler flows 171 131. Gembruch U. Huisman TW. Gudmundsson S. Raichlen JS. J Am Coll Cardiol 1991. Am J Physiol 1983. Wladimiroff JW. Stewart PA. Tonge HM. 6: 277–81. Sutton MGS. Gembruch U. Quantitative assessment of right and left ventricular growth in the human fetal heart: a pathoanatomic study. 7: 411–15. Fetal adrenal and middle cerebral artery Doppler velocimetry in high-risk pregnancy. Pediatr Res 1987. 93: 471–5. 141. vein and inferior vena cava in normal human fetuses at 12–15 weeks of gestation. 133. Circulation 1984. Doppler ultrasound assessment of cerebral blood flow in the human fetus. 135. de Ridder MA. Fetal cardiac flow velocities in the late 1st trimester of pregnancy: a transvaginal Doppler study. Wladimiroff JW. 16: 413–18. 70: 935–41. 17: 1357–9. Br J Obstet Gynaecol 1986. 134. Demonstration of fetal coronary blood flow by Doppler ultrasound in relation to arterial and venous flow velocity waveforms and perinatal outcome – the ‘heart-sparing effect’. Diedrich K. 132. 138. Ultrasound Obstet Gynecol 2000. Annu Rev Physiol 1979. 139. 244: H656–63. Rudolph AM. 13: 667–76. 16: 425–31. Blood flow in pulmonary veins: I. Stewart PA. Huisman TW. Doppler study on pulmonary venous flow in the human fetus. 143. Reichek N. van Splunder IP. 9: 162–72. Stallard T. Reiss I. 136. 22: 557–66. Wladimiroff JW. Rajagopalan B. . 41: 383–95. Laudy JA. Breborowicz GH. Cardiac functional changes in the human fetus in the late first and early second trimesters. Fetal Diagn Ther 1999. Ultrasound Med Biol 1993. Teitel DF. Ultrasound Obstet Gynecol 2000. Marsal K. Cardiovasc Res 1979. Baschat AA. Coronary artery blood flow visualization signifies hemodynamic deterioration in growth-restricted fetuses. Effects of birthrelated events on central blood flow patterns. Huff DS. 142. Ultrasound Obstet Gynecol 1995. Lee GD. 140. Dubiel M. Studies in dog and man. 19: 441–5. Iwamoto HS. Ultrasound Obstet Gynecol 1997. Friend JA. Thornburg KL. Rudolph AM. 14: 86–91. Normal fetal pulmonary venous blood flow velocity. Gortner L. Baschat AA. Gortner L et al. Filling and arterial pressures as determinants of RV stroke volume in the sheep fetus. Ultrasound Obstet Gynecol 1996. Wladimiroff JW.

.

either slightly modifying the pediatric cardiologic planes or creating ‘new’ fetal planes with more flexibility. On the right side. as well as to the late calcification of the ribs. During this movement the connection of the inferior vena cava with the right atrium is checked (venoatrial connection). This chapter concentrates on planes we proposed some years ago3 (Figure 12. atrioventricular valves. Different approaches for such an examination were proposed in the past. reducing the disadvantage of the different fetal positions on fetal examination (Figure 12.1).3 and Video clip 12. Owing to the fluid-filled lungs in prenatal life. By tilting the transducer slightly.g. The umbilical vein enters the liver and continues to the right side into the portal sinus. the rhythm. a detailed evaluation of the different cardiac structures is no longer the monopoly of a few centers but is available at the screening level. and the contractility of the fetal heart – can all be analyzed (Table 12. i.e. and can be easily learned. which is located between the stomach and diaphragm. the liver and the inferior vena cava are found. pulsed Doppler. .4). ventricles.2). The upper abdomen should then be visualized in a cross-sectional plane (Figure 12. Therefore.2). in order to distinguish the right and left sides of the fetus. the descending aorta. color Doppler. A fictitious anterior–posterior line is drawn. This plane is important since it has an accuracy in detecting a wide range of fetal heart defects. the four-chamber view (Figure 12. and the different methods based on Doppler (e. the M-mode examination. They include different approaches such as two-dimensional examination by the transabdominal or transvaginal route (in early pregnancy). Heart malformations associated with abnormalities of the situs are known to be more severe and complex.5). Furthermore. The transducer is then moved slightly cranially to visualize the next plane. The basis of every fetal echocardiographic examination is still the precise assessment of heart structures using real-time ultrasound. In this chapter. and interventricular and interatrial septa is achieved (Figure 12. the confluence of the three liver veins toward the inferior vena cava are seen. the simultaneous visualization of both atria. The examiner begins with the assessment of the fetal position in utero. the size. In this slightly oblique plane.1). The correct Examination of the upper abdomen The assessment of the upper abdomen is an integral part of the fetal echocardiographic examination (Table 12. tissue Doppler). Four-chamber view The four-chamber view is the most important plane in the examination of the fetal heart. different insonation views of the four-chamber plane can be obtained. the inferior vena cava lying anterior to the aorta. The main cardiac structures – the position.1–4 The advantage of the latter is that they are easier to learn in obstetric ultrasound. However. a careful analysis of the upper abdomen provides a better orientation when examining the fetal heart. not only at centers with high-level ultrasound machines but also at primary and secondary institutions with equipment available for performing antenatal screening. In recent years a huge improvement has been observed in image resolution.12 The examination of the normal fetal heart using two-dimensional echocardiography Rabih Chaoui Ultrasound techniques used in fetal cardiology are numerous. On the left side is the stomach. the basics of a complete cardiac examination using different planes in two-dimensional ultrasound are reviewed.1). power Doppler. the accuracy of screening for heart defects in the fetus has not increased in pace with this technical development. dividing a left and a right side. and the spleen.

174 Fetal Cardiology LV VCS WS Ao DA Apd Aps RA 1 LA RV RV 3 Ao 2 PV RA LV 1 VCI RV Ao 2 3 TP (a) Figure 12. Table 12. right ventricle. In recent years.13 and 12.1 plane • • • • Checklist for the upper abdominal Filled stomach is on the left side Aorta is on the left side of the spine Liver is on the right side Inferior vena cava is on the right side of the spine. VCS. In unclear cases. left pulmonary artery. Apd. pulmonary trunk. TP. RV. an imaginary anteroposterior line is drawn. with the heart axis pointing to the left.9). superior vena cava. and cardiothoracic ratio.3.6 The analysis of heart contractility enables the detection of hypokinesia of the myocardium. RA. the assessment of the cardiac axis was added as a new parameter in the analysis of heart position. Once the four-chamber plane is visualized. the examiner should use cardiac measurements derived in the fourchamber view.1 5 TP WS Aod Aoa VCS 4 VCS DA Acd 4 Acd TP Ao TP PV RA LV RV DA Ao VCS 5 VCI (b) (a) The different cross-sections from the four-chamber view to the five-chamber and pulmonary views (compare with Figures 12. Ao. Aoa. An abnormal heart rhythm is easily detected with realtime sonography. generally due to atrioventricular valve insufficiency associated with right atrium dilatation. and 12. one-third of the heart is on the right side and two-thirds are on the left. but its classification is effectively performed using M-mode. measuring the heart length. Aps. especially those involving the great vessels. Aod. aorta. width. After localization of the heart’s left and right sides.2. the cardiac axis is at 45 ± 15° and is abnormal in many heart defects. WS. the cavities are then compared to each other. ductus arteriosus. ventral and lateral to the aorta • Inferior vena cava receives the hepatic veins and is connected to the right atrium plane should include both patent atrioventricular valves connecting the atria with the corresponding ventricles (Figure 12. one would continue to inspect the heart structures. as well as the atria and ventricles.14). VCI. descending aorta. (b) A cross-section of the upper thorax to obtain the three-vessel view (compare with Figures 12. Analysis of the heart size is important in order to distinguish between cardiomegaly. ascending aorta. After the general examination of the heart.5 Compared to the sagittal axis. dividing the thorax into two equal left and right sides. PV. spine. 12. LV. right pulmonary artery. From reference 3. pulmonary vein. DA. left ventricle. inferior vena cava.6). area. right atrium. and the normal heart in a small thorax in growth-restricted fetuses. In normal levocardia. Important landmarks to remember are that the lumen of .

Figure 12. the esophagus can often be recognized as an echogenic circular structure (Figure 12. the esophagus dilates and mimics a second vessel in front of the aorta. the right ventricle is slightly smaller than the left. The line divides the left and right sides.2 Cross-section of the upper abdomen. The foramen ovale ‘flap’ is the free part of the septum primum.Examination of normal fetal heart using two-dimensional echocardiography 175 Table 12. but then disappears after swallowing is ended. pulsating structure. The left atrium is thus the cardiac structure situated most posteriorly in the chest and is recognized by the connections of the pulmonary veins and the leaflet of the foramen ovale. See also corresponding Video clip 12. The first cardiac structure ventrally adjacent to the aorta and esophagus is the left atrium. Owing to the rightto-left shunt at the atrial level.4). and the foramen ovale flap bulges into the left atrium. Visualized are the stomach (ST) and the descending aorta (AO) on the left (L) and the inferior vena cava (VCI) on the right.and five-chamber views toward the pulmonary view.1. which closes during embryological development of the septum primum. Anterolaterally to the spine. Just anteriorly and close to the aorta.2 Checklist for the four-chamber view plane • • • • • • • • • Position of the heart in the thorax Cardiac axis Size of the heart Rhythm Contractility Size of the left and right atria Size of the left and right ventricles Size relationship of the left and right sides Position and function of the tricuspid and mitral valves • Continuity of the interventricular septum • Position and form of the interatrial septum and the valve of the foramen ovale • Connections of the pulmonary veins to the left atrium Figure 12. the descending aorta is recognized as a circular.3 The approach to obtain the different planes from the abdominal plane to the four. the flap bulges into the . During swallowing.

6).4 Cross-section of the thorax visualizing the apical four-chamber view. the connections of the inferior and superior venae cavae can be identified (Figure 12. In the lateral approaches the septum can be better estimated. or by tilting the transducer into a longitudinal plane. IAS). Many features can be used to differentiate the right from the left ventricle. By slightly angulating the transducer cranially and/or caudally. In front of the spine the descending aorta (Ao) is seen and in front of it the esophagus (Es. the right ventricle appears as the most anterior cardiac structure. Directly behind the sternum. Both atria are nearly equal in size. LV). mitral (MV) and tricuspid valves (TV). .e. The left ventricle is adjacent and posterior to the right ventricle. LA). and is best seen using a left-sided approach to the heart. whereas the right atrial appendage is pyramidal in shape with a broad base. This structure is semilunar. and reaches the apex of the heart. left atrium. foramen ovale. whereas the inner shape of the left ventricle is smooth. Another feature is visualization of the appendages: the left atrial appendage is finger-like and has a narrow base.5 Four-chamber view seen from the left side (left) and from the right (right). showing a wide variation in its size and shape (Figure 12. but are not identified reliably under many conditions. the following structures can be seen: right and left ventricles (RV. interventricular and interatrial septa (IVS. The lumen of the left ventricle is longer than that of the right ventricle. The right ventricle is trabeculated and the cavity is irregular. the ostium secundum). and is the most left-sided cardiac structure.7 In hypoplastic left heart syndrome. In this plane. paradoxical movements of this flap can be seen to the right side. arrow). The right atrium is on the right side of the left atrium and communicates with the latter via the foramen ovale (i.7).176 Fetal Cardiology Figure 12. The appendages can be visualized in a plane slightly cranial to the four-chamber view. right and left atria (RA. and are best recognized by the vein connections. The advantage of the fetal examination is that the four-chamber view can be visualized in different fetal positions. The right Figure 12.

ventricle shows a short lumen. as well as the bulging of the flap of the foramen ovale (FO) (spetum primum). The septum begins as a wide thickened structure at the apex of the heart. with a muscular part in the lower two-thirds. this thin membranous part is not correctly visualized by an apical approach. the heart should be examined using a lateral view. Around 20 weeks of gestation. Both ventricles are separated by the interventricular septum. Figure 12. The ventricles can also be recognized by their corresponding atrioventricular valves: the left ventricle receives the mitral valve and the right ventricle the tricuspid valve. . In these conditions. See also corresponding Video clip 12. and a membranous part at the junction with the atrioventricular and semilunar valves. Using a cine-loop. During diastole the opened valves are well recognized. the visualization of different phases of the heart cycle is easier. should also be measured by the lateral approach. The tricuspid valve inserts slightly more apically than the mitral valve on the interventricular septum. ranging between 2 and 4 mm during gestation. mainly owing to the moderator band (septomarginal trabeculum coursing from the interventricular septum to the lower free wall of the right ventricle).2.6 Apical four-chamber view in systole (left) and diastole (right). This is due to the development and anatomic structure of the septum.7 Parasagittal right-sided longitudinal view demonstrating the connection of the superior (VCS) and inferior vena cava (VCI) to the right atrium (RA). This drop-out effect sometimes leads to a falsepositive suspicion of septal defects (Figure 12. and becomes thinner as it reaches the level of the atrioventricular valves. Thickness of the septum.8).Examination of normal fetal heart using two-dimensional echocardiography 177 Figure 12. allowing better visualization of the septum.

Once the four-chamber plane is visualized.9). but some authors recommend the short-axis view (Figure 12. the size of the aortic root. the examiner tilts the transducer slightly cranially (Figures 12.g. In this plane the examiner checks (Table 12.9) and focuses attention on the center of the heart in the left ventricle. The pulmonary trunk can be visualized by further tilting of the transducer cranially (Figure 12.and five-chamber planes (Figure 12. as well as the opening movements of the aortic valve (most ventricular septal defects can be detected in this plane). and the ascending aorta.3) the continuity of the septum and aorta. the angulation of the aorta and septum. left persisting superior vena cava) Visualization of left and right ventricular outflow tracts In the next planes. At the level of the atrioventricular valves. the five-chamber view and the pulmonary view. Within the aortic root. This “drop-out” effect is due to the insonation angle parallel to the membranous septum. The aorta arises as a vessel continuing the ventricular septum but pointing slightly to the right side. The other border of the aortic wall shows a close connection to the mitral valve.3 Checklist for outflow tract assessment • Normal connection of the aorta to the left ventricle and pulmonary trunk to the right ventricle • Both vessels cross over each other • Compare caliber of pulmonary trunk (PT) and aorta (PT > aorta) • Assess opening excursion of aortic and pulmonary valves • Continuity of ventricular septum to the aortic root • Normal course and caliber of great vessels and superior vena cava in the upper thorax • Assess aortic isthmus and ductus arteriosus • Rule out atypical vessels (e. Table 12.178 Fetal Cardiology Figure 12. This plane is easy to obtain by non-cardiologists and can be obtained by successive tilting from the four.3). the arising of the aorta from the left ventricle and the pulmonary trunk from the right ventricle are visualized. Once the five-chamber . Assessment of the ventriculoarterial concordance is mandatory in analyzing heart anatomy.8 Apical four chamber in systole (left) and diastole (right). the aortic valve can be recognized as an echogenic dot. The pericardium of the heart is recognized as a slight double layer around the outer cardiac wall. a tiny amount of pericardial fluid can be seen and should not be diagnosed as an abnormal effusion.10). By angulating the transducer more laterally (right).3 and 12. where the mitral valve connects with the ventricular septum. the intact septum is better recognized.

9 “Five-chamber view” (left) with the aorta (Ao) arising from the left ventricle (LV) and “pulmonary view” (right) with the pulmonary artery arising from the right ventricle (RV). In these conditions. Whereas the three arterial branches are seen by visualizing the aortic arch. In some heart defects involving the great arteries. the examiner checks for the correct connection of the right ventricle and pulmonary trunk. the examiner focuses attention during further tilting of the transducer on the connection of the right ventricle with the descending aorta (vessel toward the spine). pulmonary trunk. the pulmonary arteries are best visualized by obtaining a short-axis view of the heart.9). continuing as the ductus arteriosus (Figure 12.Examination of normal fetal heart using two-dimensional echocardiography 179 Figure 12. the right ventricle (RV). Compare with Figure 12. The vessel then arising is the pulmonary trunk. TP. VCS. The pulmonary trunk crosses perpendicularly over the ascending aorta and then becomes the vessel on the left.1. and the bifurcation of the right and left pulmonary arteries for the pulmonary trunk. On its right side. During the tilting movement from the five-chamber view. superior vena cava. it is important to identify correctly which vessel is involved. This is achieved by visualizing the five-chamber view. and the pulmonary trunk (TP) with its bifurcation into the right (APD) and left (APS) pulmonary arteries.8 and the valve is seen as a white dot showing opening and closing movements. and then rotating the transducer . as well as the crossing of the pulmonary trunk (absent in transposition of the great arteries. view is obtained. two vessels in cross-section can be recognized: the ascending aorta and the superior vena cava. Figure 12. the aorta and pulmonary trunk are differentiated by the arising of the stem vessels for the aorta. where both vessels show a parallel course). The size of the pulmonary trunk has a slightly larger caliber compared to that of the aorta.10 Short-axis view visualizing the aorta (Ao) in the center and around it the right atrium (RA).

12). and bifurcation around it (Figure 12. in teaching fetal echocardiography to the non-cardiologist.14. In the next adjacent plane to the left. Under the ascending aorta. the superior vena cava is recognized. In this plane the aortic arch appears to emerge from the center of the heart and shows a circular shape (‘candy cane’). especially in abnormal cases. the three vessels of the pulmonary trunk – the ductus arteriosus. The first is that these planes can be obtained in only a very few fetal positions: dorsoanterior or dorsoposterior. and the right atrium. The right pulmonary artery is seen very easily in its course under the aortic root toward the right lung. The examiner can be confronted with two problems while obtaining both of these planes. Therefore.12 in order to obtain a plane from the right hip to the left shoulder. This plane is therefore called the three-vessel view.1b) which was later called the ‘three-vessel view’. Longitudinal views of the outflow tracts The longitudinal planes of the outflow tracts are visualized to assess the aortic and the ductus arteriosus arches. the continuity and form of the arches are seen. right ventricle. as well as the brachiocephalic vessels arising from the aorta to the head and upper extremity. The trachea prior to its bifurcation is recognized as a circular structure with an echogenic wall adjacent to and on the right side of the aortic isthmus and anterior to the spine. The examiner obtains a parasagittal plane slightly to the left. and superior vena cava – can be seen (Figures 12. and can thus visualize the aortic arch (Figure 12.1). a cross-section of the right pulmonary artery can be recognized.14).3). including the aortic valve and the descending aorta. Video clip 12. Under the aortic arch the right pulmonary artery is seen in cross-section. and the ductus arteriosus arch courses perpendicularly to connect with the descending aorta. The pulmonary artery (TP) arises from the anteriorly positioned right ventricle (RV) and courses toward the descending aorta. This arch has a nearly perpendicular shape and resembles a hockey stick. pulmonary trunk. The second is that both planes are very close to one another. with the aorta in the center. See also corresponding DVD clip 12. Longitudinal view of the ductus arteriosus arch.3. Figure 12.3. One can easily recognize the three vessels to the head and upper limbs. On the left and right sides of . demonstrating the ‘circle and sausage’ sign. resembling a candy cane. In front of the trachea and on the right side of the aortic arch. recognized as having a more angular shape (‘hockey stick’).13 and 12.180 Fetal Cardiology Ascending Ao 1 2 3 rpa Isthmus 5 Descending Ao Figure 12. and inexperienced examiners can easily be confused. In these planes.9 The aortic and the ductus arches are seen in a tangential cross-section and build a V-form pointing to the posterior thorax on the left side of the spine (Figure 12. visualizing the above structures in a tangential cross-sectional approach. aortic arch with aortic isthmus. The right ventricle and the pulmonary valve are seen anteriorly.9 Three-vessel view From the four-chamber plane. In this sagittal cross-section of the upper thorax.11 and Video clip 12. the longitudinal view of the pulmonary trunk with the ductus arteriosus can be recognized (Figure 12.11 Longitudinal view of the aortic arch. we proposed an easier plane3 (Figure 12.10). the transducer is moved parallel in the direction of the upper thorax.

many outflow tract disorders can easily be detected. the position of the trachea in comparison to these vessels can be used as a landmark to distinguish the correct left-sided aortic arch from an abnormal right-sided aortic arch. resulting from 22q11 delections). allowing easier detection of possible abnormalities.10 Visualization of the thymus in front of these vessels can also be proposed as a hint to detect defects associated with malformations of the CATCH-22 group (cardiac defects. the aorta (AO) and the pulmonary trunk (TP). the examiner can also recognize a structure with a less echogenic appearance. lung tissue can be recognized (Figure 12. the pulmonary trunk (TP). or the presence of a fourth vessel on the left of the pulmonary trunk. cleft palate. as can the continuity of the aortic arch. The assessment of the threevessel view has been stressed in the past few years because. which is the thymus (Figure 12. these vessels. ductus arteriosus (DA) and the superior vena cava (VCS). It was then proposed to call this plane the three-vessel–trachea view.13).Examination of normal fetal heart using two-dimensional echocardiography 181 Figure 12.13 Cross-section of the upper thorax demonstrating the three vessels: the superior vena cava (VCS). . as a left persisting superior vena cava. In front of these vessels the thymus (THYM) is clearly recognized as an echodense structure compared to the neighboring lungs (L).13). In front of these three vessels.14 Three-vessel view in the upper thorax visualizing the aorta (AO) and isthmus. abnormal facies. hypocalcemia. Using color Doppler.9 Furthermore.11 The advantage of this plane is that it allows the visualization of both arches in most positions of the fetus. the visualization of antegrade flow within both outflow tracts under normal conditions can be easily distinguished from the retrograde flow in one outflow tract when severe outflow Th TP AO ym us L DA Isthmus Sp VCS Trachea R Figure 12. The slightly larger size of the pulmonary trunk compared to the size of the aorta can be checked. with this view. thymic hypoplasia. The trachea is recognized as a circle with an echogenic wall on the right side of the two great vessels and behind the vena cava.

18).17). similarly to the details explained in this chapter. regional. Several authors represented in this book have recently contributed to publishing guidelines of the International Society of Obstetrics and Gynecology12 for performing the ‘basic’ and ‘extended basic’ cardiac examination. especially with the advent of the exciting acquisition technique related to a beating fetal heart. the aortic and ductal arches. etc. It is expected that. in the near future. Figure 12. in addition to the four-chamber view plane.15). The basic cardiac screening examination relies on the four-chamber view of the fetal heart. These can be visualized offline both in cine loop and as still images.16).14 Once a volume data set is successfully obtained it can be visualized as three orthogonal planes (Figure 12. as tomographic parallel slices (Figure 12. or rendered as a 3D volume (Figure 12. This increases the accuracy in Three. as a single plane (Figure 12. . From a 3D/4D data set. the five-chamber. cross-sectional views can be obtained at any desired orientation. Spatial and temporal image correlation (STIC) allows the acquisition of a cardiac volume with one cardiac cycle.16). when feasible. The ‘extended basic’ cardiac examination includes. and the three vessel–trachea view.15 Four-dimensional data set as a STIC volume. and international societies have published recommendations on how to examine the fetal heart during the screening ultrasound. Recommendations and guidelines for two-dimensional examination of the fetal heart In recent years several local. detecting major cardiac anomalies above those detectable in the four-chamber view plane. and depth. the visualization of both the left and right ventricular outflow tracts.182 Fetal Cardiology obstruction on this side is present. the short-and long-axis views.15. called spatiotemporal image correlation (STIC). three-dimensional (3D) and 4D facilities have been used in fetal echocardiography. Each plane of interest (‘anyplane’) can be demonstrated by scrolling through the volume as the four-chamber (Figure 12.and four-dimensional ultrasound and cardiac planes Over the last few years. this plane will be added to the fourchamber view in the routine assessment of the fetal heart. direction.13. The examiner can thus demonstrate the data set as three orthogonal planes and chooses in the cine loop the moment of interest within the cardiac cycle.

Examination of normal fetal heart using two-dimensional echocardiography 183 Figure 12. Figure 12. . The examiner can modify the interslice distance (upper left image) to get the information of interest. the volume shows all parallel planes from the upper abdomen toward the three vessels. In this example.17 From the volume data set (Figure 12.15) the examiner can demonstrate the heart with multiple parallel slices using the tomography imaging tool. here the four-chamber view in systole.16 From the volume data set (Figure 12.15) the examiner can display one single plane of interest and the time of interest within the cardiac cycle.

. Ultrasound Obstet Gynecol 1997. Ultrasound Obstet Gynecol 2006. 20: 546–52. 54: 92–7. The prenatal diagnosis of congenital heart disease—a practical approach for the fetal sonographer. The three vessels and trachea view (3VT) in fetal cardiac scanning. Ultrasound Obstet Gynecol 2002. 23: 573–6. Abuhamad A. Hoffmann J. [Ultrasound measurements of the fetal heart in the 4-chamber image plane]. 10. Abnormal three-vessel view on sonography: a clue to the diagnosis of congenital heart disease in the fetus. 22: 380–7. Bowie JD. Abuhamad A. Yagel S. Ultrasound Obstet Gynecol 2004. Heling KS. Carroll BA. Cardiac screening examination of the fetus: guidelines for performing the ‘basic’ and ‘extended basic’ cardiac scan. Heling KS et al. Chaoui R. Kalache KD. Arbel R. 1997. Spatio-temporal image correlation (STIC): new technology for evaluation of the fetal heart. J Ultrasound Med 1991. Gynakol Geburtshilfliche Rundsch 1994. 1986. New developments in fetal heart scanning: three. Philadelphia: Lippincott-Raven. Ultrasound Obstet Gynecol 2002. [in German] 7. Devore GR. The storage of digitalized volume data instead of single images.3 A sagittal view demonstrating the aortic arch with the arising three brachiocephalic vessels. Video clip 12. Heling KS. 10: 567–77. Legends for the DVD Video clip 12. Yoo SJ. Figure 12. and the three vessel view. Heling KS. Both atria and ventricles are connected by the mitral and tricuspid valve. Smith RS. Anteby EY. Raveh D. Ultrasound Obstet Gynecol 2003. Hoffmann H. Bollmann R. Comstock CH. [in German] 4. Lancaster: MTP Press. 23: 535–45. Automated multiplanar imaging: a novel approach to ultrasonography. five-chamber. Semin Fetal Neonatal Med 2005.1 A parallel sweep from the upper abdomen to the upper thorax showing the different planes as the four-chamber. 14. 2.2 Transverse view of the thorax demonstrating an apical four-chamber plane. Bollmann R. 34: 145–51. [Sonoanatomy of the fetal heart. [in German] 9. J Ultrasound Med 2004. Geburtshilfe Frauenheilkd 1994. J Clin Ultrasound 1985. 15. 10: 423–6.2 in fetal cardiac defects. Kim ES et al. 16. 20: 340–5. showing normal excursion. Ultrasonographic left cardiac axis deviation: a marker for fetal anomalies. Platt LD. Proposal of simple crosssectional planes for the non-cardiologists]. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatiotemporal image correlation (STIC). 5. DeVore GR. Achiron R. A Practical Guide to Fetal Echocardiography. Kachalia P. Ultraschall Klin Prax 1991. Manual of Fetal Echocardiography.and four-dimensional fetal echocardiography. Bollmann R. 13: 229–45. Chaoui R. Chaoui R. 85: 187–91. Kirk JS. 12. 9: 173–82.184 Fetal Cardiology References 1. and the usefulness of this technique in getting a second opinion or a later offline analysis. Heling KS. 27: 107–13. [Ultrasound measurements of the diameter of the aorta and pulmonary trunk of the fetus]. Chaoui R. Lee W. Heling KS. are a few of the potentials of this new promising technique.18 The volume data set allows furthermore spatial demonstration of the heart. 11. 8. In utero sonographic appearance of the atrial septum primum and septum secundum. Chaoui R. 3. here showing a spatial view of the cardiac cavities in the four-chamber plane. 6. 13. Allan LD. It can be imagined that in the future the planes presented in this chapter will be obtained either directly during the examination or by acquiring a volume data set and demonstrating offline the different planes. Lee YH. Adams DB. Sklansky MS. Falkensammer P. 6: 59–67. Obstet Gynecol 1995. International Society of Ultrasound in Obstetrics and Gynecology. Preliminary studies reported the possibility of automated examination16 of the heart starting from such a volume data set after standardizing the positions of the planes. Video clip 12. Chaoui R. Absent or hypoplastic thymus on ultrasound: a marker for deletion 22q11.

Intrauterine. The characteristic changes in structural anatomy. rather than technical obstacles.4. rather than postnatal. is the limiting factor in early imaging and detection of structural anomalies. It should always be borne in mind that the majority of cases of CHD do not cluster in families or populations. once the province of the embryologist and pathologist.2–5 Therefore. have dramatically increased our ability to visualize and examine the developing fetal heart.6 The overall risk for aneuploidy in a fetus with CHD is estimated to be 30%. In addition. but rather occur in low-risk patients.13 First and early second trimester fetal heart screening Simcha Yagel. early detection of CHD allows for the correct timing and place of delivery. the prenatal diagnosis of CHD is often considered as an indication for fetal karyotyping. high-resolution transvaginal as well as new transabdominal (TAS) probes. at the same time it is also one of the lesions most commonly missed during fetal scans. Sarah M Cohen.2–6 Major chromosomal abnormalities are found in spontaneously aborted fetuses (57%). and (3) in selected cases there is a possibility of pharmacological therapy with subsequent improvement in fetal condition. the systematic investigation of detailed normal fetal cardiac anatomy during the first and early second trimesters of pregnancy has now shifted prenatal diagnosis of cardiac defects into that period. Cases of heart anomalies detected before 15–16 weeks’ gestation by transabdominal sonography (TAS) have been reported since the 1980s. Benefits of early fetal echocardiography Certain advantages are offered by the earlier diagnosis of CHD: (1) early confirmation of normal cardiac anatomy may help to relieve the anxiety of high-risk patients. Baruch Messing. (2) there is sufficient time for fetal karyotyping and genetic counseling for parents with affected fetuses.1 The epidemiology of CHD is discussed in Chapter 8. or facilitate earlier and safer termination of pregnancy in cases where severe anomalies are detected. along with substantial improvements in magnification imaging and signal processing.7 The increasing acceptance of transvaginal sonography (TVS) and the use of high-frequency (4–9 MHz). and Reuwen Achiron Introduction Congenital heart disease (CHD) is the most common major congenital malformation. The terms ‘sonoembryology’8 and ‘embryography’9 have been coined and used in this context. and appropriate neonatal care facilities may be planned and arranged in advance. While CHD is one of the most clinically significant malformations amenable to prenatal ultrasonographic diagnosis. and may reduce early morbidity and mortality. this rate can be as high as 66%. affecting approximately 6:1000 live births. like all screening programs. Finally. Indeed. the point has been reached where embryonic development. .3 midtrimester fetuses (18%). a fetus diagnosed with CHD is at increased risk of chromosomal abnormalities. Thus. Targeted scanning for fetal heart anomalies was usually scheduled at 20–22 weeks’ gestation and performed by the conventional transabdominal approach in specialized referral centers. can now be studied in great detail. and live births affected by significant CHD (12%). transfer ensures the neonate an optimum condition for surgery.10–12 Early screening must be repeated at mid-gestation to rule out developmental CHD.5 When extracardiac malformations are present.

the sine qua non of fetal viability. when the crown–rump length (CRL) is about 3 mm.34 echocardiography had proved to be a reliable tool in the diagnosis of CHD.35 The importance of a normal fourchamber view (FCV) was emphasized by Copel et al. In our ultrasound unit we use 4–9-MHz high-frequency probes.13 Formation of the septae and arterial and venous connections are completed only after 8 weeks’ gestation. certain maneuvers and manipulations are frequently needed in order to obtain the best planes and views in a minimal examination time. ventricular outflow tract views have . and understanding of the anatomic landmarks of fetal cardiovascular development. and fetal position often dictates the views that can be imaged. we38 and others39. Certain views may be limited by the fixed linear axis of the ultrasonographic probe during the transvaginal examination. and that the FCV was abnormal in 96% of fetuses with structural heart defects. are essential before the assessment of anomalous development can be attempted early in the course of pregnancy. can be clearly discerned from the first days of the 6th gestational week.13 Description. whether by the transvaginal or transabdominal approach.13 The defined cardiac structures visible earliest in the first trimester are probably the mitral and tricuspid valves. in their antepartum obstetrical guidelines.15 making small structures more distinguishable.32. or comparing the two. In our experience.40 failed to reach the high sensitivities in the range of 90% reported by Copel et al. The heart gradually attains a tubular structure that resembles a triocular cavity at the end of the 10th gestational week. It is primarily the obstetric and gynecologic sonographer who is well acquainted with the TVS technique and can perform a systematic evaluation of the heart and all other fetal organs. By 56 days post-fertilization the fetal heart is formed and a four-chamber structure established. The aorta can be seen at the end of the 9th week.37 Although the FCV was advocated as the preferred screening view for detection of CHD. The size of the embryonic heart has also been correlated with that of the CRL and abdominal diameter. more than 90% of CHD could be ruled out. recommended inclusion of the FCV in routine prenatal anomaly scanning programs.24 Our experience with early cardiac visualization has been favorable: full cardiac anatomy is discernible in 95% of cases at 11–12 weeks. or both. abnormalities may be either missed or suspected where they do not exist.14–24 The cardiovascular system begins to mature during the third developmental week.14. Linear correlation was found between gestational age and measurements of the right and left ventricles.33 The extended echocardiographic examination Already by the early 1980s.24–31 Two factors critically influence the accuracy of early fetal cardiac imaging: the machine and the operator. and its role as such is well established. and by the end of the 12th week the brachiocephalic and carotid arteries are readily seen in every fetus. The advantage of these highfrequency transducers is the ability to provide a clearer image because of greater axial and lateral resolution. knowledge. It has been shown that each atrioventricular valve can be individually visualized by 10 weeks’ gestation. Consequently. and their normal offset position19 (with the tricuspid valve more apical than the mitral) as well as the three vessels and trachea. Shapiro et al17 performed transvaginal measurements of the fetal heart from 14 weeks’ gestation to term. and the ductus arteriosus.36 who showed that if TAS revealed normal views. which provide a higher resolution than the 5-MHz probes that are widely used. the American Institute of Ultrasound in Medicine (AIUM) and the American College of Radiology (ACR).35 Therefore. and in many cases the echocardiographic examination requires less than 5 minutes of the scanner’s time with a favorable fetal lie. the average duration of fetal examination is approximately 30 minutes. CHD was confirmed in over 80% of cases.24 While complete echocardiography including all scanning parameters cannot be reliably performed at 10 gestational weeks. When patients were referred with a suspected diagnosis of FCV abnormalities. Without this knowledge.17 Since the 1990s many groups have examined the feasibility of early transvaginal fetal heart scanning. and the first beats of the embryonic heart. and incorporation of the four-chamber view of the fetal heart into screening made possible the detection of 60% of severe cardiac anomalies. ventricular outflow tracts. Several investigators have described normal early fetal heart anatomy. with recognition of our ability to adequately visualize normal cardiac anatomy and appreciation of our limitations. and in 100% at 13–15 weeks’ gestation with favorable fetal lie. The ratio between the fetal heart and the transverse diameter of the chest was almost constant at that period. aortic arch.186 Fetal Cardiology Normal early anatomy and its visualization Organogenesis occurs over a relatively short period: by the end of the first trimester organogenesis is complete and all the major organ systems can be imaged by ultrasound.24 The superior and inferior venae cavae can be reliably visualized from 11 to 15 weeks’ gestation and the pulmonary veins by 12–14 weeks’ gestation. it is possible in 90% of cases by 12–14 weeks and in 100% by 15 weeks. the technique of TVS requires a substantial amount of experience. With regard to the operator.36 Prompted by the limitations discovered in the FCV approach.

45 the three-vessel and trachea (3VT) plane of insonation as the fifth short-axis view in fetal echocardiography. all the pertinent views are readily visualized. beginning with the caudal upperabdomen view. should be visualized. and position. umbilical veins. we published44. the aortic arch and trachea were not included. vitelline veins. and additionally the examination should demonstrate normal outflow tracts: two great vessels approximately equal in size crossing each other at right angles from their origins as they exit from their respective ventricular chambers. Yoo et al42. and its application to prenatal echocardiographic diagnosis of congenital heart defects.44. are visualized (Figure 13.45 We consider one of the greatest advantages of our novel approach to be the ease with which the examiner can scan the fetal heart. It was shown to shorten the time required to examine the aortic arch. and subsequently performed an elective abortion . Both atrioventricular valves should be observed to open and move freely. moving cephalad. the next is the traditional four-chamber view. It is obtained easily by moving the transducer cephalad and slightly oblique from the FCV. Cross-sections of the superior vena cava (SVC).52. without observed pericardial effusion or hypertrophy.54 The fetal venous system Examination of the fetal venous system is important for comprehensive understanding of the fetal cardiovascular system. and the heart occupying one-third of the thorax. the 3VT reveals the main pulmonary trunk (MPA) in direct communication with the ductus arteriosus (DA). It is based on the familiar segmental approach to cardiac evaluation. simplified and streamlined fetal cardiac examination. further evaluation is performed in collaboration with the pediatric cardiologist and management team. The venous system develops from three paired veins. At the end of the first trimester. to complement the four traditional planes currently in use. The 3VT is the most cephalad transverse view. A transverse section of the aortic arch is seen to the right of the MPA and DA. axis. and posterior to it the trachea. Atria approximately equal in size with the foramen ovale flap in the left atrium. When properly executed. the most time-consuming aspect of fetal echocardiography.45–51 Applying this methodology allows the examiner to obtain all the views necessary for the extended fetal cardiac examination as recommended by the ISUOG guidelines. using the transabdominal approach. a crosssectional view of the major vessels. Four cardiac chambers.41 Whenever a cardiac abnormality is suspected or detected. should be observed. with the tricuspid valve leaflet inserting on the septum closer to the cardiac apex as compared to the mitral valve. which remains the key to the fetal echocardiographic examination. We routinely schedule a second echocardiographic examination. including the 3VT. without compromising diagnostic effectiveness.1). as well as the moderator band at the right ventricular apex. visualized on a plane crossing the fetal upper mediastinum. we suggest a fetal echocardiographic examination based on five transverse planes. at 20–22 weeks’ gestation for all our patients. The third is the plane commonly termed the five-chamber view in which the aortic root is visualized. several abnormal connections of the fetal venous system are recognized and are classified as follows: pathologies of the cardinal vein.38. The examination should include general observation of normal cardiac situs. Gembruch et al55 used a 5-MHz transvaginal probe to diagnose complete atrioventricular canal defect and complete heart block at 11 weeks’ gestation. The first and most caudal plane is a transverse view of the upper abdomen. and the fourth transverse view reveals the bifurcation of the pulmonary arteries. In our center we demonstrated that five shortaxis views.43 suggested the three-vessel view of the fetal upper mediastinum.First and early second trimester fetal heart screening 187 been incorporated into screening programs since the early 1990s.1). as a quick and accurate method to identify the great vessels and diagnose certain anomalies. with the majority of the heart in the left chest. should be seen.45–51 In 99% of patients this view was quickly and easily obtained from the familiar FCV. To improve and simplify comprehensive fetal heart examination.39 The International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) has published guidelines41 for the performance of fetal echocardiography.53 This approach. Following Yoo’s study. However. the former being perhaps the most demanding aspect of fetal heart examination. By sliding the transducer cephalad in one continuous motion. The ventricular septum should be examined apex to crux and shown to be intact. and atrial septum primum present.45 The clinical applicability of the 3VT has been demonstrated. and anomalous pulmonary venous connections (for detailed description see Chapter 29). The ‘extended basic’ fetal echocardiography examination proposed in the ISUOG guidelines includes all the above. Diagnostic early echocardiography The first reports of diagnostic transvaginal fetal echocardiography date back to 1990. while facilitating and expediting thorough fetal heart examination (Figure 13. Ventricles of approximately equal size without cardiac wall hypertrophy. was recently reviewed.

showing the pulmonary trunk (P). (III) The ‘five-chamber’ view. RPA) and cross-sections of the ascending and descending aorta (AO and AO with arrow. superior vena cava (SVC). (V) The three-vessel and trachea (3VT) plane of insonation. RV) and atria (LA. RT right). LV) and atria (RA. RA) and a cross-section of the descending aorta (AO with arrow). and the trachea (T) (reproduced with permission from reference 44). distal aorta ((D)Ao). left. ductus arteriosus (DA). foramen ovale (FO). proximal aorta ((P)Ao). (II) The four-chamber view of the fetal heart. showing the fetal stomach (ST). respectively). (IV) The slightly more cephalad view showing the main pulmonary artery (MPA) and the bifurcation of left and right pulmonary arteries (LPA. Polygons show the angle of the transducer and are assigned to the relevant gray-scale images (LT.188 Fetal Cardiology LT P (P)Ao RT DA V (D)Ao T SVC LT RT IV LPA MPA AO RA RPA SVC AO T DA AO PA LT RT LV RV RA LA AO III LT RT II PV LV RV RA LA AO FO PV I LT RT ST LI AO SP Figure 13. and stomach with the five planes of insonation superimposed. showing the right and left ventricles (RV. showing the aortic root (AO). liver. (I) The most caudal plane. The color image shows the trachea. and liver (LI). cross-section of the abdominal aorta (AO). spine (SP). heart and great vessels. left and right ventricles (LV.1 The five short-axis views for optimal fetal heart screening. and pulmonary veins (PV) to the right and left of the aorta (AO). LA). .

56 Gembruch et al55 and our group58 reported experiences with transvaginal echocardiography in both lowand high-risk patients.24–32. Arrow indicates canal defect. both at 14 weeks’ gestation. From the series and case reports that followed over the next several years. The same group later reported a series of 10 fetuses with CHD diagnosed by the same technique at 12–16 weeks’ gestation.61. 13 gestational weeks. including 12 793 patients over a 5-year period. left atrium.57 A further series of eight fetuses with cardiac abnormalities detected between 10 and 12 weeks’ gestation was reported by our group.First and early second trimester fetal heart screening 189 (a) (b) (c) Figure 13. (c) Tetralogy of Fallot at 14 weeks’ gestation. Note the overriding aorta (Ao) in this ‘five-chamber view’ image. Note the differential sizes of the four heart chambers. left ventricle.2. on the basis of this information. Over the ensuing years since these seminal studies. right atrium.58 A few examples of heart defects detected during the first and early second trimesters are shown in Figure 13. LV. and an isolated VSD with pericardial effusion.5-MHz probe to diagnose a ventricular septal detect (VSD) associated with an overriding aorta.2 Congenital heart defects detected in the first and early second trimesters. RA.55–60 we learned that transvaginal echocardiography provides accurate diagnosis of practically all classes and types of CHD. reaching high sensitivities and specificities.56.62 . (b) Ebstein’s anomaly diagnosed at 13 weeks’ gestation. Bronshtein et al56 used a high-frequency 6. right ventricle. in the four-chamber view plane. early fetal echocardiography has been rigorously evaluated in large studies (Table 13. RV. (a) Atrioventricular canal defect in gray-scale. LA. originated in Israel and was reported by Bronshtein et al. One of the largest series.1).

and high definition power Doppler (Figure 13. Various post-processing modalities and their application to fetal echocardiography have also been investigated67–73 and are all more fully described in Chapter 15. 3D power Doppler (3DPD). and the post-processing tools tomographic ultrasound imaging (TUI) and inversion mode (IM). D. color flow mapping. Ao.64–66 Briefly. Another advantage of 3D/4D modalities is the digital archiving and sharing capabilities designed into these systems. Of particular note is the spatiotemporal image correlation (STIC) modality (see Chapter 14).45 Planes not accessible to direct scanning with conventional 2DUS can now be obtained and evaluated. and should preferably be performed with the fetus in a quiet state.5) 11–13+6. GVd.). 3D/4DUS modalities are also applicable to screening fetal echocardiography examinations. congenital heart disease.2. 92 0.9 0. and are acquired in a single sweep. making STIC an excellent tool for early fetal echocardiography (see Chapters 14 and 15 for details). col. a wedge-shaped scanning volume is acquired from the fetal upper abdomen moving cephalad through the fetal thorax by tilting the transducer through approximately 15–25°.3). it is stored to the system database and can be transferred by internet link to any connected computer for analysis. such as the en face view of the interventricular septum and the coronal atrioventricular valves plane. Within a well-executed volume. 87 84. 22 11–13+6.5 to 15 seconds. four-chamber view.68 Other examples of 3D/4DUS modalities applicable to fetal echocardiography screening examinations are the acquisition tools B-flow.67.63 The various techniques of volume acquisition and analysis are described in detail in this volume in Chapters 14 and 15. showing the ‘surgical plane’ through the level of the heart valves (see Video clip 13. The three-dimensional revolution in fetal echocardiography Three. the volume contains the complete cardiac cycle (with or without color Doppler) and is available for post-processing analysis.5 7.44. AVV.2) 11–16 (13. RLVIT. This will introduce artifacts that compromise scan quality.2 70 63. Once the volume is acquired. right/left ventricular inflow tracts.and four-dimensional ultrasound (3D/4DUS) have revolutionized fetal echocardiography.4 12. pulmonary arteries.85 0. right/left ventricular outflow tracts. Aortic root. The five planes of fetal echocardiography are contained within this volume. Once acquired.1 of early STIC examination.2 Number of patients 12 793 660 5967 22 050 330 160 2165 3094 Comas Gabriel/2002 Segmental approach + (high-risk)30 D + col McAuliffe/2005 (high-risk)26 Smrcek/2006 (mixed-risk)24 Becker/2006 (medium-risk)25 FCV + RLVOT + AVV Five planes + D + col FCV + RLVOT + RLVIT + col + VW + AVd + GVd CHD. AoR. Therefore. all the planes necessary for complete fetal cardiac scanning are available for evaluation. AVd. FCV.76 14. great vessel diameters. Doppler.190 Fetal Cardiology Table 13.6 1. atrioventricular valve. The application of 3D/4D to diagnosis and evaluation of the fetus with congenital heart disease as it applies to specific groups of lesions appears in Chapter 15. RLVOT. younger fetal age shortens the examination time. Slower scanning speed will provide higher resolution. VW. but the fetus will have more opportunity to move or breathe. atrioventricular valve diameters. 22 77 50 78 64–85 79. aorta. Acquisition speed varies from 7. ventricular width.36 0.1 Studies of early screening echocardiographic examination Author/year Bronshtein/1993 (mixed-risk)56 Achiron/1994 (low-risk)32 Kirk/199461 Yagel/1997 62 Echocardiographic views employed FCV + RLVOT FCV + RLVOT FCV + AoR FCV + RLVOT + PA + Ao Gestational age at Sensitivity (%) Population prevalence examination (weeks) of CHD (%) 12–16 13–15 14+ 13–022 12–17 (14. PA. This opens unlimited possibilities for onsite and offsite multidisciplinary .

right aortic arch. DA. four-dimensional ultrasound (4DUS) B-flow image. LCA. Abnormal rates may aid not only in raising suspicion of CHD. Perhaps most important for screening echocardiography. superior vena cava. left carotid artery. quality review of screening programs. NT has been shown to be an effective tool not only in the identification of fetuses at high risk for chromosomal anomalies. LSVC. Arrow indicates Kommerell’s diverticulum.3 (a) Right aortic arch defect diagnosed at 15 weeks’ gestation. this allows outlying or poorly served areas to be reached more effectively. left superior vena cava. 4D high-definition color flow Doppler image.First and early second trimester fetal heart screening 191 (a) (b) (c) Figure 13. RA. manifested by hydrothorax and ascites (Figure 13. MPA. but also in predicting the fetuses possibly destined to undergo spontaneous abortion. creating a large new patient population and concomitant demand80–85 (for details see Chapter 42). the diagnosis of fetal tachycardia or arrhythmia early in pregnancy provides the opportunity for medical treatment for patients who desire to continue the pregnancy.76–79 Significant fetal tachycardia (above the upper 95% confidence interval) often induces cardiac decompensation.76–79 Natural course and in utero development of congenital heart disease In terms of the aforementioned. RSCA. since volumes acquired ‘in the field’ by local practitioners can be analyzed offsite by fetal cardiology specialists. SVC. main pulmonary artery.74.4). aberrant right subclavian artery. consultations. most particularly fetal echocardiography. In addition. with or without associated anomaly. right carotid artery. but also those at high risk for CHD. Extreme deviations of the HR from the norm convey a poor prognosis. RCA.76 Nomograms for embryonic HR are helpful in the diagnosis of bradycardia or tachycardia in the first trimester. (c) The 3VT plane of insonation.80–85 NT screening will refer 3–5% of patients for comprehensive fetal echocardiography. Embryonic and fetal heart rate Quantitative evaluation of embryonic and fetal heart rate (HR) in cases of suspected CHD is both feasible and important. which signal the need for a detailed anatomic evaluation. Branching of the vessels from the right aortic arch. ductus arteriosus.75 Nuchal translucency and the early detection of congenital heart disease The increasing acceptance of nuchal translucency (NT) measurement screening among low-risk patients will necessarily affect fetal organ screening programs. and teaching. (b) The same patient in posterior view. delayed or even missed diagnoses in some cardiac malformations occur despite detailed echocardiographic examination by experienced .

tetralogy of Fallot (n = 1). Overall. 80 malformations (78%) were detected at the initial TAS scan at mid-trimester.5 Cumulative percentage of congenital heart defects (CHD) identified at transvaginal (TVS). with resolution of hydrops. and third trimester scans and postnatally (reproduced with permission from reference 62).62. Our group undertook a study that attempted to further characterize the development of cardiac defects in utero.4 (a) Supraventricular tachycardia (SVT) to 292 beats/min. and 10 malformations (15%) were detected postnatally. some CHD may develop in utero. and an additional seven (7%) were found in the third trimester. 100 Group A Group B 90 80 70 60 TVS at 13–16 weeks TAS at 20–22 weeks Third trimester Postnatal Figure 13. ventricular septal defect (n = 1). for an overall rate of 7. and to evaluate the effects of this newly recognized entity on the accuracy of prenatal diagnosis of CHD.86–95 Thus. signs of cardiac failure included increased scalp edema (reproduced with permission from reference 79). which Cumulative % of CHD . The fetus responded well to transplacental treatment of the SVT. In group B. transabdominal (TAS).62 Possible causes for delayed diagnosis may be classified into three major categories: limited resolution. Three additional anomalies (4%) were found during the third trimester. and hypertrophic cardiomyopathy (n = 1). cardiac rhabdomyoma (n = 2). whereas 15 (15%) were diagnosed postnatally. In group A (Figure 13.192 Fetal Cardiology (a) (b) Figure 13. diagnosed at 13 weeks’ gestation and associated with severe fetal hydrops. Therefore. aortic coarctation (n = 1). Patients were divided into two groups: 6294 who had initial TVS at 13–16 weeks’ gestation. operators.31. (b) At presentation. of a mixed high. Congenital heart disease was diagnosed in 168 babies: 66 in group A and 102 in group B. followed by TAS at 20–22 weeks. and cannot completely rule out subsequent diagnosis of structural heart disease in late gestation.62 In a retrospective study. and all newborns were examined by certified pediatricians. or even in the postnatal period.and low-risk population for CHD. Both groups were subsequently examined in the third trimester.6:1000. Evidence has accrued that progression of cardiac disease may occur and may be observed in utero with advancing gestational age. leading to a varying appearance over time. subaortic stenosis (n = 1). The 10 anomalies that were diagnosed only during the third trimester included: aortic stenosis (n = 2). a normal echocardiographic appearance of the heart at any gestational age does not always mean that subsequent development can be assumed to be normal. 42 malformations (64%) were detected at the first TVS examination. and 15 126 who had initial TAS at 20–22 weeks. we reviewed the medical records of 22 050 pregnant women and their newborns. sealed foramen ovale (n = 1). 85% of the affected children were diagnosed prenatally. and 11 (17%) were diagnosed during the subsequent TAS.5).

it must be stressed that repeated examination at mid-trimester to rule out developmental CHD is absolutely essential. Lesions that may evolve and progress in utero are of major interest. Precautions and recommendations The information outlined above clearly demonstrates that high-frequency transvaginal echocardiography provides a comprehensive assessment of the fetal heart by 13–15 weeks’ gestation. The small size of the specimens after pregnancy interruption during the first or mid-trimester can render this task difficult. in our experience. This is the time when the combination of the sufficient anatomic size of the heart and the higher resolution. i. leading to late onset of CHD (category B). However.96 In comparison. and while counseling their parents. narrowing of the outflow tract will first prompt ventricular asymmetry. By using dilatation and evacuation we found that confirmation was possible in only 62% of our cases. Legend for the DVD Video clip 13. follow-up examinations are of major importance and should be performed throughout pregnancy. moving cephalad . developing fetus. chamber hypoplasia. ascites. serious defects may develop even after mid-trimester. or both. and cause more severe hemodynamic disturbances in the small.59 This requires certain expertise and careful inspection of the products of conception. The benefits of early cardiac diagnosis may still be preserved at that period.4 This should be borne in mind while considering the management of fetuses with complex CHD. cardiac abnormalities diagnosed early in pregnancy tend to be more complex than those detected in the second half of pregnancy. and even filtering to avoid loss of a fetal heart.e. Therefore. an apparently normal appearance of the heart at that time does not exclude major CHD.First and early second trimester fetal heart screening 193 may be related to both instrumentation and fetal size and position (category A). especially in high-risk patients. Abnormal pressure gradients may result in focal hypoplasia and structural remodeling. Large studies have confirmed the feasibility and high sensitivity of early fetal echocardiography. We prefer to perform early fetal echocardiography at 13–15 weeks’ gestation.3. which can shift early screening for CHD to this gestational age. therefore providing a reliable diagnosis of the major cardiac defects at this early stage of pregnancy. because of the complexity and lethality of many of the anomalies amenable to diagnosis early in pregnancy.38 Furthermore. These forms of outflow tract lesions may not appear obvious during the first half of pregnancy. only 2–3% of the anxious patients will remain who will need a repeated scan because of inadequate visualization. and a huge generalized hygroma enveloping the entire fetal body surface. fibroelastosis. Similarly. but limited focal range of TVS. Furthermore. is expected to give best results. spontaneous miscarriage occurs frequently. The speed has been slowed to allow for annotation. Postmortem confirmation of the diagnosis is almost impossible after vacuum evacuation of the uterus. may ensue. a common feature in seven of eight fetuses in one of our previously reported series59 was the demonstration of fluid accumulation. Examples of such lesions include major vessel stenosis and ventricular outflow tract obstruction. although the heart will have completed its structural development by the end of the first trimester. physicians and patients should be aware that rarely. First is the fact that pathologic confirmation of an echocardiographic diagnosis of fetal CHD is essential if the pregnancy is terminated. irrespective of the technique used for termination.95 it has the disadvantage of being an inpatient procedure that carries considerable physical and psychological morbidity. and errors in diagnosis (category C). as was also described by Gembruch et al. However. For example. This is noteworthy. some caveats concerning early cardiac diagnosis should be emphasized. Second. and this is probably the result of a combination of limited ultrasound resolution as well as erroneous diagnosis. which measures about 7–8 mm at this gestational age. The FCV and great arteries are uniformly available for examination by 12 weeks’ gestation. considerable experience in the techniques of TVS and fetal echocardiography is obligatory before attempting early diagnosis of CHD. since only a sonographer who is well acquainted with both of these techniques can perform transvaginal echocardiography. Later. only two of 23 fetuses with CHD detected during the second trimester had such fluid accumulation. mainly because the process of arrested valve growth is not significant enough to be delineated so early by ultrasound. pleural–pericardial effusion. In addition. progression of lesions in utero.6 as are abnormal karyotypes.1 The video clip shows the MPR screen of a volume data set acquired with STIC in a 15 weeks’ fetus. Beginning from the four chamber view the atria and ventricles are seen. For example. Isolated VSDs are probably the most commonly missed CHD during prenatal sonographic scanning. While termination of pregnancy by prostaglandins permits a more gentle extraction of the embryo or fetus so that pathologic confirmation may be achieved in nearly 100% of cases. the incidence of CHD in second trimester abortions is high. Third.2. as the FCV can be imaged in 100% of cases and the extended examination can be completed in 98%. Consequently. which can predominate anatomically.

Detection of fetal structural abnormalities at the 11-14 week ultrasound scan. Achiron R. Ferencz C. Shapiro I. 17: 17–40. 20. RA: left and right atria. Ultrasound Obstet Gynecol 2000. Tr: trachea. Brizot ML. Rt: left and right. 87: 594–600. Am J Obstet Gynecol 1988. 11: 180–4. SVC: superior vena cava. 2. 25. Ultrasound Obstet Gynecol 2001. Snijders RJM. Jaeggi ET. Circulation 1985. Jones A. 1998: 349–404. then the main pulmonary artery and the ductus arteriosus. 18. Verheijen PM. Peisner DB. Evaluation of fetal heart dimensions from 12 weeks to term. RV: left and right ventricles. Strobino BA. 121: 798–803. J Ultrasound Med 2006. Abbreviations: Lt. the descending aorta. Haak MC. 165: 688–91. Early fetal echocardiography: heart biometry and visualization of cardiac structures between 10 and 15 weeks’ gestation. Becker R. Coltri A. 114: 79–86. ductus arteriosus. Wegner RD. Detailed screening for fetal anomalies and cardiac defects at the 11-13-week scan. The incidence of congenital heart disease. Geipel A et al. LPA. Accuracy and limitations of transabdominal fetal echocardiography at 12-15 weeks of gestation in a population at high risk for congenital heart disease. Trines J. Smrcek JM. Am J Obstet Gynecol 2005. Leibovitz Z et al. Comparative analysis of pattern. Shi C. 8. Significance of cardiac defects in the developing fetus: a study of spontaneous abortuses. Hoffman JI. Lancet 1992. 16. 22. References 1. Simpson JM. Ursell PC. Prenat Diagn 2002. Gabbay U et al. Fetal aortic arch measurements between 14 and 38 weeks’ gestation: in-utero ultrasonographic study. Sholler GF. A close look at embryonic development with the high frequency transvaginal transducer. Ultrasound Obstet Gynecol 1998. Bonnet D. Kiserud T. Gembruch U. Raju S.versus postnatally diagnosed major congenital heart disease: a population-based study. which shows the main pulmonary artery. 24. main pulmonary arteries. Next the three vessels and trachea plane is seen. Philadelphia: WB Saunders. Blaas HG. Detection of transposition of the great arteries in fetuses reduces neonatal morbidity and mortality. Sharland GK. Ultrasound Obstet Gynecol 1998. D’Amelio R. Martinez JM et al. Cullen M. 29. J Pediatr 1989. Steiger RM. Nicolaides KH. Chromosomal anomalies in fetal congenital heart disease. Gosden CM. Neiman UL. 72: 1232–6. Sonoembryology: an organ oriented approach using a high-frequency vaginal probe. J Am Coll Cardiol 2002. LV. Am J Obstet Gynecol 1991. Chervenak FA. Van Vugt JM. Early prenatal diagnosis of major cardiac anomalies in a high-risk population. Ultrasound Obstet Gynecol 1995. Zimand S. Larson EJ. 5. Timor-Tritsch IE. 158: 409–13. The cardiovascular system. Farine D. J Thorac Cardiovasc Surg 2001. Chita SK. Maxwell DJ. MPA: right. Fetal Diagn Ther 2000. Copel JA. In utero ultrasonographic measurements of fetal aortic and pulmonary artery diameters during the first half of gestation. Eik-Nes SK. 11: 22–33. Jones OD. Transvaginal ultrasound embryography. 3. Moore KL. 15: 226–30. Timor-Tritsch IE. 28. Butera G et al. aortic valve. 6: 240–9. Sharland GK. Byrne JK. Green JJ et al. Ultrasonographically detectable markers of fetal chromosomal abnormalities. Lisowski LA. DeVore GR. Obstet Gynecol Clin North Am 1990. stomach and heart from 7 to 12 weeks of gestation: a longitudinal ultrasound study. 9. 15: 20–31. 25: 173–82. 23: 916–18. Allan LD. 30. 22: 586–93. McAuliffe FM. Biometry of the fetal heart between 10 and 17 weeks of gestation. Early fetal echocardiography—a reliable prenatal diagnosis tool. Continuing the sweep cephalad the left and right pulmonary arteries are shown. Silverman NH. Rosen MG. 23.194 Fetal Cardiology the aortic valve is seen in the “A” plane (upper left quadrant) and the orthogonal “B” plane. Golan-Porat N. 340: 704–7. Carvalho MH. left. trachea. 11. Achiron R. Gioriandino C. 69: 494–7. 26. Campbell S. Kaplan S. 6th edn. . DA: ductus arteriosus. 14. 18: 286–98. and the azygos vein. 27. 7. Ultrasound Obstet Gynecol 2006. 27: 613–18. 1: 8–11. In: The Developing Human: Clinically Oriented Embryology. 159: 676–81. Circulation 1999. Semin Ultrasound CT MR 1990. Fetal echocardiography using transvaginal and transabdominal probes during the first period of pregnancy: a comparative study. Prenat Diagn 1991. Fetal echocardiography: the prenatal diagnosis of a ventricular septal defect in a 14 week fetus with pulmonary artery hypoplasia. FOF: foramen ovale flap. Ultrasound diagnosis of fetal anomalies. 15. 6. Ultrasound Obstet Gynecol 1991. Az: azygos vein. Congenital cardiovascular malformations associated with chromosome abnormalities: an epidemiologic study. Hegesh J et al. Lockhart S. 13. The frequency of aneuploidy in prenatally diagnosed congenital heart disease: an indication for fetal karyotyping. 17: 380–5. LA. 12. Lopes LM et al. Reimers FT. Twisk JW. Neill CA. Stoutenbeek P et al. Masala L et al. AoV. Prenatal diagnosis of congenital heart disease affects preoperative acidosis in the newborn patient. 39: 1890–900. Boughman JA et al. Berry C. the aortic arch. How successful is fetal echocardiographic examination in the first trimester of pregnancy? Ultrasound Obstet Gynecol 2002. management and outcome of pre. J Clin Ultrasound 1990. AoA: aortic arch. Nield LE et al. Prenat Diagn 2002. Berg C. Am J Obstet Gynecol 1988. Firpo C. 10. Galindo A. 19. Fetal cardiac measurements derived by transvaginal and transabdominal crosssectional echocardiography from 14 weeks of gestation to term. Cooper SG. Smrcek JM. where the foramen ovale flap is seen. Also available in post-processing is the rendered image. 22: 1–4. AV. RPA. 11: 69–75. 20: 9–13. BJOG 2000. Hellevik LR. Rottem S. Callaghan N. Persaud TVN. 17. 193: 1253–9. Am J Cardiol 2001. Degani S. Dolkart L. 107: 1492–7. superior vena cava. 21. Hoffman JI. 4. 12: 404–18. Comas Gabriel C. Transvaginal fetal echocardiography in early pregnancy: normative data. Early development of the abdominal wall. Obstet Gynecol 1987.

Zimmer EZ. Wiener Y. Ho SY. 64. Pediatrics 1980. Ultrasound Obstet Gynecol 2002. 10: 697–702. DeVore GR. Fetal cardiac abnormalities detected by transvaginal sonography at 12-16 weeks’ gestation. Achiron R. Zimmer EZ. 157: 648–55. 35. 60. 55. Bronshtein M. Yagel S. 59. Development of congenital lesions in mid or late gestation. 56: 139–43. Siegler E. Ultrasound Obstet Gynecol 2005. Carnforth. Prenatal screening for congenital heart disease. 56. 82: 225–9. 75: 496–98. Del Bianco A. Doppler assessment of fetal aortic isthmus blood flow in two different sonographic planes during the second half of gestation. Am J Obstet Gynecol 1992.First and early second trimester fetal heart screening 195 31. Detection rate of early fetal echocardiography and in utero development of congenital heart defects. Fetal ultrasonography. 43. J Perinat Med 2006. Yoo SJ. Echocardiographic studies of the human fetus: prenatal diagnosis of congenital heart disease and cardiac dysrhythmias. Prenat Diagn 1990. Yagel S. 61. Hobbins JC. 50. Simple approach to prenatal diagnosis of transposition of the great arteries. Garmel SK. Mandruzzato G. Gerlis LM. Smrcek JM. Lipitz S. 172: 825–30. 39. Arbel R. Heredia F. Chatterjee M. Ultrasound Obstet Gynecol 2001. Copel JA. Circulation 1997. Achiron R. Sequential chamber localization – logical approach to diagnosis in congenital heart disease. Ultrasound Obstet Gynecol 2007. 54. 24: 174–8. Transvaginal echocardiographic examination of the fetal heart between 13 and 15 weeks’ gestation in a low-risk population. Anderson RH. 1995: 95–120. Martinez JM. Fetal echocardiographic screening for congenital heart disease: the importance of the four-chamber view. 38: 327–40. AJR Am J Roentgenol 1999. Ultrasound Obstet Gynecol 2002. Bronshtein M. D’Afton ME. Lacerenza N et al. Glaser J. 292: 1717–19. Gamzu R et al. Bronshtein M. 20: 553–7. 54. 17: 367–9. Ultrasound Obstet Gynecol 2006. Kleinman CS. 9: 173–82. Achiron R. Rotstein Z. Weissman A. BMJ 1992. Prenatal diagnosis of ventricular septal defect and overriding aorta at 14 weeks’ gestation. Anteby EY. Obstet Gynecol 1991. Heggesh J et al. Yoo SJ. Drugan A. Falkensammer P. 32. J Ultrasound Med 2006. 49. Hegesh J. 96: 550–5. Van Praagh R. Cohen SM. Comstock CH et al. West J Med 1993. 47. 19: 361–2. Ultrasound Obstet Gynecol 2003. 84: 69–72. Kim ES et al. Macartney FJ. Extended fetal echocardiographic examination for detecting cardiac malformations in low risk pregnancies. using transvaginal sonography. Obstet Gynecol 1993. 26: 170–4. Fetal echocardiography: accuracy and limitations in a population at high and low risk for heart defects. Obstet Gynecol 1994. Zalel Y. Poblete P et al. Shinebourne EA. Zimmer EZ. In: Jurkovic D. International Society of Ultrasound in Obstetrics and Gynecology. Bald R. Young HL. 46. Valsky DV. Weissman A. Platt LD. Bussani R. Am J Obstet Gynecol 1987. Glossary and commentary. 27: 107–13. Cohen SM. Mashiach S. Vinals F. 63. 78: 374–8. Firsttrimester diagnosis of fetal congenital heart disease by transvaginal ultrasonography. 20: 340–5. Maieron A. 28: 22–5. 65: 1059–67. Anomalies of the fetal aortic arch: a novel sonographic approach to in-utero diagnosis. Sequential segmental analysis in complex fetal cardiac abnormalities: a logical approach to diagnosis. Shinebourne EA. Congenital heart defects: natural course and in utero development. 45. 34: 309–12. Sanders SP et al. Chita SK. Four chamber view plus three-vessel and trachea view for a complete evaluation of the fetal heart during the second trimester. Kirk JS. Knopfle G. Jauniaux E. Obstet Gynecol 1990. Ultrasound Obstet Gynecol 2003. eds. 26: 105–11. Hansmann M. Benettoni A. Raveh D. Kyoung SC. 166: 1473–80. UK: Parthenon Publishing. 38. Estroff JA. Three-vessel view of the fetal upper mediastinum: an easy means of detecting abnormalities of the ventricular outflow tracts and great arteries during obstetric screening. Examination of the fetal heart by five short-axis views: a proposed screening method for comprehensive cardiac evaluation. Br Med J 1986. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Pilu G. Gembruch U. 62. Int J Cardiol 1988. The three vessels and trachea view (3VT) in fetal cardiac scanning. Tynan MJ. Jaffe CC et al. Shapiro I. Ascenzo R. 34. 22: 380–7. 40. Abnormal three-vessel view on sonography: a clue to the diagnosis of congenital heart disease in the fetus. 53. Obstet Gynecol 1994. Allen LD. J Ultrasound Med 1994. Yagel S. Crawford DC. 52. Cardiac screening examination of the fetus: guidelines for performing the ‘basic’ and ‘extended basic’ cardiac scan. Carvalho JS. 5: 202–5. Prenatal screening for cardiac anomalies: the value of routine addition of the aortic root to the four-chamber view. 37. 48. Achiron R. Lorber A. . 36. Green J et al. 33. Antepartum obstetrical guidelines. 42. Figueras F et al. 58. Early ultrasound diagnosis of fetal congenital heart defects in high-risk and low-risk pregnancies. Russo S. Rustico MA. Yoffe N. Prenat Diagn 2004. The three-vessel and tracheal view of the fetal heart: an in utero sonographic evaluation. 304: 671–4. Utrasound Obstet Gynecol 1997. Yagel S. Rotstein Z. Rotstein Z et al. Hegesh J. Ultrasound diagnosis of congenital anomalies in early pregnancy. 13: 783–9. Bromley B. First trimester diagnosis of fetal congenital heart disease by transvaginal two-dimensional and Doppler echocardiography. Berg C. Gelernter I. Milo S et al. Riggs TW. Sklansky MS. 159: 273–85. Rotstein Z et al. Allan LD. Achiron R. Br Heart J 1976. 57. Achiron R. 51. Leopold G. Ultrasound Obstet Gynecol 2005. Geipel A et al. 84: 427–31. Weissman A. Lee YH. Circulation 1977. Terminology of congenital heart disease. 22: 358–67. 29: 81–95. 5: 241–2. Ultrasound and Early Pregnancy. Del Rio M. 25: 187–96. J Ultrasound Med 1986. Early fetal endocardial fibroelastosis and critical aortic stenosis: a case report. 44. Giuliano A. Ultrasound Obstet Gynecol 2006. Spatio-temporal image correlation (STIC): new technology for evaluation of the fetal heart. 41. Achiron R. Yagel S. Ultrasound Obstet Gynecol 1995. Vinals F. The role of the three vessels and trachea view (3VT) in the diagnosis of congenital heart defects.

Hansmann M. 90. Allan LD. Bonnet D et al. Lopes LM. Tadmor O. Hansmann M. Am J Obstet Gynecol 1995. Sciaky-Tamir Y. Br Med J 1999. Am J Obstet Gynecol 2006. DeCaro E. Evolution of echocardiographic findings in the fetus. Vargas G. Evolution of the aorta in intrauterine life. Vassallo M. Cardiac malformations in spontaneous abortuses. Todors T. Sciarrone A. Giuliano A. Hochner-Celnikier D et al. Structural and functional cardiac abnormalities identified prior to 16 weeks’ gestation in fetuses with increased nuchal translucency. Lee W. 78: 330–4. 25: 341–3. 68. Michailidis GD. 82. Ultrasound Obstet Gynecol 2002. Rendering in fetal cardiac scanning: the intracardiac septa and the coronal atrioventricular valve planes. 352: 343–6. First-trimester fetal heart block and increased nuchal translucency: an indication for early fetal echocardiography. 66. . 74. A new approach to the characterization of ventricular septal defects in the fetus. Am J Obstet Gynecol 2005. Am J Obstet Gynecol 2003. Mashiach S. Makrydimas G. Snijders RJM. Bald R. Van Vugt JM. 78. Sharland GK. 93. Allan LD. Gerlis LM. 69. Allan LD. Ultrasound Obstet Gynecol 1999. 77. 52: 471–3. J Ultrasound Med 2006. Bartelings MM. Lee W et al. Pulmonary stenosis with intact ventricular septum: documentation of development of the lesion echocardiographically during fetal life. 88. 91. Reller MD. Paladini D. Hyett JA. Ultrasound Obstet Gynecol 2003. 203(Suppl): 354. Left heart obstructive disease: changes in echocardiographic appearance during pregnancy. Hornberger LK. 10–14 weeks of gestation as a marker for major cardiac defects. Ultrasound Obstet Gynecol 2001. Obstet Gynecol 1991. Prenat Diagn 2005. Sklansky MS. Fetal supraventricular tachycardia diagnosed and treated at 13 weeks of gestation: a case report. Heart rate as a predictor of first-trimester spontaneous abortion after ultrasoundproven viability. Botta G et al. 72. Porat S. 20: 14–21. Knopfle G. McDonald RW. Demarie D. Gembruch U. 85. J Am Coll Cardiol 1995. Anteby EY. 79. Souka A. Transvaginal ultrasonography at early pregnancy cannot be used alone for targeted organ ultrasonographic examination in a high-risk population. Lopes MA et al. A novel algorithm for comprehensive fetal echocardiography using 4-dimensional ultrasonography and tomographic imaging. Lancet 1998. Gittenberger-De Groot AC. UK multicentre project on assessment of risk of trisomy 21 by maternal age and fetal nuchal-translucency thickness at 10-14 weeks of gestation. Ultrasound Obstet Gynecol 2003. 23: 573–6. Haak MC. Gaglioti P. 96: 391–2. Goncalves LF. Perdu M. Sahn DJ et al. 21: 65–8. 76. 67. Yagel S. Caponetto S. Examination of the fetal heart by four-dimensional (4D) ultrasound with spatio-temporal image correlation (STIC). The ‘inplane’ view of the inter-ventricular septum. Three-dimensional power Doppler (3DPD) ultrasound in the diagnosis and follow-up of fetal vascular anomalies. Crawford DC. Early diagnosis of fetal congenital heart disease by transvaginal echocardiography. Romero R. Anteby E. DeVore GR. Tynan MJ. Hamani Y. Economides DL. Perdu NL. Karidas C. Cardiac malformations in first-trimester fetuses with increased nuchal translucency: ultrasound diagnosis and postmortem morphology. 19: 355–60. Vinals F. 18: 325–8. Russo MG. Simpson JM. Espinoza J. Circulation 1997. Snijders RJM. Ribaidone D. Fourdimensional ultrasonography of the fetal heart with spatiotemporal image correlation. Tynan M. Kusanovic JP. 75. Platt LD. Sebire N. Polanco B. Brizot ML. Matitiahu A et al. Cohen SM. Goncalves LF et al. Ant J Cardiol 1988. 92. 71. Masturzo B. Abuhamad A. A new approach to fetal echocardiography: digital casts of the fetal cardiac chambers and great vessels for detection of congenital heart disease. Rice MJ. Achiron R. 25: 1129–32. Baschat AA. Progressive pulmonary stenosis in the fetus: two case reports. The natural history of the hypoplastic heart syndrome. Sotiriadis A. 172: 971–5. 25: 739–45. Am J Perinatol 1993. 318: 81–5. Achiron R. Yagel S. 3: 310–17. Ultrasound Obstet Gynecol 2006. Revel A. Nuchal translucency and fetal cardiac defects: a pooled analysis of major fetal echocardiography centers. Int J Cardiol 1985. Presbitero P. Increased nuchal translucency at 81. Lee W. 24: 415–24. Tartaglione A. 194: 274–81. Nicolaides KH. Knopfle G. Ultrasound Obstet Gynecol 1994. 14: 311–14. Sanders SP. Ultrasound Obstet Gynecol 2006. 10: 424–7. Sharland G. 189: 1792–802. Pongiglione G. 27: 336–48. Espinoza J. J Ultrasound Med 2005. First-trimester fetal heart block: a marker for cardiac anomaly. Detailed three-dimensional fetal echocardiography facilitated by an Internet link. 70. 83. Sharland GK et al. Noble P. 23: 1052–5. Prenat Diagn 2003. Espinoza J. 24: 72–82. The ‘spin’ technique: a new method for examination of the fetal outflow tracts using three-dimensional ultrasound. Goncalves LF. Ultrasound Obstet Gynecol 1997. Mandujano L. 25: 25–31. Prenatal diagnosis of congenital heart disease using four-dimensional spatio-temporal image correlation (STIC) telemedicine via an Internet link: a pilot study. In utero pulmonary artery and aortic growth and potential for progression of pulmonary outflow tract obstruction in tetralogy of Fallot. 94. Marasini M. 73. J Ultrasound Med 2004. Nicolaides KH. 84. 21: 302–5. Weissman A. Using fetal nuchal translucency to screen for major congenital cardiac defects at 10-14 weeks of gestation: population based cohort study. 87. Int J Cardiol 1989. 7: 29–35. 96. Hyett JA. 192: 89–95.196 Fetal Cardiology 65. Ultrasound Obstet Gynecol 2004. Goncalves LF. Automated multiplanar imaging: a novel approach to ultrasonography. 95. 22: 470–8. 28: 266–74. Huggon IC et al. Endocardial fibroelastosis secondary to critical aortic stenosis: natural course and evolution in utero. Achiron R. 80. Ultrasound Obstet Gynecol 2005. Gembruch U. 25: 947–56. Benachi A. Chaiworapongsa T et al. 89. Yagel S. Br Heart J 1984. J Clin Utrasound 1993. 86. 10: 242–6. Ultrasound Obstet Gynecol 1993. Ron M.

volume data sets can be examined using the standard planes of section of two-dimensional ultrasonography (2DUS).5. and cardiac rhythm. Indeed. and (3) if the examiner accidentally makes a mistake during the review of the volume data set. the ‘digital specimen’ is not damaged forever. or B-flow imaging. Technology The term 4DUS is used to describe volume data sets that incorporate information about the three spatial dimensions plus the temporal dimension. Therefore.31 In this chapter. and (4) faster heart rates than in adult and pediatric patients. and Roberto Romero Introduction Prenatal evaluation of the fetal heart is one of the most challenging components of the obstetrical ultrasound examination. Juan Pedro Kusanovic. (2) the direction of blood flow can be analyzed in volume data sets acquired with color or power Doppler.32 4DUS of the heart presents particular technical challenges. the venous return to the heart. and oligohydramnios may have a negative impact on image quality. Jimmy Espinoza. and coarctation of the aorta.7–12 Failure to diagnose a life-threatening cardiac disorder in utero may negatively impact survival. sophisticated three-dimensional rendering techniques can be applied to display ‘digital casts’ of cardiac chambers and great vessels that look very similar to postmortem casts obtained by injecting silicone rubber into cardiovascular structures.1). 4DUS of the fetal heart may help to overcome the dependency on operator skills which is characteristic of fetal echocardiography performed by conventional 2DUS.7–12. hypoplastic left heart syndrome.13–16 Four-dimensional ultrasonography (4DUS) with spatiotemporal image correlation (STIC) allows examiners to acquire volume data sets of the fetal heart using gray-scale imaging only or with the addition of blood flow information from color Doppler. abdominal scars. which includes a detailed assessment of the four-chamber view (Table 14. Moreover. The ‘digital heart’ can be oriented on the screen to be displayed in a standardized position.2 the connections of the great vessels to the ventricular chambers.14 Four-dimensional ultrasound examination of the fetal heart by spatiotemporal image correlation Luís F Gonçalves. and eventually improve the prenatal characterization of complex congenital cardiac defects. since there is evidence that prompt intervention after delivery is associated with improved outcomes in disorders such as transposition of the great arteries.1. factors such as maternal obesity. thus. as well as novel planes that are only possible by volumetric imaging.17–29 Once acquired. extensive training is required to develop the skills necessary to examine the fetal heart effectively.30 Advantages of the ‘digital specimen’ when compared to the ‘actual specimen’ include: (1) functional information is preserved since heart beats are included in the volume data set.3–6 Thus. and all that it takes to begin the examination all over again is to reset it to its original state with the click of a button. other challenges that are particular to the examination of the fetal heart include: (1) the fact that the fetus is located within a uterus inside another body (the mother) and.19. since the phases of . anterior placentas. (3) less than ideal fetal position that cannot be controlled by the operator. power Doppler. Besides the limited amount of time that is usually available for a comprehensive ultrasound examination in clinical practice. after which standard planes of section are obtained with the use of ‘digital scalpel’ tools. These volume data sets can be thought of as ‘digital specimens’ of the fetal heart. (2) frequent movement or breathing during the examination. operator skill is considered by many as the most important factor affecting prenatal diagnosis of congenital heart disease. akin to actual heart specimens that are examined by pathologists during a necropsy. we will review several techniques that can be used to examine the fetal heart by 4DUS with STIC in normal fetuses as well as in those with congenital heart disease. Wesley Lee. Sonographers must conduct a comprehensive examination.

a region of interest is selected on the screen. This image is not visible on the system during STIC acquisition. an ordered sequence of volume data sets is displayed on the screen as a continuous cine loop containing all phases of the cardiac cycle.4 The end result is a volume data set containing three-dimensional plus temporal information that allows the interactive display of cardiac structures in any plane of section as well as three-dimensional reconstruction by rendering techniques.43 Once the two-dimensional image is optimized. axis and position Heart occupies a third of thoracic area Majority of heart in left chest Four cardiac chambers present No pericardial effusion or hypertrophy Atria Atria approximately equal in size Foramen ovale flap in left atrium Atrial septum primum present Ventricles Ventricles about equal in size No cardiac wall hypertrophy Moderator band at right ventricular apex Ventricular septum intact (apex to crux) Atrioventricular valves Both atrioventricular valves open and move freely Tricuspid valve leaflet inserts on ventricular septum closer to the cardiac apex than to the mitral valve Figure 14.22. This process is repeated for all phases of the cardiac cycle. are merged into the same volume data set. and the data are ready for examination. and the fetal heart rate is calculated.26. volume acquisition may take from 5 to 15 seconds.1 Raw data image used to calculate the fetal heart rate.1). and we favor those characterized by low persistence. obese patients). gating is quickly performed immediately after the volume scan. After image rearrangement. In adults and children.27. a cardiac gating signal can be obtained by simultaneous recording of an electrocardiogram. Harmonic imaging may improve image quality in selected cases (e.33–41 STIC is a gating algorithm based on the analysis of cardiac motion. but helps to understand the technique (reproduced from reference 22 with permission from Elsevier). Depending on the manufacturer. other gating methods have been proposed to overcome this problem. while the patient is still on the scanning table. slow sweep. and the volumes can be saved on a hard drive for later review or transmission to a remote diagnosis center. Because of the long acquisition time (7. the cardiac cycle must be not only included in the final volume data set. The final quality of the volume data set is heavily dependent on prior adjustment of two-dimensional grayscale and color Doppler parameters.1 Basic cardiac screening examination.1). and volume acquisition is performed with a single automated sweep of the transducer (Video clip 14.27 Although retrospective. Since electrocardiographic signals are difficult to obtain in the fetus. but also synchronized with the acquired spatial information. and high frame rate Magnification should be performed prior to acquisition.5.40 . Beat-to-beat changes of the heart rate would appear as shortening or elongation of the above motion pattern.40 The fundamental fetal heart rate is directly extracted from the volume data set (Figure 14. This image is generated after the spatiotemporal image correlation (STIC) acquisition is performed as a single.35.22 The examination can be performed either in the presence of the patient or offline. the beating heart draws a motion pattern.23 Fetal echocardiography presets are available in most ultrasound systems. Frames acquired during the same phase of the cardiac cycle. This pattern is analyzed in terms of periodical changes of gray-scale information.42. The STIC algorithm detects the fundamental heart rate based on the rhythmic pattern of changes in cardiac diameter and uses this information for gating.5–15 seconds from the left to the right end of this image). This synchronization is termed cardiac gating.198 Fetal Cardiology Table 14. Adapted with permission of the American Institute of Ultrasound Medicine from reference 2 General Normal cardiac situs.5. This particular image is orthogonal to the original two-dimensional frame. although from a different position in space. Volume acquisition Volume acquisition is a crucial aspect of the 4DUS with STIC. high contrast. Information from these raw data is used to rearrange the two-dimensional frames.g.

Under these circumstances. (4) three-vessel view. the four-chamber view plane). if the examiner is mainly interested in obtaining images of the aortic and ductal arches. with the spine oriented at approximately the 6 o’clock position on the screen). (2) four-chamber view. Adjustments may be necessary when examining smaller or larger fetuses. will be available for the examination. and three-vessel and trachea views. ribs. ROIs should be set as narrow as possible and include only the information that the examiner is interested in. Therefore. For second trimester fetuses. acquisition time ranges from 5 to 15 seconds. Since this is not always possible. the left and right ventricular outflow tracts. When an examiner begins to use 4DUS with STIC. Vinals et al27 evaluated this approach in 100 volume data sets acquired by sonologists with little experience in examination of the fetal heart. For volumes acquired using transverse sweeps through the fetal chest. acoustic shadowing from the spine and ribs frequently compromises fetal heart imaging. and the transverse section of the fetal abdomen containing the stomach. Fetal position The ideal fetal position to examine the fetal heart is with the fetus lying on its back (i. aorta and interior vena cava (IVC). trachea view. threevessel. volume data sets of sufficient quality may also be obtained when the spine is up. the tendency is to select a large ROI. superiorly. including not only the heart. Ideally. and abdominal breathing by the mother may cause motion artifacts.27 inferiorly. (3) fivechamber view. Depending upon equipment manufacturer. Visualization rates for the four-chamber view. This results in a volume data set with lower spatial resolution. The lowest Setting the acquisition angle The acquisition angle determines the amount of information acquired in the ‘z’ or azimuth plane. in the presence of intense fetal movements or breathing. and (5) three-vessel view and trachea view (Figure 14.5 seconds and adjust the acquisition time according to the presence or absence of active fetal movement or breathing. Conversely.2 and Video clip 14.e. For acquisitions performed in the transverse plane (e. This practice maximizes both the frame rate during acquisition as well as the final temporal resolution of the volume data set. images from the upper mediastinum all the way down to the upper abdomen should be ideally included in the volume data set. acquisition angles between 25° and 30° should suffice. who was not involved in volume acquisition. five-chamber. provided that it is not positioned between 11 and 1 o’clock. the three-vessel view. Unfortunately. A specialist in fetal echocardiography. Selecting a region of interest The region of interest (ROI) determines the width and height of the volume data set (x and y planes). high-quality volume data sets are acquired with sagittal sweeps through the fetal thorax. In practice. Mothers may also move during acquisition.27 This practice ensures that all standard planes of section. including the three-vessel and . we have found it useful to ask mothers to momentarily withhold moving or breathing during acquisition. we preset the equipment to sweep for 10 or 12. Setting the acquisition time Acquisition time determines the speed that is used by the transducer to sweep through the ROI. examined the volume data sets. volume data sets should be acquired using the longest acquisition time possible to improve spatial resolution. and the three-vessel and trachea view ranged from 81 to 100%. and amniotic fluid.g. This is particularly important during color or power Doppler acquisitions. but also the surrounding anatomical structures such as the lungs. The reader is reminded that the selection of wide ROIs is associated with lower than expected frame rates during acquisition and this may negatively impact the temporal resolution of the final volume data set.4D ultrasound examination of the fetal heart by spatiotemporal image correlation 199 Tips to optimize volume acquisition23 Original plane of acquisition Volume acquisition using transverse sweeps across the fetal thorax are preferred when the examiner is interested in evaluation of the four-chamber. examiners may be forced to select a shorter acquisition time to minimize motion artifacts. this approach allows examiners to visualize the transverse planes of section proposed by Yoo et al44 and Yagel et al45: (1) transverse view of the upper abdomen. Thus. Examination of the fetal heart with STIC Scrolling through the volume data set The most basic approach to examine a volume data set of the fetal heart acquired with STIC is to scroll through the volume data set from top to bottom along the original plane of acquisition.2).

right atrium. (c) the five-chamber view.46 . IVC.2 The five transverse planes of section proposed by Yagel et al for examination of the fetal heart45: (a) the transverse view of the upper abdomen. left ventricle. trachea. and (e) the three-vessel and trachea view. LA. aorta. (e) visualization rates were observed for structures located either at the upper abdomen or upper mediastinum. PA. Ao. (d) the three-vessel view. left atrium. right ventricle. superior vena cava. T. This occurred because acquisition angles were not properly set. LV. RA. This methodology has also been used for remote diagnosis of congenital heart disease (TELE-STIC). inferior vena cava. pulmonary artery.26. SVC. RV. that is. (b) the four-chamber view.200 Fetal Cardiology (a) (b) (c) (d) Figure 14. the sweep angle was not wide enough to include information from the upper mediastinum through the upper abdomen in some cases.

Issaquah. Siemens Medical Solutions.47 With STIC. and contrast can also be adjusted to optimize image quality.3 (b) Tomographic ultrasound imaging (TUI) of a normal fetal heart in systole (a) and diastole (b).50 Hue. multiple slices of the beating heart can be visualized and examined at the same time. Philips Medical Systems. Please note that the fivechamber view was better visualized during systole. Bothell. 5CH: five-chamber (reproduced with permission from reference 49). since motion information is preserved. The plane marked by the dotted line is not displayed.49 The user can adjust the number and position of slices with specific software controls until the desired planes of section can be visualized in the same image. FO. The overview image on the left upper panel of each figure shows the orthogonal sagittal plane to the sections that are being displayed. In this volume data set. The center slice is marked with an asterisk (*) and each subsequent plane to the right or left is marked with numbers ranging from minus −4 to +4. brightness. left pulmonary artery. interventricular septum. akin to display methods commonly used in computed tomography and magnetic resonance imaging.48. Medison. Tomographic Ultrasound Imaging. four-chamber. LPA. .4 and Video clip 14. GE Healthcare. the five transverse planes of section proposed by Yagel et al45 for examination of the fetal heart are visualized. WI. iSlice. Multi-Slice View. This approach represents an alternative to manually scrolling through the volume data sets to obtain standard cardiac views.48. 4CH. Milwaukee.3 and 14. IVS.48 (a) Figure 14. Accuvix. WA). WA. Each line represents a slice. foramen ovale. and may decrease the time spent in the evaluation of cardiac anatomy during screening and consultative examinations by allowing examiners to simultaneously view the spatial relationships of the transverse views described by Yoo et al44 and Yagel et al45 (Figures 14.3). This technology allows an examiner to automatically obtain a series of tomographic parallel images on a single screen. Seoul. Korea.4D ultrasound examination of the fetal heart by spatiotemporal image correlation 201 Automated multiplanar slicing Several ultrasound manufacturers now provide software to automatically slice 3D and 4D volume data sets (Multislice View™.

Each line represents a slice.52–54 In 2003. The center slice is marked with an asterisk (*) and each subsequent plane to the right or left is marked with numbers ranging from −4 to +4.45 The optimal interslice distance to automatically slice the volume data set and simultaneously visualize these planes of section changes with gestational age (Table 14.22.49 Manual adjustment of the interslice distance when all views are not obtained by automatically slicing the volume data set with the equipment improves the visualization rates of the transverse planes of section proposed by Yoo et al44 and Yagel et al. we proposed and subsequently validated a technique designed to simultaneously display the long-axis view of the left ventricular outflow tract and . In this volume data set.21.202 Fetal Cardiology (a) (b) Figure 14. the three planes of section proposed by Chaoui and McEwing51 for examination of the fetal heart are visualized (reproduced with permission from reference 49). The volume data sets were acquired using B-mode and color Doppler imaging.4 Tomographic ultrasound imaging (TUI) of a normal fetal heart in systole (a) and diastole (b). In addition. the visualization rate for the five-chamber view improves by approximately 14% over what is possible by gray-scale imaging alone for volumes acquired with color Doppler imaging.49 Systematic approach for visualization of the great vessels There are now several approaches to systematically demonstrate the great vessels in volume data sets acquired with STIC. Previous studies have demonstrated that the five basic axial planes of section for examination of the fetal heart can be automatically obtained in the majority of patients. The plane marked by the dotted line is not displayed.2). The overview image on the left upper panel of each figure shows the orthogonal sagittal plane to the sections that are being displayed.

The spatial relationships of the standard planes of section to the four-chamber view52.22 The technique was developed for volume data sets acquired by sweeping the fetal thorax on the transverse plane of section using the four-chamber view as the starting point.2 3.37 4.58 as well as the geometric changes that the fetal heart undergoes as gestational age progresses54 have been studied as a preliminary step for the development of such algorithms.82 3.53 According to this concept.75 SD (mm) 0. The latest approach to obtaining standard planes of section by manipulation of the volume data set has been denominated ‘automated sonography’. and may prove helpful in the identification of specific patterns of congenital heart disease in the future.21.4 4.5 and Video clip 14.34 0.6 and Video clip 14.7 1. three-vessel and trachea view. and superior vena cava.20 More recently. and is illustrated in detail in Figure 14.56 0.37 Gestational age (weeks) the short-axis view of the right ventricular outflow tract on panels A and B of the multiplanar display. Systematic approach for visualization of the aortic and ductal arches Ideal visualization of the aortic and ductal arches requires that the volume data set be acquired with sagittal sweeps through the fetal chest. the ductus arteriosus. and both outflow tracts on the same screen. which are considered part of an integral examination of the fetal heart. Details of the algorithm are described in Figure 14. simultaneous visualization of the short axis of the aorta. Rendering techniques for visualization of intracardiac structures and valves Rendering techniques can be applied to the visualization of intracardiac structures to obtain a depth perspective of . The original technique describing how to systematically visualize the aortic and ductal arches was originally published by Bega et al55 in 2001.2 2. Figure 14.65 0. The algorithm was developed to overcome a limitation of automatic slicing using tomographic ultrasound imaging alone.52. In a study that included 227 volume data sets of fetuses with (n = 14) and without (n = 138) congenital heart disease. Spin technique DeVore et al18 proposed in 2004 a technique to display the ascending aorta and transverse aortic arch.6.6 3. The technique is simple. Espinoza et al54 developed an enhanced algorithm combining tomographic ultrasound imaging and standardized manipulations of the volume data set to simultaneously display the three-vessel and trachea view.4 2.8 illustrates the application of the spin technique to identify an abnormal vessel visualized in the three-vessel and trachea view as a persistent left superior vena cava draining into a dilated coronary sinus.27 0.3 Maximum (mm) 1. and can be applied to any structure of interest during the examination of volume data sets of the fetal heart. which is that the simple use of parallel planes to the four-chamber view of the heart does not allow visualization of the long-axis view of the left outflow tract and the short-axis view of the aorta. and is illustrated in Figure 14. simply by positioning the reference dot in the center of the structure and ‘spinning’ the volume data set around the y axis until the structure is ‘opened up’ and visualized in its entirety. left outflow tract.5 show the clinical application of this technique for the diagnosis of transposition of the great arteries. and four-chamber view was achieved in 78% (152/195) of the volume data sets from fetuses without and 40% (8/20) of those with congenital heart disease.26 0. the standard planes of section for examination of the fetal heart can be automatically displayed by the equipment once the four-chamber view is selected as the reference plane.63 4.4 2. the main pulmonary artery and bifurcation of the right and left pulmonary arteries. the four-chamber view. The lower visualization rates of standard planes of section in cases of congenital heart disease may reflect abnormal spatial relationships between cardiovascular structures.3 5. Figure 14.7 and Video clip 14.4D ultrasound examination of the fetal heart by spatiotemporal image correlation 203 Table 14.6 5.14 1.4.6.3 Mean (mm) 1.4 4.2 12–15 16–19 20–24 25–29 30–34 35–40 Interslice distance according to gestational age. Adapted with permission from reference 50 n 11 10 44 15 17 6 Minimum (mm) 0.78 2.

transverse) and B (sagittal plane) are displayed. if necessary. (e) The volume data set is now rotated around the y axis. and the right side of the heart is on the right side of the image. which will display the left ventricular outflow tract. this will open up the continuity between the interventricular septum and the anterior wall of the aorta. the anterior leaflet of the mitral valve is seen in continuity with the posterior wall of the aorta. (a) The first step in this technique consists of making sure that the left side of the heart is located on the left side of the image. Only panels A (original plane of acquisition. on both the transverse and the sagittal planes. The reference dot is then positioned in the crux of the heart.204 Fetal Cardiology (a) (b) Figure 14. (f) Once the reference dot is moved above the aortic valve. this will anchor the three orthogonal planes for the next rotation movement. (c) The volume is then rotated around the z axis counter-clockwise until the angle between the apex and the transducer is approximately 30–40°.5 Technique to systematically visualize the left and right outflow tracts in volume data sets acquired with STIC. (d) The crucial step is to position the reference dot in the interventricular septum. the short-axis view of the right ventricular outflow tract is displayed on the sagittal plane. left ventricular outflow tract (reproduced with permission from reference 23). (c) (d) (e) (f) . rotate the volume around the y axis until this is achieved. LVOT. midway between the crux of the heart and the apex. (b) The volume data set is rotated around the z axis until a perfect apical four-chamber view is obtained.

when necessary. (c). the three-vessel and trachea view in panel B. using the ‘Slices’ option. The parallel lines determine the position of the eight orthogonal planes to the plane containing the ‘overview image’. where the fetal aorta was aligned with the crux of the heart in the vertical plane. (b) Volume data sets were adjusted to display the four-chamber view in panel A. (d) Only three planes were selected. In panel C the image was rotated to display the aorta in a vertical position. and the four-chamber view on panel D (reproduced with permission of the American Institute of Ultrasound in Medicine from reference 54). one plane to the left (‘−1’) and one to the right (‘+1’) to the plane crossing the reference dot. including the plane that crosses the reference dot (which is labeled with an asterisk in the software). The reference dot was positioned in the aorta allowing visualization of the coronal view of the descending aorta in panel C. These images were magnified using the four-panel ‘Display format’. This allowed simultaneous visualization of the short axis of the aorta in panel A. This allowed visualization of the longitudinal view of the ductal arch in panel B. This allowed simultaneous visualization of the ductal arch in panel A. the three vessel and trachea view in panel B. and the four-chamber view in panel D. (f) The ‘Rotation Y’ was selected by clicking on the bar. the image was moved until the reference dot was positioned in the center of the aorta. the five-chamber view in panel C. and the five-chamber view was rotated by scrolling on the y axis until the left outflow tract was visualized in panel B. . The ‘Adjust’ option was selected to align the ‘−1’ plane with the ductal arch and the ‘+1’ plane with the external edge of the aorta.6 (a) An ‘overview image’ is shown in the upper left corner.4D ultrasound examination of the fetal heart by spatiotemporal image correlation 205 (a) (b) (c) (d) (e) (f) Figure 14. the long axis of the left outflow tract on panel C. (e) In panel A.

the right upper panel image was simply rotated counter-clockwise around the z axis (curved arrow) (reproduced from reference 22 with permission from Elsevier). or to obtain realistic 4D images of particular structures of interest.22 This view can also be . (c) the area under examination. ‘thick slice’ rendering was applied to the atrioventricular valves to visualize the leaflets en face. The resulting image in the left upper panel was a sagittal view of the aorta.56 Rendering techniques can also be used to optimize contrast of myocardial borders. septa. and valves. (c) To demonstrate the ductal arch. as if the examiner was observing them from the ventricular chambers.17.206 Fetal Cardiology (a) (b) Figure 14. (a) 3D multiplanar image of the fetal thorax in the original acquisition plane (sagittal).9 and Video clip 14.22. Minor adjustment of this image around the y axis was required to demonstrate the aortic arch. In Figure 14.7.7 Three-dimensional (3D) multiplanar slicing of the aortic and ductal arches.3.(b) The reference dot was manipulated in the right upper panel and moved to the center of the aorta (white arrow).

(a) Three-vessel view showing the pulmonary artery (PA). the superior vena cava (SVC).4D ultrasound examination of the fetal heart by spatiotemporal image correlation 207 (a) (b) (c) (d) Figure 14. . (d) In order to know where this vessel was draining to. the ascending aorta (Ao). (c) The volume was rotated around the y axis. the reference dot was moved to the drainage site. this maneuver opens up the vessel so that it can be visualized in the sagittal plane. (b) The reference dot was moved from the pulmonary artery to the center of the structure under interrogation. (e) The volume was spun once more around the y axis. This confirmed the diagnosis of a persistent left superior vena cava (PLSVC).8 (e) Application of the spin technique18 to the identification of the abnormal vessel marked with ‘?’. revealing that the drainage site was the coronary sinus (CS). the descending aorta (DAo) and an abnormal structure that is not commonly visualized to the left of the pulmonary artery in this view.

in a case of complete atrioventricular canal. The key to obtain the rendered image (right lower corner) is the position and size of the region of interest (ROI). TV.208 Fetal Cardiology (a) Figure 14. This volume data set was rendered using a combination (mix) of 60% gradient light and 40% surface mode. MV. In this case.9 (b) ‘Thick slice’ rendering of the atrioventricular valves of a normal fetus. RV. Figure 14. which is the direction that the computer software will use to convert the voxels along the projection path to pixel information to be displayed on the twodimensional screen.57 Yagel et al57 were able to consistently obtain this rendered view in 93% of 136 normal fetuses examined between 21 and 26 weeks of gestational age. encompassing only the atrioventricular valves.8 we demonstrate the use of ‘thick slice’ rendering with inversion mode to emphasize the abnormal insertion of the tricuspid valve in a case of Ebstein’s anomaly. The green line indicates the direction of view. In Figure 14. In five of 35 abnormal cases examined in their study. and in a case of a hypoplastic tricuspid valve associated with pulmonary atresia. left ventricle.19 This technique has been reported as helpful in characterizing the hypokinetic . right ventricle (reproduced with permission from reference 23). this rendered view provided additional diagnostic information regarding the relative position of the great vessels to the atrioventricular valves as well as the appearance of the semilunar valves. LV. the position of the green line indicates that the atrioventricular valves are being visualized from the ventricular chambers. mitral valve.11 and Video clip 14. obtained by selecting the direction of rendering from the atrial chambers. tricuspid valve.10 presents a comparison of atrioventricular valves visualized en face in a normal case.

12 and Video clip 14.17. In the tricuspid stenosis with ventricular septal defect case.14 shows rendered views of the atrial and ventricular septum in the sagittal plane using a technique described by Yagel et al. Figure 14. Note the abnormal apical insertion of the tricuspid valve and the small right ventricle (reproduced with permission from reference 23). and. as a result of the depth perspective provided by rendering.60 or B-flow imaging. Figure 14.58 Figure 14. a cross-section of the dilated left pulmonary artery is also visualized.28 as well as by rendering gray-scale .57 This view may contribute to further elucidation of atrial and ventricular septal defects. This mode provides great contrast for visualization of the myocardium and atrioventricular valves.24. which is constricted at the level of the pulmonary valves.13 and Video clip 14.24. the poststenotic dilatation.10 show the pulmonary artery.11 ‘Thick slice’ rendering of the atrioventricular valves using inversion mode in a fetus with Ebstein’s anomaly. diastole). This is a common finding in absent pulmonary valve syndrome associated with tetralogy of Fallot. These images can be obtained by acquiring volume data sets with color Doppler.59 power Doppler17. Loss of continuity of the interventricular septum is observed during systole (white arrowhead) (reproduced from reference 22 with permission from Elsevier). Rendering techniques for visualization of the great vessels Several rendering algorithms have been described for visualization of the three-dimensional structure and spatial relationships of great vessels and venous return to the heart. Figure 14. motion of the ventricular wall in a case of congenital left ventricular aneurysm. The bright echogenic spot in the center of the image of the atrioventricular septal defect case (green arrowhead.4D ultrasound examination of the fetal heart by spatiotemporal image correlation 209 Figure 14. corresponds to the atrial septum secundum viewed from above. tricuspid stenosis (bottom). the tricuspid valve is very small with limited excursion during diastole.36.25. atrioventricular septal defect (middle).10 Comparison of three-dimensional surface rendering of the atrioventricular valves: normal fetus (top).19.9 show another example of ‘thick slice’ rendering to visualize an overriding aorta in a case of tetralogy of Fallot associated with absent pulmonary valve syndrome.

15 and Video clip 14. In order to visualize the great vessels leaving the heart. and B-flow imaging (Figure 14. whereas structures that are normally echogenic before gray-scale inversion (e.18 and Video clip 14. inversion mode (Figure 14.12 ‘Thick slice’ rendered image of the aorta (Ao) overriding the interventricular septum (IVS) in a case of absent pulmonary valve syndrome associated with tetralogy of Fallot.11 show the technique that we frequently use in our unit to visualize the crisscrossing of the outflow tracts as they leave the ventricular chambers. anechoic structures such as the heart chambers. we reorient the heart on the screen whenever necessary by rotating the volume dataset around the y and z axes until the four-chamber view is in the apical position.40. without the need for color Doppler. The algorithm used for rendering was the ‘gradient light’ mode (reproduced with permission from reference 23). which illustrates the three-dimensional rendering of a case of interrupted IVC with azygos continuation. In Figure 14. the user must set the direction of view to project the heart from the great vessels towards its base. . the application of this technique for the diagnosis of transposition of the great arteries is demonstrated. Inversion mode and B-flow Inversion of gray-scale voxels to visualize blood pools from cardiac structures was originally proposed by Nelson et al. including gamma curve correction to optimize contrast resolution. the direction of view was set from left to right. power Doppler (Figure 14. or B-flow imaging (Figure 14. using panel B as a reference. bones) appear anechoic. The stenotic pulmonary valve annulus (PVA).16c).19. With inversion mode.16b). from the heart base touching the diaphragm to the great vessels close to the neck). the rendering box is selected and adjusted in panel B to include the whole heart within the region of interest (in this case.g.62 Crisscrossing of the pulmonary artery over the ascending aorta as these vessels leave the ventricular chambers is best visualized in volume data sets acquired using transverse sweeps through the fetal thorax. The resulting rendered image is shown in panel D. power Doppler. Sagittal acquisitions are preferred when the objective is to visualize the aortic and ductal arches.40 in 1996. and bladder appear echogenic in the rendered images. Figure 14.16a).63 Figure 14.210 Fetal Cardiology Figure 14. vessel lumen.19. and the left side of the heart is displayed on the left side of the screen. B-flow technology digitally enhances weak blood reflector signals from vessels and.13). and a cross-section of the dilated annulus of the left pulmonary artery (LPA) are demonstrated in a single image. gallbladder.16d and Video clip 14. suppresses strong signals from the surrounding tissues.12). An example of the applicability of this technique is provided in Figure 14. at the same time. volume data sets with inversion mode. volume data sets are acquired using the fourchamber view as the starting point. The technique allows examiners to obtain 4D rendered images of cardiovascular structures from volume data sets acquired with gray-scale only. This principle has recently been incorporated into commercially available ultrasound equipment and is known as inversion mode.25. Postprocessing adjustments are performed as necessary.13 ‘Thick slice’ rendered image of the right ventricle (RV) and pulmonary artery (PA) in a case of absent pulmonary valve syndrome associated with tetralogy of Fallot. Next. and gray-scale threshold and transparency to improve image quality. The algorithm used for rendering was the ‘gradient light’ mode (reproduced with permission from reference 23).61.17.16 shows examples of the normal pulmonary artery crisscrossing over the aorta in volume data sets acquired with color Doppler (Figure 14. In our example. For the sake of reproducibility. stomach. the poststenotic dilatation of the pulmonary artery. Figure 14. renal pelvis. As described in the previous paragraph.

(a) The interventricular septum is rendered with the direction of view (green line) set from the right ventricle (RV).14.63–66 Due to its high sensitivity and angle independence to blood flow. Since this technology does not use Doppler methods to display blood flow. (b) The direction of view is set so that the volume is reconstructed with the IVS visualized from the left ventricle (LV). The principles utilized to obtain these images are the same as described for Figures 14.17. These factors include .19.14 Rendered images of the interventricular septum (IVS). B-flow is potentially advantageous over color or power Doppler imaging for the visualization of the great vessels and venous return to the heart. flap of the foramen ovale.4D ultrasound examination of the fetal heart by spatiotemporal image correlation 211 (a) (b) Figure 14.65 as well as for visualization of small vessels such as the coronary arteries67 and aortopulmonary collateral branches in cases of pulmonary atresia with a ventricular septal defect. the aortic and ductal arches of a normal fetus have been rendered using a volume data set acquired with B-flow imaging and the gradient light algorithm. Limitations Factors that affect image quality of conventional twodimensional ultrasonography are likely to impact the quality of STIC volume data sets.15–14. it is angle independent and does not interfere with frame rate as much as color or power Doppler.20 and Video clip 14. Flap.28 In Figure 14.

212 Fetal Cardiology Figure 14. power Doppler (b).16 Crisscrossing of the outflow tracts as they exit the ventricular chambers. as well as volume data sets acquired with B-mode imaging but rendered using inversion mode (reproduced with permission from reference 23).15 Technique to obtain rendered images of the outflow tracts using color Doppler. The rendered images were obtained with the technique described in Figure 14. (a) (b) (c) (d) Figure 14. The rendering box is adjusted to encompass the heart and great vessels.11. and B-flow (d) (reproduced with permission from reference 23). The volume data sets were acquired with color Doppler (a).11 and are explained in detail in Video clip 14. . gray-scale (then rendered with inversion mode) (c). The pulmonary artery always crosses in front of the aorta. The same technique can be applied to volume data sets acquired with power Doppler and B-flow imaging. The direction of view (green line) is set to project the rendered image from anterior (pulmonary artery) to posterior (aorta and ventricular chambers).

normal crisscrossing of the great arteries is observed.g. tetralogy of Fallot. significant fetal movement during acquisition artificially shifted the connections of the outflow tracts to appear as exiting from the right ventricle. The volumes were acquired through a transverse sweep of the fetal chest. and maternal obesity. pulmonary atresia . overriding ventricular septal defect. double outlet right ventricle. whereas in transposition of the great arteries.18 ‘Digital casts’ of the aortic and ductal arches obtained with inversion mode (reproduced with permission from reference 62). which was misinterpreted during analysis of the four-dimensional volume data set as double outlet right ventricle.4D ultrasound examination of the fetal heart by spatiotemporal image correlation 213 (a) (b) Figure 14. truncus arteriosus.22 We have observed such an artifact in a case of a fetus with transposition of the great arteries. In the normal fetus. unfavorable fetal position. Fetal movement and sudden changes in fetal heart rate (e. the vessels leave the ventricles in parallel. using power Doppler 4D ultrasonography with STIC.14 (reproduced with permission of the American Institute of Ultrasound in Medicine from reference 20). In this case. we caution that. Figure 14. The technique utilized to render the volume datasets is explained in Figure 14. early gestational age.68 Therefore.g. fetal arrhythmias) during acquisition are additional factors that may affect the technique and cause misregistration of the information required for precise reconstruction of moving cardiac structures. in the case of a suspected malalignment defect (e.17 Rendered images from volume data sets of the fetal heart in a normal case (a) and transposition of the great arteries (b).

(a) An image from the right side of the heart shows that the arch of the azygos vein joins the superior vena cava (SVC) before entering the right atrium. aortic arch. size. Video clip 14. The ROI defines the width and the height of the volume data set (x and y planes).19 Three-dimensional images of a fetal heart rendered with inversion mode in a case of interrupted inferior vena cava with azygos vein continuation associated with omphalocele. and (4) 3D and 4D rendering of cardiovascular structures to visualize the relationships.69 Legends for the DVD Figure 14.30.2 Standard planes of section that can be obtained by scrolling from the top to the bottom of the volume data set. (2) the use of techniques to systematically visualize the outflow tracts in volume data sets acquired using the four-chamber view image as the starting point. ductus arteriosus (reproduced with permission from reference 23). the finding should be confirmed by conventional two-dimensional ultrasonography prior to establishing the final diagnosis. This technology allows: (1) navigation through the volume data set and examination of the fetal heart in the absence of the patient. with a ventricular septal defect) that is observed first by volumetric imaging. 3D and 4D rendering of the great vessels.25. and course of the outflow tracts in normal fetuses and those with congenital heart disease. (b) Posterior view of the fetal heart shows a dilated azygos vein located to the right of the descending aorta.15. . Ao arch. has been previously possible only during postmortem examination by injection of silicone rubber to produce pathological casts of the cardiovascular system.214 Fetal Cardiology (a) (b) Figure 14. Video clip 14. using a similar acquisition and rendering technique as illustrated in Figure 14. illustrating the placement of the region of interest (ROI) and acquisition of the volume data set with STIC. DA.1 Two-dimensional video clip of the four-chamber view. in particular. acquired using transverse sweeps through the fetal thorax. Video clip 14. Conclusion In this chapter we describe a practical approach for the examination of the fetal heart by 3D/4D ultrasonography with STIC.20 Rendered image of the aortic and ductal arches obtained from a volume data set acquired with B-flow imaging. The arch of this vein forms a Y image with the aortic arch before joining the SVC (reproduced with permission from reference 62). (3) examination of the fetal heart using a tomographic approach similar to that used to read computed tomography and magnetic resonance imaging examinations.3 Tomographic ultrasound imaging of a normal volume data set of the fetal heart.

88: 387–91. PA. Wong SF. 17: 601–7. The key to obtain the rendered image (right lower corner) is the position and size of the region of interest. Ultrasound Obstet Gynecol 2003. 99: 916–18. Video clip 14. 2. Curr Opin Obstet Gynecol 2006. Note the abnormal apical insertion of the tricuspid valve. Ultrasound Obstet Gynecol 2006. 12. the poststenotic dilatation of the pulmonary artery and a crosssection of the dilated annulus of the left pulmonary artery are demonstrated in a single image. Factors influencing the prenatal detection of structural congenital heart diseases. Ultrasound Obstet Gynecol 2003. the position of the green line indicates that the atrioventricular valves are being visualized from the ventricular chambers. Burch M.4 and Video clip 14. 27: 107–13.14 Four-dimensional visualization of the aortic and ductal arches in a volume data set acquired with B-flow imaging.5 Demonstration of transposition of the great arteries using the technique described in Figure 14. Butera G et al. Labels for this image are provided in Figure 14. 6. Ultrasound Obstet Gynecol 2001. Sklansky MS. Video clip 14. Video clip 14. Heart 2002. Circulation 1999. Tegnander E. Rodeck CH. 9. Medearis AL. . 10. Heling KS. Performance of the basic fetal cardiac ultrasound examination. Ultrasound Obstet Gynecol 2006. Ultrasound Obstet Gynecol 2000. Rustico MA. 13. 14. Benacerraf B. Video clip 14.12 Crisscrossing of the outflow tracts in a volume data set acquired with B-flow imaging. RV. demonstrating the overriding aorta. Hunter S. 5. 21: 19–25. Video clip 14. 11. 18: 177–84.10 ‘Thick slice’ rendering of the right ventricle and pulmonary artery in a case of absent pulmonary valve syndrome associated with tetralogy of Fallot. which is the direction that the computer software will use to convert the voxels along the projection path to pixel information to be displayed on the two-dimensional screen.4. Falkensammer P.9. as well as for volume data sets acquired with B-mode imaging but rendered using inversion mode. The examiner’s ultrasound experience has a significant impact on the detection rate of congenital heart defects at the second-trimester fetal examination. References 1. Cincotta RB et al. Ao. Video clip 14. 17: 370–9. This volume data set was rendered using a combination (mix) of 60% gradient light and 40% surface mode. The same technique can be applied to volume data sets acquired with power Doppler and B-flow imaging. Heads A. 8. New developments in fetal heart scanning: three. Early screening for fetal cardiac anomalies by transvaginal echocardiography in an unselected population: the role of operator experience. 10: 567–77. left ventricle.4D ultrasound examination of the fetal heart by spatiotemporal image correlation 215 Video clip 14. 84: 294–8. DeVore GR. 87: 67–9. International Society of Ultrasound in Obstetrics and Gynecology. Heart 2000. Semin Fetal Neonatal Med 2005. Carvalho JS. Platt LD. Coltri A. Spatio-temporal image correlation (STIC): new technology for evaluation of the fetal heart. J Ultrasound Med 1998. Video clip 14. Manning N et al. Video clip 14.and four-dimensional fetal echocardiography. Copel JA et al. Fetal echocardiography: factors that influence imaging of the fetal heart during the second trimester of pregnancy. The stenotic pulmonary valve annulus. Chaoui R. Robson S.4 Demonstration of the technique to systematically visualize the outflow tracts in volume data sets acquired using transverse sweeps throught the fetal thorax with STIC. The green line indicates the direction of view. Mavrides E. Deng J. The algorithm used for rendering was the ‘gradient light’ mode. Prenatal diagnosis of coarctation of the aorta improves survival and reduces morbidity. Video clip 14. Cardiac screening examination of the fetus: guidelines for performing the ‘basic’ and ‘extended basic’ cardiac scan. Chan FY. 3.7 ‘Thick slice’ rendering of the atrioventricular valves of a normal fetus.8 ‘Thick slice’ rendering of the atrioventricular valves using inversion mode in a fetus with Ebstein anomaly. Eik-Nes SH. 28: 8–14. D’Ottavio G et al. 12: 659–63. pulmonary artery. right ventricle. Prenatal diagnosis of congenital heart disease in the northern region of England: benefits of a training programme for obstetric ultrasonographers. Bear MB et al.11 Technique to obtain rendered images of the outflow tracts using color Doppler. Wyllie J. Lee W. Current applications of fetal cardiac imaging technology. Bonnet D. Heart 2002. 4. Shinebourne EA et al. LV. Video clip 14. This mode provides great contrast for the visualization of the myocardium and atrioventricular valves. DeVore GR. 7. The algorithm used for rendering was the ‘gradient light’ mode. Benettoni A.6 Demonstration of a technique to visualize the aortic and ductal arches in 4D volume data sets acquired using sagittal sweeps through the fetal chest (described originally by Bega et al55 for volume data sets acquired by 3DUS). Improving the effectiveness of routine prenatal screening for major congenital heart defects.13 Technique to render the aortic and ductal arches using volume data sets acquired with gray scale imaging only and rendered with inversion mode. Allan L. Franklin O. In this case. Detection of transposition of the great arteries in fetuses reduces neonatal morbidity and mortality. J Ultrasound Med 1993. Isolated major congenital heart disease. 16: 614–19.9 ‘Thick slice’ rendering in a case of absent pulmonary valve syndrome associated with tetralogy of Fallot. encompassing only the atrioventricular valves. 22: 380–7. aorta.

Three-dimensional echocardiographic evaluation of fetal heart anatomy and function: acquisition. 23. Deng J. 68: 481–4. Giuliano A. Mandujano L. Gonçalves LF. Simpson J. 23: 159–64. 38. Ultrasound Obstet Gynecol 2004. Korai A. B-flow and B-flow spatio-temporal image correlation in visualizing fetal cardiac blood flow. Nelson TR. Ultrasound Med Biol 2001. Romero R. Chen CP. Rodeck CH. Messing B. 24. 25: 25–31. 22: 336–44. J Ultrasound Med 2004. Prenatal diagnosis of pulmonary atresia: impact on clinical presentation and early outcome. Ultrasound Med Biol 1996. Espinoza J. 27: 43–50. Reddy VM et al. Gonçalves LF. Campobasso G. Valsky DV. DeVore GR. Volpe P. Ultrasound Obstet Gynecol 2001. 23: 473–81. Paladini D. 26. 40. Ruff CF. 25. Croat Med J 2005. Gardener JE. Gonçalves LF. analysis. 18. Espinoza J. Shih JC. 103: 1269–73. Sklansky MS. Ultrasound Med Biol 2000. Yates R. 37. Ultrasound Obstet Gynecol 2006. 24: 415–24. Allan LD. Herberg U. Breuer J. Ultrasound Obstet Gynecol 1997. The ‘spin’ technique: a new method for examination of the fetal outflow tracts using three-dimensional ultrasound. 32. J Ultrasound Med 2005. Br Heart J 1992. Ultrasound Obstet Gynecol 2003. A systematic approach to prenatal diagnosis of transposition of the great arteries using 4-dimensional ultrasonography with spatiotemporal image correlation. The role of tissue harmonic imaging in fetal echocardiography. Novel application of 4D sonography with B-flow imaging and spatiotemporal image correlation (STIC) in the assessment of the anatomy of pulmonary arteries in fetuses with pulmonary atresia and ventricular septal defect. 80: 665–8. Spatio-temporal image correlation (STIC): a new tool for the prenatal screening of congenital heart defects. Deng J. Romero R. 44. Dynamic free-hand threedimensional fetal echocardiography gated by cardiotocography. Ultrasound Obstet Gynecol 2005. and display. Pretorius DH. Brekke S. Pretorius DH. Deng J. Vargas G. Detailed assessment of fetal ventricular septal defect with 4D color Doppler ultra- 30. 26: 1021–32. Improved surgical outcome after fetal diagnosis of hypoplastic left heart syndrome. Tissue Doppler gated (TDOG) dynamic three-dimensional ultrasound imaging of the fetal heart. Gonçalves LF. Examination of the fetal heart by four-dimensional (4D) ultrasound with spatio-temporal image correlation (STIC). 22: 493–502. Eik-Nes SH. Spatio-temporal image correlation (STIC): innovative 3D/4D technique for illustrating unique and independent information and diagnosing complex congenital heart diseases. 39. 24: 192–8. 23: 1225–31. Lee YH. A new approach to fetal echocardiography: digital casts of the fetal cardiac chambers and great vessels for detection of congenital heart disease. 28. Circulation 2001. 22: 388–94. 22: 979–86. 42. 36. Kovalchin JP. Lee W et al. Stanziano A et al. Birkett AG et al. Sklansky MS. Barker C. Examination of the fetal heart by five short-axis views: a proposed screening method for comprehensive cardiac evaluation. 17. Yagel S. Gonçalves LF. Yagel S. Lees WR. Deng J. 24: 72–82. Achiron R. J Ultrasound Med 1996. Fetal echocardiography in three and four dimensions. Ultrasound Obstet Gynecol 2004. 28: 40–6. Torp HG. 34. Pooh RK. Ultrasound Obstet Gynecol 2003. Birkett AG. Vinals F. Four-dimensional fetal echocardiography with spatiotemporal image correlation (STIC): a systematic study of standard cardiac views assessed by different observers. Chaoui R. Tworetzky W. Yoo SJ. Ultrasound Med Biol 2001. Espinoza J et al. Fagg NL. 27: 51–9. 31. 20: 209. Conversion of umbilical arterial Doppler waveforms to cardiac cycle triggering signals: a preparatory study for online motion-gated three-dimensional fetal echocardiography. 22. Vassallo M. Am J Cardiol 1997. Ultrasound Obstet Gynecol 2004. 19. 9: 173–82. sound using spatio-temporal image correlation technology. Tegnander E. Espinoza J. 17: 367–9. Kim ES et al. Romero R et al. 33. Hagen-Ansert S. 17: 323–31. Kalache KD et al. Heling KS. Bezold LI et al. 46: 812–20. 27: 336–48. Ultrasound Obstet Gynecol 2002. 46: 808–11. Fourdimensional ultrasonography of the fetal heart with spatiotemporal image correlation. Cook AC. 45. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatiotemporal image correlation (STIC). Cohen SM. Chaiworapongsa T et al. 14: 1025–9. 35. Richards R. Deng J. 20. Lee W. Platt LD. 25: 97–8. 41. 92: F199–203. Linney AD et al. Lee W. Polanco B. Four-dimensional ultrasonography of the fetal heart using color Doppler spatiotemporal image correlation. Ultrasound Obstet Gynecol 2004. Vinals F. Croat Med J 2005. Terminology of three-dimensional and fourdimensional ultrasound imaging of the fetal heart and other moving body parts. 16. 29. Three-vessel view of the fetal upper mediastinum: an easy means of detecting abnormalities of the ventricular outflow tracts and great arteries during obstetric screening. Simultaneous use of two ultrasound scanners for motion-gated three-dimensional fetal echocardiography. Lewin MB. Deng J. Tzifa A. 43. Harmonic imaging in fetal echocardiography. 27. Tibby S. Nelson TR. Ultrasound Obstet Gynecol 2006. 23: 535–45. Ultrasound Obstet Gynecol 2005. Espinoza J. Gonçalves LF. Prenatal diagnosis of congenital heart disease using four-dimensional spatio-temporal image correlation (STIC) telemedicine via an Internet link: a pilot study. Online motion-gated dynamic three-dimensional echocardiography in the fetus—preliminary results. Use of casts in the necropsy diagnosis of fetal congenital heart disease. 15: 1–9. Ultrasound Obstet Gynecol 2003. Romero R et al.216 Fetal Cardiology 15. . Sklansky M. Am J Obstet Gynecol 2003. 21. J Am Soc Echocardiogr 2001. Poblete P. Usefulness of gated three-dimensional fetal echocardiography to reconstruct and display structures not visualized with two-dimensional imaging. J Matern Fetal Neonatal Med 2005. Giuliano A. Dynamic three-dimensional gray-scale and color Doppler ultrasound of the fetal heart for dynamic diagnosis. J Ultrasound Med 2004. 189: 1792–802. Goldberg H. Tartaglione A et al. Arch Dis Child Fetal Neonatal Ed 2007. McElhinney DB. Hoffmann J.

Yang CK. DeVore GR. 68. Polanko B. Gonçalves LF et al. J Ultrasound Med 2001. 81: 465–7. 66. Gonçalves LF. Deane C. 61. J Radiol 2000. Bega G. Wachsberg RH. 63. Deng J. Extended field-of-view and B-flow ultrasound: fashion or future? Ultrasound Obstet Gynecol 2000.and 4-dimensional obstetric ultrasonography? J Ultrasound Med 2006. Kusanovic JP. 19: 114–22. 25: 947–56. 20: 131–6. Vinals F. Chaoui R. 47. Weskott HP. Abuhamad A. Espinoza J. 26: 366. . Int J Cardiol 1989. 24: 696–8. 53. Gonçalves LF. Ultrasound Q 2003. 52. J Ultrasound Med 2005. 58. Dynamic threedimensional color Doppler ultrasound of human fetal intracardiac flow. Nien JK. Three-dimensional extended imaging: a new display modality for three-dimensional ultrasound examination. Sullivan ID et al. Benachi A. 56. Ultrasound Obstet Gynecol 2006. Abuhamad A. The use of the minimum projection mode in 4-dimensional examination of the fetal heart with spatiotemporal image correlation. Ho SY. Ultrasound Obstet Gynecol 2005. McEwing R. J Ultrasound Med 2004. [B-flow—a new method for detecting blood flow]. 26: 244–51. Prenatal diagnosis of congenital left ventricular aneurysm by fourdimensional ultrasonography with spatio-temporal image correlation (STIC). Bonnet D et al. Lee W et al. Ultrasound Obstet Gynecol 2006. Lee W et al. 21: 81–93. What does 2-dimensional imaging add to 3. Gonçalves LF.4D ultrasound examination of the fetal heart by spatiotemporal image correlation 217 46. Gonçalves LF. Ultrasound Obstet Gynecol 2002. Tomographic ultrasound imaging of the fetal heart: a new technique for identifying normal and abnormal cardiac anatomy. Kalache KD. Espinoza J. 21: 59–65. Lev-Toaff A et al. Ultraschall Med 2000. 49. Henri P. Simple approach to prenatal diagnosis of transposition of the great arteries. Ultrasound Obstet Gynecol 2004. 25: 691–9. Ascenzo R. B-flow. 28: 266–74. 51. Chiang YC. Cardiovascular cavities cast in silicone rubber as an adjunct to post-mortem examination of the heart. Ultrasound Obstet Gynecol 2006. Espinoza J. Application of threedimensional ultrasonography in the evaluation of the fetal heart. Falkensammer P. Rendering in fetal cardiac scanning: the intracardiac septa and the coronal atrioventricular valve planes. [B-flow ultrasonographic imaging of circulating blood]. Yates R. Anderson RH. Paladini D. Ultrasound Obstet Gynecol 2006. 24: 1685–96. Shih JC. 28: 345–7. [in German] 67. 23: 573–6. A novel algorithm for comprehensive fetal echocardiography using 4-dimensional ultrasonography and tomographic imaging. 69. Ultrasound Obstet Gynecol 2005. [in French] 65. Lee CN. 25: 428–34. Ultrasound Obstet Gynecol 2003. Kilner PJ. Lee W et al. 34: 39–55. Ngai CS. 23: 1337–48. Ultrasound Obstet Gynecol 2005. 15: 96–7. Leung KY. 57. 22: 99–107.and fourdimensional reconstruction of the aortic and ductal arches using inversion mode: a new rendering algorithm for visualization of fluid-filled anatomical structures. 28: 399. Automated sonography: defining the spatial relationships of standard diagnostic fetal cardiac planes in the second trimester of pregnancy. Espinoza J. a non-Doppler technology for flow mapping: early experience in the abdomen. 17: 455–6. Espinoza J. The role of spatiotemporal image correlation (STIC) with tomographic ultrasound imaging (TUI) in the sequential analysis of fetal congenital heart disease. Espinoza J. A novel method to improve prenatal diagnosis of abnormal systemic venous connections using three. Three-dimensional power Doppler ultrasound of the fetal great vessels. Chaoui R. 50. 54. Automated multiplanar imaging: a novel approach to ultrasonography. Vassallo M. Poblete P et al. Chan BC et al. J Perinat Med 2006. 64. 59. Tranquart F.and four-dimensional ultrasonography and ‘inversion mode’. 20: 307–13. Four-dimensional ultrasonography of the fetal heart using a novel Tomographic Ultrasound Imaging display. Three cross-sectional planes for fetal color Doppler echocardiography. 27: 555–61. 48. 55. 62. Gonçalves LF. 28: 22–5. J Ultrasound Med 2004. Early visualization of the fetal coronary arteries by four-dimensional ultrasonography with B-flow imaging and spatiotemporal image correlation (STIC). Espinoza J et al. Three. Romero R et al. Kusanovic JP et al. Yagel S. J Ultrasound Med 2006. 60. Kuhlman K. Ultrasound Obstet Gynecol 2006. Ultrasound Obstet Gynecol 2001. Gonçalves LF. Sglavo G et al.

.

1–5 B-flow B-flow is an old–new technology that images blood flow without relying on Doppler shift. The introduction of ‘virtual planes’ to fetal cardiac scanning has helped sonographers obtain views of the fetal heart not generally accessible with a standard 2D approach. This reconstruction takes place directly following the scan in a matter of seconds. the systolic peaks that define the heart cycle4 (Figure 15. The methodology of STIC is comprehensively described in Chapter 14 of this text. However. Israel Shapiro. This cine-like file of a beating fetal heart can be manipulated to display any acquired scanning plane at any stage in the cardiac cycle (Figure 15. mathematical algorithms are applied to the volume data to detect systolic peaks that are used to calculate the fetal heart rate. B-flow directly depicts blood cell reflectors. With the advent of faster frame rates and computer processing. there is no doubt that 3D/4D gives us another look at the fetal heart. Briefly. or network. It avoids some of the pitfalls of Dop- . with particularly dramatic improvement in fetal echocardiography. Three. rather. In this chapter we will summarize the 3D/4D acquisition and post-processing modalities in 3D/4D fetal echocardiography. automated volume acquisition is made possible by the array in the transducer performing a slow single sweep. The result is a reconstructed complete heart cycle that displays in an endless loop. demonstrating their use through normal and anomalous case examples. personal computer (PC).and four-dimensional techniques and their application to fetal cardiac scanning Acquisition modalities Spatiotemporal image correlation Spatiotemporal image correlation (STIC) acquisition is an indirect motion-gated offline scanning mode. There is insufficient evidence to determine whether 3D/4D cardiac scanning will improve the accuracy of fetal echocardiography screening programs. tissue Doppler. and saved to the scanning machine. and Dan V Valsky Background Three. tomographic ultrasound imaging (TUI)). 25° acquisition would contain 1500 B-mode images.and four-dimensional (3D/4D) applications in fetal ultrasound scanning have made impressive strides in the past two decades. Sarah M Cohen. color and power Doppler. and to professional training. It appears that 3D/4D ultrasound applications will make a significant contribution to our understanding of the developing fetal heart in both normal and anomalous cases. A 10-second. VOCAL (virtual organ computer-aided analysis).4 Following acquisition. Recent technological developments in motion-gated scanning allow almost real-time 3D/4D cardiac examination. to parental counseling.15 Three.1). STIC acquisition can be combined with other applications by selecting the appropriate setting before acquisition (B-flow. allowing the STIC acquisition to be reviewed with the patient still present and repeated if necessary. The volume of interest (VOI) is acquired over a period of 7. and with its offline networking capabilities may improve healthcare delivery systems by extending the benefits of prenatal cardiac screening to poorly served areas. 3D/4D echocardiography may facilitate screening methods. recording a single 3D data set consisting of many 2D frames one behind the other. to interdisciplinary management team consultation.5 to about 30 seconds at a sweep angle of approximately 20–40° (depending on the size of the fetus) and frame rate of about 150 frames per second. inversion mode. high definition flow Doppler) or with post-processing visualization modalities (3D volume rendering. B-flow is an outgrowth of B-mode imaging. correlated to the internal trigger.and four-dimensional ultrasound in fetal echocardiography: a new look at the fetal heart Simcha Yagel.2). The B-mode images are arranged in order according to their spatial and temporal domain.

1 sec. By simultaneous analysis of the tissue movements. The resulting image is a live gray-scale depiction of blood flow and part of the surrounding lumen. 1 sec.6 This makes it an invaluable tool in fetal echocardiography. . the software identifies the beginning of each cycle and sets the time that each frame was acquired in respect of the beginning of the cycle. Though each frame composing the object was acquired in a different cycle. Figure 15. thus acquiring many frames per slice.3 and Video clip 15. without color Doppler flow information. (a) An object is contracting in a cyclical manner (4 seconds per cycle). 2 sec.1 Schematic demonstration of spatiotemporal image correlation (STIC) technology. Cycle duration.1). The shape is constructed from frames arranged side by side according to their position in the object (hence spatiotemporal). The scale applicable to fetal cardiac examination is discussed in the text. The shape of the object is presented at four points during the cycle.220 Fetal Cardiology (a) 0 sec. creating sensitive ‘digital casts’ of blood vessels and cardiac chambers (Figure 15. 3 sec. B-flow is also a sensitive acquisition tool for volume measurement. 2 sec. Knowing the time and position of each frame the software reconstructs the 3D shape of the complete object in each phase of the cycle (3). First cycle (total 4 seconds) Second cycle (b) Time from beginning of cycle Total cycle duration: 4 seconds Frames acquired in first 2D slice 1 2 3 2 Frames acquired in first 2D slice 1 Contracting object scanned in three consecutive slices Combining frames of identical phase in the cycle from consecutive slices frame’s spacial position is restored 3 Frames acquired in first 2D slice 0 sec. such as aliasing and signal drop-out at orthogonal scanning angles. number of slices. 3 sec. and number of frames per slice were chosen to simplify illustration. B-flow modality is a direct volume non-gated scanning method able to show blood flow in the heart and great vessels in real-time. (b) The object is scanned in three consecutive slices adjacent to each other (1). Assume that the contraction rate is too high to scan the whole object in conventional real-time 3D. In this example four frames are recorded in each slice (2). their phase in respect of the beginning of the cycle is identical (hence spatiotemporal). At least one complete cycle is recorded in real-time 2D ultrasound. pler.

one-color Doppler that is most effectively joined with static 3D scanning. 2 sec. or electronically by using a phased matrix of elements.8 Extreme care must be taken when working with Doppler applications in post-processing.Three. (Reproduced with permission from Yagel S.10 obviating the necessity of reconstructing a mental picture of the idiosyncratic course of an anomalous vessel from a series of 2D planes. and the most recent development. 1 sec. uses high resolution and a small sample volume to produce images with two-color directional information. resulting in a moving volume resembling real-time 3D. Ultrasound Obstet Gynecol 2007. It depicts flow at a lower velocity than color or power Doppler. TUI) with color. for example mild tricuspid regurgitation occurring very early in systole or very briefly. can be combined with static 3D direct volume non-gated scanning to obtain 3D volume files with two-color Doppler information or one-color 3DPD. thereby combining high resolution bidirectional flow Doppler with the anatomic acuity associated with power Doppler.1 Continued (c) The system completes its task by creating an endless loop animation composed of the consecutive reconstructed volumes of the cycle. This joins the Doppler flow to cardiac events2 and provides all the advantages of analysis (multiplanar reconstruction (MPR) rendering. resulting in a volume file that reconstructs the cardiac cycle. the newest development in color Doppler applications. The procedure takes only a few seconds. while the vascular portion of the scan is isolated for evaluation. Using the ‘glass body’ mode in post-processing.4). 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart.) 3D/4D with color Doppler. Color Doppler can be used more effectively in 3D/4D when combined with STIC acquisition7 in fetal echocardiography. Shapiro I.9. . see above). with color flow information.5 and Video clip 15. In 3DPD the vascular tree of the fetal abdomen and thorax is reconstructed. to avoid misinterpretation of flow direction as the volume is rotated. while retaining the advantage of flow directional information. It can be used with static 3D or 4D gated acquisition (STIC) and the glass-body mode. Cohen SM. Indeed.2 3DPD uses Doppler shift technology to reconstruct the blood vessels in the VOI. This has been shown to aid our understanding of the normal and anomalous anatomy and pathophysiology of vascular lesions11 (Figure 15. to produce high-resolution images of the vascular tree with bidirectional color coding (Figure 15. either by means of a small motor in some systems. however. isolated from the rest of the volume. (d) Schematic demonstration of the multiple slices through the heart acquired during a single STIC scan. 3 sec. The operator can scroll spatially to any plane in the volume (but not temporally: in this case. with less ‘blooming’ of color for more realistic representation of vessel size.and four-dimensional ultrasound in fetal echocardiography 221 (c) 0 sec. Valsky DV. The dedicated transducer automatically changes its scanning angle. as above. 29: 81–95. 3DPD is directionless. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. Color or power Doppler. It is particularly sensitive for imaging small vessels. © International Society of Ultrasound in Obstetrics and Gynecology. the stored reconstructed volumes are now available for analysis with post-processing techniques as described in the text. surrounding tissue is not shown. scanning is incomplete today without color Doppler. and 3D high definition power flow Doppler Color and power Doppler have been extensively applied to fetal echocardiography. high definition flow Doppler. color Doppler with STIC is more effective. 3D power Doppler (3DPD).2). High definition power flow Doppler. This combination of modalities is very sensitive for detecting intracardiac Doppler flow signals throughout the cardiac cycle. Reconstructed cycle Repetition of reconstructed cycle (d) Figure 15.

Valsky DV. By applying multiplanar reconstruction (MPR) the operator optimizes the four-chamber view (FCV) plane. Ultrasound Obstet Gynecol 2007. © International Society of Ultrasound in Obstetrics and Gynecology. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. y. or teaching.) systolic and diastolic flow are observed at the same time owing to the sensitivity of the modality. Any of the stored information can be shared for expert review. in addition to the original volume. this is generally a wedge-shaped chunk of the targeted area. adjusting the image both spatially along the x. when used with STIC acquisition. . and to the desired stage of the cardiac cycle. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. are stored to the patient file. inaccessible in 2D cardiac scanning.14–17 These views.6). and TUI 3D/4D volume sets contain a ‘block’ of information. Cohen SM. interdisciplinary consultation. For example. the ductus venosus is shown to remain filled in both systole and diastole. The navigation point is placed on the interventricular septum in the A-plane. parental counseling. as well as ‘virtual planes’ that are generally MPR. either as static images or 4D motion files. once obtained. 29: 81–95. (Reproduced with permission from Yagel S. various methodologies have been proposed to optimize the acquired volumes to demonstrate the classic planes of fetal echocardiography12.222 Fetal Cardiology Figure 15.2 The four-chamber view from a STIC acquisition in a third-trimester fetus in systole (a) and diastole (b). the B-plane shows the septum ‘en face’. 3D rendering. Post-processing modalities In post-processing. and z axes.13 (Figure 15. and the C-plane shows a coronal plane through the ventricles. Shapiro I.

For example. an anomalous vessel that might be disregarded in cross-section is confirmed in the longitudinal plane. and umbilical vein (UV).Three. See also corresponding Video clip 15. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. © International Society of Ultrasound in Obstetrics and Gynecology.e.1.2) or 3D volume rendering. the operator places the bounding box tightly around the interventricular septum. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Brachiocephalic trunk (BT). 3D rendering is another analysis capability of an acquired volume.and B-frames of the MPR display. the operator can view complex cardiac anatomy in corresponding transverse and longitudinal planes simultaneously. for example opening and closing of the atrioventricular valves. Inferior vena cava (IVC) is indicated. By comparing the A. The operator can determine whether the plane will be displayed from the left or right.) Figure 15. ductus venosus (DV). referred to as A (upper left).3 B-flow image of normal heart and aortic arch. Shapiro I. and any plane may be viewed at any time-point throughout the reconstructed cardiac cycle loop. for example. Ultrasound Obstet Gynecol 2007. In MPR the screen is divided into four frames. inferior vena cava (IVC). (Reproduced with permission from Yagel S. the fourth frame (lower right) will show either the volume model for reference. The reference dot guides the operator in navigating within the volume. Valsky DV. © International Society of Ultrasound in Obstetrics and Gynecology. the same plane can be displayed at any stage of the scanned cycle. the thickness of the slice will determine the depth .and four-dimensional ultrasound in fetal echocardiography 223 Figure 15. From a good STIC acquisition5 the operator can scroll through the acquired volume to obtain sequentially each of the classic five planes12 of fetal echocardiography. Ultrasound Obstet Gynecol 2007.4 3D power Doppler of the heart and major vessels. Valsky DV. Cohen SM. the septum from within the left or the right ventricle. such as imaging the fetal face in surface rendering mode. Noted are the carotid artery (CA). the operator displays 2D planes in either MPR mode (Figure 15. Shapiro I. By moving the point the operator manipulates the volume to display any plane within the volume. Each of the three frames shows one of the three orthogonal planes of the volume. if temporal information was acquired. Cohen SM. The rendered image in the D-frame will show an ‘en face’ view of the septum. left common carotid (LCC). and left subclavian artery (LSA) are seen projecting from the aortic arch (AoA). with the A-frame showing a good four-chamber view. The operator places a bounding box around the region of interest within the volume (after arriving at the desired plane and time) to show a slice of the volume whose depth reflects the thickness of the slice. or the rendered image. So. B. i. and C. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. 29: 81–95. 29: 81–95. The cycle can be run or stopped ‘frame-by-frame’ to allow examination of all phases of the cardiac cycle.) In order to analyze this effectively. (Reproduced with permission from Yagel S. aorta (AO). as it is anchored at the point of intersection of all three planes. In fetal echo it is readily applied to 4D scanning. It is familiar from static 3D applications.

or 30%. superior mesenteric artery. CA.e. SMA. i. descending aorta. pulmonary veins. The computer mouse is used to manually define the contours of the measured object (for example a heart ventricle) at each plane serially.13 The operator can . results in 12 planes available for measurement. the operator can opt for the system to draw the contours automatically. conversely.e. dAo.7). giving sequential views from −4 through +4. See also corresponding Video clip 15. according to varying degrees of sensitivity. The saved volume file is rotated 180° about a fixed central axis though a preset number of rotation steps based on the Screening examination of the fetal heart with three. celiac artery. Or. fluid-filled spaces such as the cardiac chambers now appear white.21.8). similar to 3DPD. TUI is a more recent application that extends the capabilities of MPR and rendering modes.) of the final image. Cohen SM. This multislice analysis mode resembles a magnetic resonance imaging or computer-assisted tomography display. Ultrasound Obstet Gynecol 2007. (Reproduced with permission from Yagel S. Many research teams have applied 3D ultrasound and STIC acquisition to fetal echocardiography. Alternatively. 9. as well as evaluation of congenital anomalies. 15. Nine parallel slices are displayed simultaneously from the plane of interest (the ‘zero’ plane). the distance of one plane to the next.and four-dimensional ultrasound Guidelines Guidelines for the performance of fetal heart examination have been published by the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG). and others. The thickness of the slices. while the myocardium has disappeared. for example. IM has the additional advantage of showing the stomach and gallbladder as white structures. i. UV. 6.224 Fetal Cardiology operator-chosen angle of rotation.22 IM analyzes the echogenicity of tissue (white) and fluid-filled (black) pixels in a volume and inverts their presentation. giving a more complete picture of the fetal heart (Figure 15. ductus venosus. the system reconstructs a contour model of the target. and various techniques have been put forward to optimize the use of this modality.5 Normal heart and great vessels: STIC acquisition with high definition power flow Doppler. 3D/4D ultrasound can enhance ‘basic’ and ‘extended basic’ fetal cardiac scans. Setting the rotation angle at 15°. A well-executed STIC acquisition5 contains all the necessary planes for evaluation of the five classic transverse planes of fetal echocardiography. lungs.12. The upper left frame of the display shows the position of each plane within the region of interest. © International Society of Ultrasound in Obstetrics and Gynecology. In fetal echocardiography it can be applied to create ‘digital casts’23 of the cardiac chambers and vessels. This application has the advantage of displaying sequential parallel planes simultaneously.18–20 Inversion mode Inversion mode (IM) is another post-processing visualization modality that can be combined with static 3D or STIC acquisition. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. DV. Shapiro I. relative to the reference plane. can be adjusted by the operator. for example to show texture of the trabeculations within the right ventricle (Figure 15.24 These guidelines for ‘basic’ and ‘extended basic’ fetal cardiac scanning can incorporate 3D/4D applications. 29: 81–95.20 Figure 15. umbilical vein. which may prove useful in the evaluation of fetal heart function. Once an outline is drawn around each plane of the target. IVC inferior vena cava. which can aid the operator in navigating within a complex anomaly scan. IM can be joined with STIC and VOCAL to quantify fetal cardiac ventricular volumes.2. including heart. This post-processing modality has been applied to volume calculation of numerous fetal organs. Valsky DV. VOCAL mode is a semiautomated 3D measurement mode that performs rotational measurement of volume. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. PV. it can produce a reconstruction of the extracardiac vascular tree.

25.6 The five planes of fetal echocardiography (reproduced with permission.26 This can have important implications in . will contain artifacts within the scan volume.1 for full explanation. ST LI AO SP examine the fetal upper abdomen and stomach. it must be remembered that STIC acquisition that was degraded by maternal or fetal movements. However. and finally the three-vessel and trachea view. Performed properly. above. including fetal breathing movements. These capabilities were applied by Vinals and colleagues to increase delivery of prenatal cardiac scanning to poorly served areas.Three. the bifurcation of the pulmonary arteries.and four-dimensional ultrasound in fetal echocardiography 225 LT P (P)Ao RT DA V (D)Ao T SVC LT RT IV LPA MPA AO RA RPA SVC AO T DA AO PA LT RT LV RV RA LA AO III LT RT LV RV RA LA II PV FO PV AO LT RT I Figure 15. then scroll cephalad to obtain the familiar four-chamber view. from reference 6). the five-chamber view. See Chapter 13. Applications Among the most attractive facets of 3D/4D scanning are the potential for digital archiving and sharing of examination data over a network. Slight adjustment along the x or y axes may be necessary to optimize the images. Figure 13. they were subsequently sent over an internet link and analyzed by expert examiners in central locations. this methodology will provide the examiner with all the necessary planes to conform to the guidelines. Local practitioners in distant areas acquired and stored 3D volume sets at their centers.

thus reducing the chance of acquisition degradation from fetal movements. Cohen SM.7 Normal interventricular septum in 3D rendering mode. the outflow tract view is imaged in the A plane. these parameters have eluded practical prenatal quantification. 29: 81–95. Espinoza and colleagues introduced a novel algorithm combining STIC and TUI16 to image the diagnostic planes of the fetal heart simultaneously. Functional evaluation of the fetal heart: ventricular volumetry Evaluation of fetal heart functional parameters has long challenged fetal echocardiographers. ejection fraction. coining the term ‘automated multiplanar imaging’ or AMI. many of the pediatric and adult measures are based on end-systolic and end-diastolic ventricular volumes: stroke volume. In frame A (a) the bounding box is placed tightly around the septum with the active side (green line) on the right. Without electrical trace or clinically applicable segmentation methods to determine the ventricular volume. and facilitate visualization of the long-axis view of the aortic arch. © International Society of Ultrasound in Obstetrics and Gynecology. Shapiro I. as the smaller fetal heart can be scanned in a shorter acquisition time.226 Fetal Cardiology Figure 15. subsequent analysis will be prone to lower image quality and the introduction of artifacts.14 combining MPR and STIC acquisition to analyze acquired volumes and simplify demonstration of the ventricular outflow tracts. as with any post-processing technique. 3D ultrasound opens new avenues for . DeVore and colleagues presented the ‘spin’ technique. Valsky DV. Ultrasound Obstet Gynecol 2007. if the original volume was suboptimal. The D-frame (b) shows the septum ‘en face’: note the rough appearance of the septum from within the trabeculated right ventricle.28 increasing demand for fetal cardiac screening programs. STIC acquisition is amenable to younger gestational ages. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.14 Abuhamad proposed an automated approach to extract the required planes from an acquired volume.27. and a computer-automated program could present those planes once the appropriate volume block is acquired. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart.15 Most recently. (Reproduced with permission from Yagel S.15 Based on the idea that the scanned 3D volume contains all possible planes of the scanned organ. Beginning from the four-chamber view. Nuchal translucency screening programs will refer approximately 3–5% of patients for fetal echocardiography as high-risk. the operator acquires a VOI from a transverse sweep of the fetal mediastinum that includes the sequential planes of fetal echocardiography. While duplex and color Doppler flow nomograms have been quantified and are long-established in 2D fetal echo. and outflow tract and adjacent vessels are then examined by placing the reference point over each vessel and rotating the image along the x and y axes until the full length of each vessel is identified. for example the sequential scanning planes of fetal echocardiography. and cardiac output. In post-processing.) increasing the penetration of prenatal ultrasound services in poorly served or outlying areas of many countries. Using this technique. it should be possible to define the geometric planes within that volume that would be required to display each of the diagnostic planes of a given organ. all the other planes are in constant anatomic relationship to this plane. However.

right).32 We recently published20 a methodology combining STIC acquisition to determine the end-systolic and enddiastolic stages in the cardiac cycle. away from the patient and the time constraints of a busy clinic. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.8 Tomographic ultrasound imaging (TUI): the −4 plane (top row. center) shows the FCV while the zero plane (asterisk.3). right) shows the outflow tract view and the +3 plane shows the great vessels (bottom. Examination volumes are stored for later analysis. then inversion mode to isolate the fluid-filled ventricular volume. Bhat and colleagues used non-gated static 3D acquisition and STIC to obtain mid-diastolic scans of fetal hearts.29. Shapiro I. Cohen SM.050 g/cm3) to obtain the mass.31. In cases of congenital heart disease (CHD).) exploration into ventricular volumetry20. middle row.20 It was found that both the inversion mode and VOCAL analysis were highly dependent on operator-determined threshold parameters. 29: 81–95. which was measured using VOCAL analysis (Figure 15. A similar study of cardiac mass is under way. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. The result was multiplied by myocardial density (1. The resulting volumes allowed quantification of stroke volume and ejection fraction. (Reproduced with permission from Yagel S. © International Society of Ultrasound in Obstetrics and Gynecology.9 and Video clip 15. other professionals can be invited to view the examination.30 and mass measurement. Three. Ultrasound Obstet Gynecol 2007. and applied VOCAL analysis to determine cavity volume. which affect the intensity of signal to be colored and included in the volumetry. Valsky DV.and four-dimensional ultrasound in fetal echocardiography 227 Figure 15.and four-dimensional ultrasound in the diagnosis of congenital heart disease One of the great advantages of a 3D/4D system is its digital archiving capabilities.Three. They can do this anywhere that an .

IAS) planes. In addition. and parents. and the coronal atrioventricular (CAV) plane of the cardiac valves’ annuli. bottom right. The first examiner can consult with the attending physician. image data contains and makes available many scanning planes that are not readily accessible in 2D ultrasound.3. restrictive foramen ovale. Cohen SM. This block of spatial and temporal Segmental approach The segmental approach to congenital heart disease has helped to standardize the description of cardiac lesions. See also corresponding Video clip 15. The interventricular and interatrial septa (IVS. Valsky DV. In addition.) internet link is available. Shapiro I. 29: 81–95. alignment of the ventricles and great vessels. The term ‘virtual planes’ was coined to refer to these rendered scanning planes. Virtual planes As described above. . Many teams have applied 3D/4D ultrasound capabilities to the diagnosis of congenital cardiovascular malformations. a properly executed STIC acquisition results in a volume ‘block’. Each of the modalities and applications described above lends itself to different facets of this complex endeavor. Ultrasound Obstet Gynecol 2007. Complex malformations can be elucidated through interdisciplinary discussion and made clear to laymen. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. cardiologist. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. and evaluation of the atrioventricular (AV) valves. The resulting measurements appear in the box. surgical or other management teams. © International Society of Ultrasound in Obstetrics and Gynecology.228 Fetal Cardiology Figure 15. it has contributed to understanding the pathophysiology of the malformed developing fetal heart.9 STIC acquisition combined with inversion mode (IM) and virtual organ computer-aided analysis (VOCAL) for fetal cardiac ventricle volumetry.17 They were shown to have added value in the diagnosis of ventricular septal defect. (Reproduced with permission from Yagel S. which is a reconstructed complete cardiac cycle. have been investigated and applied to the evaluation of CHD. genetic counselors. stored data from cases of CHD are invaluable teaching materials for professional education.

and four-dimensional ultrasound in fetal echocardiography 229 and subsequently to our conceptualization and diagnostic imaging.34 Essentially.4).10a and Video clip 15.33 we will follow this sequence in describing the application and added value of 3D/4D in the diagnosis of CHD.4. The present case was an intradiaphragmatic variation with drainage of the pulmonary veins to the portal vein. Interrupted inferior vena cava (IVC) with azygos continuation is shown in Figure 15. © International Society of Ultrasound in Obstetrics and Gynecology. all involve an abnormality of the AV valves. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.12 shows the use of 3D rendering of a STIC volume acquired . This confirmed that the finding was not an artifact. Shapiro I. in these anomalies the pulmonary veins do not drain into the left atrium. the ventricles. through index cases of anomalies diagnosed in our center. as well as the aorta in one three-dimensional image that would be impossible to obtain with 2D color Doppler scanning. The sequential segmental approach essentially divides the heart into three basic segments: the atria. Cohen SM. which is confirmed in the B-plane (arrow).3 and 15. AVSD has many forms. Valsky DV.11 and Video clip 15. These are divided and joined at the level of the atrioventricular valves. 3D power flow Doppler displayed the idiosyncratic vascular tree and absence of the pulmonary veins (Figure 15.3 show the use of MPR with the reference point to navigate this complex lesion. 3D/4D ultrasound can have a significant contribution to the understanding of the fetal venous system. rotation of the image in post-processing allowed overall examination of the lesion in 360°. forming a common atrioventricular junction. Atrioventricular junction: atrioventricular septal defect and tricuspid valve stenosis Atrioventricular septal defect (AVSD) is characterized by incomplete atrial and ventricular septation. The A-plane shown raised suspicion of an anomalous vessel (caret).Three. Note also the absence of pulmonary veins (compare Figure 15.10b and Video clip 15. Figure 15. and at the ventriculoarterial junctions. but rather to various other locations: the right atrium.35 Blood is shunted directly into the right supracardinal vein (which will become the superior vena cava (SVC)) blood from the lower body flows through the azygos vein to the SVC. B-flow scanning provided superior imaging of the Figure 15. or abdominal veins. (Reproduced with permission from Yagel S.10 (a) STIC acquisition in a case of total anomalous pulmonary venous connection (TAPVC). and the great arteries. This cardinal vein anomaly results from primary failure of the right subcardinal vein to connect to the hepatic segment of the IVC.5. Veins and atria: total anomalous pulmonary venous connection and interrupted inferior vena cava with azygos continuation Total anomalous pulmonary venous connection (TAPVC) is a many-faceted group of malformations affecting the pulmonary veins. (b) The heart and great vessels of this fetus: STIC acquisition and high definition power flow Doppler confirmed the characteristic vertical vein (VV). but rather the characteristic vertical vein. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart.7). great veins. B-flow acquisition provided real-time representation of the anomalous course of the IVC and connection to the fetal heart. Figure 15. See also corresponding Video clips 15. The segmental approach to diagnosis of CHD is comprehensively and concisely described elsewhere. the variations and classification are described in detail in Chapter 29. 29: 81–95. It showed the azygos vein draining into the SVC. Placement of the reference point in the suspected anomalous blood vessel in cross-section (A-frame) showed the vessel in longitudinal plane in the B-frame. Ultrasound Obstet Gynecol 2007. In this instance.) slower blood flow in the azygos vein than was demonstrated with 3DPD.

230 Fetal Cardiology Figure 15. or stenosis. with the reference point placed on the septum with the four-chamber view in the A-frame. Another group of AV valve lesions is mitral or tricuspid valve atresia. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. 29: 81–95. Valsky DV. resembling the surgical plane seen when the heart is opened in surgery. for a more detailed examination of the size and nature (and number) of the VSD(s). Cohen SM. We recommend. This virtual plane provides a three-dimensional look at the AV and semilunar valves’ annuli. © International Society of Ultrasound in Obstetrics and Gynecology. Shapiro I. (Reproduced with permission from Yagel S. are described in Chapter 19. The operator places the ‘active’ side of the box to the right or left (i. Shapiro I. the plane is slightly adjusted along the x and y axes. SVC.11 B-flow image of the heart and great vessels in a fetus with interrupted inferior vena cava with azygos continuation. Valsky DV. by placing the bounding box tightly around the level of the AV connection in the four-chamber view. Ultrasound Obstet Gynecol 2007. Cohen SM. aorta.) The coronal atrioventricular (CAV) plane from STIC acquisition with color Doppler mapping in a case of atriventricular septal defect (AVSD). Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. the rendered image (frame D) shows the AV valves with anomalous anatomy (compare normal CAV plane.e. AzV.38 By using MPR. Ventricles: ventricular septal defects Ventricular septal defects are perhaps the most common – and most commonly missed – congenital heart defect. the use of the bounding box in 3D rendering from STIC acquisition with color Doppler. ascending aorta. DV. (Reproduced with permission from Yagel S. AVSD. The natural history and in utero development of these lesions Ventriculoarterial junctions (conotruncal anomalies): transposition of the great arteries Transposition (or malposition or malalignment) of the great arteries (TGA) is the general name for a complex . Ultrasound Obstet Gynecol 2007.13 shows the CAV plane in a case of tricuspid stenosis. however. with the superior side active (frame A). 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. © International Society of Ultrasound in Obstetrics and Gynecology. dysplasia.) with color Doppler to demonstrate the anomalous intracardiac flow resulting from the AVSD. from within the left or right ventricle) and obtains an image (in the D–frame) having more depth.36 Several groups have proposed methods for evaluation of the interventricular septum. superior vena cava. PA. azygos vein. The addition of color Doppler demonstrates blood flow across the lesion and shows at what stage in the cardiac cycle and to what degree shunting occurs.5. See also corresponding Video clip 15.14 and Video clip 15. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. This ‘virtual plane’ is obtained from a STIC volume with color Doppler. 29: 81–95.12 Figure 15.6). Figure 15. ductus venosus. pulmonary artery. AO.37. the B-frame will show the septum and defect ‘en face’ (Figure 15. inset). atrioventricular septal defect. AoA.

29: 81–95. Shapiro I. Cohen SM. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG.13 Tricuspid stenosis evaluated with 3D rendering and the CAV plane. mitral valve annulus. Shapiro I. Ultrasound Obstet Gynecol 2007. The navigation point is placed on the septum in the A-plane (a).15 and Video clip 15.6. This demonstration of the anatomic variant of the anomaly . inset: mv. (Reproduced with permission from Yagel S. Valsky DV. © International Society of Ultrasound in Obstetrics and Gynecology. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. pa. tricuspid valve annulus. tv. the D-frame (b) shows the rendered IVS with flow across the defect from right to left. © International Society of Ultrasound in Obstetrics and Gynecology. ao. When the sequential segmental approach is applied to systematic diagnosis of congenital heart disease. 3D rendering with color Doppler was applied to the evaluation of suspected malalignment of the great vessels. aortic valve. 29: 81–95. pulmonary valve.7 show a case of complete dextrotransposition of the great arteries. Valsky DV.33 the morphology of each successive anatomic segment is assessed in turn.) group of anomalies with widely varying anatomic and clinical presentations. by examining the CAV (‘surgical plane’) at the level of the AV and semilunar valves’ annuli. See also corresponding Video clip 15.14 The interventricular septum (IVS) ‘virtual plane’ with color Doppler in evaluation of ventricular septal defect (VSD). Ultrasound Obstet Gynecol 2007. The bounding box is placed tightly around the level of the AV valves in the A-frame (a). 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. The B-flow scan clearly showed blood flow into the ventricles and out through the malaligned vessels. Compare normal CAV plane in diastole.and four-dimensional ultrasound in fetal echocardiography 231 Figure 15. (Reproduced with permission from Yagel S. Figure 15.) Figure 15. the D-frame (b) clearly shows the stenotic valve (arrow). The morphologic right and left atria and ventricles are established. Cohen SM. now the examiner addresses the ventriculoarterial junction and the accordance or discordance of the great arteries and ventricles. We applied B-flow scanning to the evaluation of TGA and found that it was more effective than 3DPD or inversion mode in visualizing the great vessels’ structure and relationships.Three.

20 Our normal cases showed the effectiveness of fetal heart ventricle volumetry in cardiac evaluation and quantification. aorta. M. The CAV plane is an excellent tool for evaluation of the semilunar valves.15 B-flow modality showed the parallel great vessels in a case of transposition. tricuspid annulus. Figure 15. 29: 81–95.) Functional evaluation: ventricular volumes We recently published20 novel methodology combining STIC acquisition with post-processing application of inversion mode and VOCAL to quantify end-systolic and end-diastolic ventricular volumes. and vein of Galen aneurysm. Shapiro I. Cohen SM. Cohen SM. The resulting measurements correlated strongly with gestational age and estimated fetal weight. the 4D-cine option is initiated and blood flow across the valves evaluated through the cardiac cycle. 29: 81–95. These included critical pulmonary stenosis. AO.16 Figure 15. The measured volumes were used to create nomograms for fetal stroke volume and cardiac ejection fraction. Valsky DV. Right aortic arch defect results from persistence of the right dorsal aorta and involution of the distal part of the left dorsal aorta.) The CAV plane from STIC acquisition with color Doppler mapping in a case of transposition of the great arteries and pulmonary stenosis with retrograde flow in the main pulmonary artery. Nomograms were created from right and left ventricle end-systolic and end-diastolic volumes from 100 fetuses examined between 20 and 40 gestational weeks. Ebstein’s anomaly. There are two main types. supraventricular tachycardia. AO. Once the CAV plane is obtained.17 shows a case of right aortic arch diagnosed with B–flow. aortic valve stenosis with hypoplastic aortic arch. twin-to-twin transfusion syndrome with secondary pulmonary stenosis. this modality showed the characteristic course of the aortic arch to the right of the trachea. mitral annulus. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. or ejection fraction. stroke volume. pulmonary artery. Shapiro I. Arterial trunks: pulmonary stenosis and right aortic arch The use of 3D rendering of a STIC acquisition with or without color Doppler to obtain virtual planes is discussed above. The methodology was applied to saved STIC volumes of cases with cardiac anomaly or dysfunction that involved changes in ventricular volume.232 Fetal Cardiology Figure 15. Application of this modality clearly shows the blood flow in the malaligned vessels. © International Society of Ultrasound in obstetrics and Gynecology. See also corresponding Video clip 15. with or without a retroesophageal component. valsky DV. pulmonary artery. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. Ultrasound Obstet Gynecol 2007. . PA.35 Figure 15.16 shows a case of critical pulmonary stenosis with retrograde flow in the main pulmonary artery (MPA). © International Society of Ultrasound in Obstetrics and Gynecology. (Reproduced with permission from Yagel S. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. such volumetry is not readily available aided our consultations with the parents and their attending physician. Ultrasound obstet Gynecol 2007. PA. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart.7. (Reproduced with permission from Yagel S. T. aorta.

Shapiro I. before starting 3D or 4D acquisition. Flow direction Potential pitfalls of threeand four-dimensional echocardiography 3D/4D fetal echocardiography scanning is prone to artifacts similar to those encountered in 2D ultrasound. the B-frame reveals artifacts introduced by fetal breathing movements (Figure 15. It is imperative to review suspected defects with repeated 2D and 3D scanning to confirm their presence in additional scanning planes. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. Ultrasound Obstet Gynecol 2007. y. When reviewing a STIC acquisition. flow data can be misinterpreted. When commencing scanning from the 2D plane. 3D rendering should always be used in conjunction with the A-frame 2D image for comparison. they may be present within the acquired volume block. . in 2D echocardiography. Cohen SM. Figure 15.and four-dimensional ultrasound in fetal echocardiography 233 STIC acquisition quality The quality of a STIC acquisition may be adversely affected by fetal body or ‘breathing’ movements. DA. the fetus should be in a quiet state and the shortest scan time possible employed. Application of some algorithms designed to smooth the image can lead to loss of data from the original scan. MPA. (Reproduced with permission from Yagel S. and z axes for analysis. © International Society of Ultrasound in Obstetrics and Gynecology. The quality of the original acquisition impacts on all further stages of post– processing and evaluation. and can be used for further investigation. To improve scan quality. the diagnosis was more serious than suspected by 2D echocardiography. rotation of the volume with Doppler directional flow information can mislead the operator: if the directions are reversed.18). However. the volume is usually acceptable. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. The operator must confirm any suspected pathological flow patterns by confirming the original direction of scanning. and some that are specific to 3D/4D acquisition and post-processing. An acquired volume containing Doppler flow information is available for manipulation and may be sliced and rotated around the x. Original angle of insonation The original angle at which a scan was performed will impact on the quality of all the planes acquired. Valsky DV. 29: 81–95. ductus arteriosus. whether flow was toward or away from the transducer during the acquisition scan.) Acoustic shadows Shadowing artifacts pose a particular problem to 3D/4D ultrasound. However. If the B-frame appears sound. main pulmonary artery.20 Three-dimensional rendering 3D rendering creates virtual images.Three.17 B-flow modality in a case of right aortic arch (RAoA). for example. acoustic shadows may not be apparent. Ventricle volumetry also provided insight into the pathophysiology of lesions such as supraventricular tachycardia (SVT) and vein of Galen aneurysm. among others. It is important to achieve an optimal beginning 2D plane. The pathological cases showed the potential added value of this methodology. In the case of critical pulmonary stenosis.

41 Wang and colleagues42 compared 3D and 2D scanning of fetuses in the spine-anterior position.234 Fetal Cardiology Figure 15. no large study has examined the contribution of 3D/4D to fetal echocardiography screening programs’ accuracy. This group found that only in the pulmonary outflow tract was 3D ultrasound superior to 2D.18 Artifacts and pitfalls. © International Society of Ultrasound in Obstetrics and Gynecology. and still others have described the application of various 3D/4D modalities to diagnosis or evaluation of fetal cardiovascular anomalies.) Accuracy While several studies have compared imaging yield between 2D and 3D/4D fetal echocardiography. Permission is granted by John Wiley & Sons Ltd on behalf of ISUOG. The investigators found that IM improved visualization of cases of dilated azygos or hemiazygos veins and their spatial relationships with the surrounding vascular structures. Levental and co-workers compared 2D and non-gated 3D ultrasound to obtain standard cardiac views.39 MeyerWittkopf and colleagues40 evaluated 2D and Dopplergated 3D in obtaining standard echocardiography scanning planes in normal hearts. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart. Espinoza and colleagues22 examined the added value of IM in the evaluation of anomalous venous connections. Ultrasound Obstet Gynecol 2007. 29: 81–95. . They found that 3D provided additional structural depth and allowed a dynamic 3-D perspective of valvar morphology and ventricular wall motion. Shapiro I. STIC acquisition in a 26-week fetus: the A-frame shows left ventricular outflow tract plane. Cohen SM. Valsky DV.40 In evaluating CHD. They determined that 3D had added value in a small proportion of lesions. Meyer-Wittkopf and co-workers41 evaluated gated 3D volume sets of 2D-diagnosed cardiac lesions. and others have examined the feasibility of 3D/4D and STIC in screening programs. Note that the B-frame however is degraded by fetal breathing artifacts (arrows). (Reproduced with permission from Yagel S. and compared key views of the heart in both modalities.

Ultrasound Obstet Gynecol 2006. 3. Sklansky MS. 2. 29: 81–95. descending aorta. Studies of the fetal heart. 3D and 4D ultrasound in fetal cardiac scanning: a new look at the fetal heart.Three. Video clip 15. inferior vena cava. DV. Fourdimensional ultrasonography of the fetal heart with spatiotemporal image correlation. Ultrasound Obstet Gynecol 2003. Terminology of three-dimensional and fourdimensional ultrasound imaging of the fetal heart and other moving body parts. 27: 336–48. 22: 336–44. pulmonary veins. This may provide insights into the physiological effects of fetal structural or functional cardiac defects. Note also the absence of pulmonary veins (compare Figure 15. Falkensammer P. UV. (b) The heart and great vessels of this fetus: STIC acquisition and high definition power flow Doppler confirmed the characteristic vertical vein (VV. superior mesenteric artery. Spatio-temporal image correlation (STIC): new technology for evaluation of the fetal heart. but among the generality of professionals performing fetal echocardiography. Espinoza J.2 Normal heart and great vessels: STIC acquisition with high definition power flow Doppler. no large study has been performed to examine whether the addition of 3D/4D modalities to fetal echocardiography screening programs increases detection rates of cardiac defects. Volpe P. Michailidis and co-workers44 and Vinals and colleagues25. Campobasso G. PA. . 22: 380–7. Brachocephalic trunk (BT). Lee W. AzV. Ultrasound Obstet Gynecol 2007. Video clip 15.) Video clip 15.1 B-flow of normal heart and aortic arch. bottom right. Gonçalves LF. To date. ductus venosus. 189: 1792–802. (AoA. inferior vena cava). azygos vein. Inferior vena cava (IVC) is indicated. pulmonary artery. The computer mouse was used to manually define the contours of the ventricle at sequential planes. on the developing fetus.5 B-flow of the heart and great vessels in a fetus with interrupted inferior vena cava with azygos continuation. Video clip 15. superior vena cava. umbilical vein. celiac artery. Chaiworapongsa T et al. Am J Obstet Gynecol 2003.and four-dimensional ultrasound in fetal echocardiography 235 Most recently. Stanziano A et al. Examination of the fetal heart by four-dimensional (4D) ultrasound with spatio-temporal image correlation (STIC). left common carotid (LCC). This technology has reached the stage when its reproducibility and added value in screening accuracy should be tested in large prospective studies. Zipf. DV. Ultrasound Obstet Gynecol 2003. studies will direct 3D/4D capabilities to the evaluation of fetal cardiac functional parameters. Shapiro I. Valsky DV. Gonçalves LF.3 STIC acquisition combined with IM and VOCAL analysis for fetal cardiac ventricle volumetry. Deng J. 28: 40–6. Falkensammer P. Benacerraf and colleagues43 compared acquisition and analysis times for 2D and 3D fetal anatomy scanning at 17–21 gestational weeks. Romero R.26 have shown the feasibility and success of programs based on 3D/4D examination volumes acquired in one center. This can have important public health implications in these populations. 2005. Spatio-temporal image correlation for volume ultrasound. or maternal disease such as diabetes. demonstrating flow across the defect from right to left. CA.7). Acknowledgment This chapter originally appeared as Yagel S. descending aorta. Video clip 15. dAo. It is reproduced with permission. Novel application of 4D sonography with B-flow imaging and spatiotemporal image correlation (STIC) in the assessment of the anatomy of pulmonary arteries in fetuses with pulmonary atresia and ventricular septal defect. 6. and reviewed by experts in a center connected by telemedicine internet link.4 (a) STIC acquisition in a case of TAPVC. The data archiving and networking capabilities of 3D/4D fetal echocardiography with STIC acquisition open new avenues for disseminating fetal echocardiography programs to outlying or poorly served areas. Video clip 15. which is confirmed in the B-plane (arrow). aortic arch.6 The IVS “virtuaI plane” with color Doppler in evaluation of VSD shows the rendered IVS plane. Lee W. not only by teams or in centers that have made 3D/4D their specialty. 5. Legends for the DVD 4. and left subclavian artery (LSA) are seen projecting from the aortic arch (AoA). References 1. 3D ultrasound compared favorably with 2D in mean scanning time and accuracy of fetal biometry. Application of this modality clearly shows the blood flow in the malaligned vessels. SMA. ductus venosus. In coming years. IVC. Platt LD. DeVore GR. Austria: GE Healthcare. PV. SVC. Cohen SM. The resulting measurements appear in the box. dAo. aorta.7 B-flow modality showing the parallel great vessels in a case of transposition. IVC. The A-plane showed raised suspicion of an anomalous vessel (caret). AO. Ultrasound Obstet Gynecol 2006. Video clip 15.

8. 27. Shinebourne EA. Fetal cardiac ventricle volumetry in the second half of gestation assessed by 4D ultrasound using STIC combined with inversion mode. Heling KS. In: Yagel S. Fetal ventricular mass determination on three-dimensional echocardiography: studies in normal fetuses and validation experiments. Philadelphia: WB Saunders. Automated multiplanar imaging: a novel approach to ultrasonography. Validation of volume and mass assessments for human fetal heart imaging by 4-dimensional spatiotemporal image correlation echocardiography: in vitro balloon model experiments. A new approach to fetal echocardiography: digital casts of the fetal cardiac chambers and great vessels for detection of congenital heart disease. Silverman NH. Three-dimensional power Doppler (3DPD) ultrasound in the diagnosis and follow-up of fetal vascular anomalies. Nicolaides KH. 6th edn. 17: 367–9. Hyett JA. Application of free-hand three-dimensional echocardiography in the evaluation of fetal cardiac ejection fraction: a preliminary study. Bonnet D et al. 30. Yagel S. Three. 34. Ultrasound Obstet Gynecol 2006. Snijders RJM. J Ultrasound Med 2006. Imbar T et al. Gonçalves LF. Yagel S. Achiron R. Benachi A. 24. Achiron R. Platt LD. 27: 128–33. Meyer-Wittkopf M. 23: 1151–9. J Ultrasound Med 2006. The ‘spin’ technique: a new method for examination of the fetal outflow tracts using three-dimensional ultrasound. 17. Nicolaides KH. 18. Espinoza J. London: Martin Dunitz. Ultrasound Obstet Gynecol 2006. The fetal venous system: normal embryology. Gonçalves LF. Application of three-dimensional power Doppler ultrasound in prenatal diagnosis. A novel algorithm for comprehensive fetal echocardiography using 4-dimensional ultrasonography and tomographic imaging. Ultrasound Obstet Gynecol 2004. Cohen SM. 12. 11. A novel method to improve prenatal diagnosis of 23. 21. 29. Spatio-temporal image correlation (STIC): a new tool for the prenatal screening of congenital heart defects.and four-dimensional ultrasonography and ‘inversion mode’.and four-dimensional reconstruction of the aortic and ductal arches using inversion mode: a new rendering algorithm for visualization of fluid-filled anatomical structures. 23: 473–81. Carpenter N et al. 110: 1054–60. 25: 947–56. Vinals F. Ultrasound Obstet Gynecol 2005. DeVore GR. 27: 107–13. Ultrasound Obstet Gynecol 2004. Mazor M. Hyett JA. Cohen SM. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatio-temporal image correlation (STIC). Messing B. Romero R. Cohen SM. Ultrasound Obstet Gynecol 2002. J Ultrasound Med 2004. Gonçalves LF. Mild tricuspid regurgitation: a benign fetal finding at various stages of pregnancy. Snijders RJM. Romero R. Yagel S. 318: 81–5. J Ultrasound Med 2004. Rendering in fetal cardiac scanning: the intracardiac septa and the coronal atrioventricular valve planes. 33. Schmidt S. Ultrasound Obstet Gynecol 2005. Ultrasound Obstet Gynecol 1997. Gembruch U. 23: 535–45. Increased nuchal translucency at 10-14 weeks of gestation as a marker for major cardiac defects. Ultrasound Obstet Gynecol 2003. Three-dimensional quantitative echocardiographic assessment of ventricular volume in healthy human fetuses and in fetuses with congenital heart disease. Lee W. 194: 274–81. Circulation 2004. Ultrasound Obstet Gynecol 2001. Fourdimensional ultrasonography of the fetal heart using color Doppler spatiotemporal image correlation. 24: 415–24. Hartung J. 23: 573–6. 20. J Ultrasound Med 2005. 19. 31. 9. Giuliano A. Espinoza J. 25: 428–34. The three vessels and trachea view (3VT) in fetal cardiac scanning. Abuhamad A. 1998: 349–404. 32. Ultrasound Obstet Gynecol 2005. Bhat AH. Yagel S. Gonçalves LF. Ultrasound Obstet Gynecol 2004. Peralta CF. Raveh D. Ho SY. J Ultrasound Med 2004. Achiron R. Persaud TVN. Lee W et al. Ultrasound Obstet Gynecol 2007. Br Med J 1999. 13. 14. 28.236 Fetal Cardiology 7. Corbett VN. Valsky DV et al. Vinals F. 22: 388–94. Sharland GK. 17: 22–9. Ultrasound Obstet Gynecol 2004. and physiology and the development and appearance of anomalies. Imbar T. Lung and heart volumes by three-dimensional ultrasound in normal fetuses at 12-32 weeks’ gestation. Giuliano A. Anteby EY. 15. The cardiovascular system. Sholler GF. Sciaky-Tamir Y. 25: 701–9. Carvalho JS. Ultrasound Obstet Gynecol 2001. Kivilevitch Z. Perdu M. Moore KL. anatomy. Espinoza J. Chaoui R. Csapo B. 22. Bhat AH. 16. Liu R et al. Ushakov FB. Porat S. Ultrasound Obstet Gynecol 2006. 26: 105–11. Espinoza J. Mazor M. Sklansky MS. Cavoretto P. Am J Obstet Gynecol 2006. Nicolaides KH. Espinoza J et al. Yagel S. Using fetal nuchal translucency to screen for major congenital cardiac defects at 10-14 weeks of gestation: population based cohort study. Corbett V. . 26: 606–9. 20: 340–5. 23: 546–51. 24: 72–82. Examination of the fetal heart by five short-axis views: a proposed screening method for comprehensive cardiac evaluation. 25: 25–31. 24: 696–8. Cole A. Goncalves LF et al. International Society of Ultrasound in Obstetrics and Gynecology. eds. Sequential segmental analysis in complex fetal cardiac abnormalities: a logical approach to diagnosis. Cooper SG. Sharland GK. Poblete P. Prenatal diagnosis of congenital heart disease using four-dimensional spatio-temporal image correlation (STIC) telemedicine via an Internet link: a pilot study. 2003: 321–32. Kusanovic JP. 35. 28: 266–74. 10. 25. Kalache KD. Joubin L. Esh-Broder E. Hoffmann J. Fetal Cardiology. Vargas G. Cardiac screening examination of the fetus: guidelines for performing the ‘basic’ and ‘extended basic’ cardiac scan. J Ultrasound Med 2001. Aubry MC et al. Lee W. 10: 242–6. abnormal systemic venous connections using three. Messing B. Hochner-Celnikier D et al. Polanco B. 26. 20: 317–27. Arbel R. In: The Developing Human: Clinically Oriented Embryology. Mandujano L. Romero R. 30: 142–51. Perdu NL. A nomogram of fetal lung volumes estimated by 3-dimensional ultrasonography using the rotational technique (virtual organ computeraided analysis). Ultrasound Obstet Gynecol 2005. Chaoui R. Falcon O. Ruano R.

Ultrasound Obstet Gynecol 2001. Bromley B. Cooper S. Rappe N. Economides DL. Lin LY. A new approach to the characterization of ventricular septal defects in the fetus.and three-dimensional ultrasound in normal fetuses in anterior spine positions. 17: 485–92. In: Yagel S. Sholler G. Russo MG. 238: 988–96. Birk E. 44. 39. J Ultrasound Med 1998. Radiology 2006. 37. 42. 17: 341–8. 38. Silverman NH. Detailed assessment of fetal ventricular septal defect with 4D color Doppler ultrasound using spatio-temporal image correlation technology. 2003: 201–10. Yagel S. 40. Benacerraf BR. Karidas C. Wang PH. Valsky DV. Shipp TD. Schmidt S.and four-dimensional ultrasound in fetal echocardiography 237 36. Michailidis GD. 18: 17–23. Paladini D. Meyer-Wittkopf M. Three-dimensional (3-D) ultrasonography for obtaining the four and five-chamber view: comparison with 41. London: Martin Dunitz. Three-dimensional US of the fetus: volume imaging. Sklansky MS et al. The ‘in-plane’ view of the inter-ventricular septum. Threedimensional ultrasonography of normal fetal heart: comparison with two-dimensional imaging. Int J Cardiovasc Imaging 2002. 25: 97–8. Meyer-Wittkopf M. Gembruch U. Chen GD. Sierra F. Tartaglione A. Ultrasound Obstet Gynecol 2000. 18: 325–8. Intracardiac shunt malformations. Prenat Diagn 2003. Ultrasound Obstet Gynecol 2001. Vaughan J.Three. Silverman NH. Messing B. eds. 43. Detailed three-dimensional fetal echocardiography facilitated by an Internet link. Imaging comparison of basic cardiac views between two. Simpson JM. cross-sectional (2-D) fetal sonographic screening. Pretorius DH. Three-dimensional (3D) echocardiographic analysis of congenital heart disease in the fetus: comparison with cross-sectional (2D) fetal echocardiography. 23: 1052–5. Levental M. Fetal Cardiology. Ultrasound Obstet Gynecol 2005. . 15: 397–402. Barth H. Vassallo M.

.

Situs inversus. The umbilical vein bends to the right. some basic knowledge is needed in order to understand some definitions as well as the classification of defects. the thoracic and abdominal structures develop asymmetrically. or medical intervention is performed. the inferior vena cava is connected to the right atrium and the pulmonary veins to the left atrium. Since the inferior vena cava and the right atrium are concordant. under normal conditions. the examiner checks the fetal position and the transducer orientation. In these conditions associated heart defects are rare. and some other more mild forms of cardiac malposition or isolated situs inversus may be overlooked even in a child. it can be recommended that. especially when associated with heart block and fetal hydrops. on fetal cardiac anatomy. where there are limited possibilities for the precise differentiation of structures. According to the Baltimore–Washington Infant Study. In this chapter. There are difficulties in achieving a final fetal diagnosis. This situation is called situs inversus. Furthermore. not all aspects of these abnormalities can be covered in detail. The true incidence of this situation is low. we have found Developmental aspects and normal body configuration In contrast to other embryological organs.1) and. When this mirror-image rotation of abdominal and intrathoracic organs is complete. but also not exactly known. or situs inversus completus).2 and 16. In levocardia the heart apex points to the left anterior thoracic cavity with normal atrial and ventricular arrangement (Figure 16. before making this diagnosis. and levocardia (heart on the left side) for the thoracic arrangement (Figure 16. would end as fetal death. and malpositions of the heart Compared to the most common condition of situs solitus and levocardia. ultrasound.1). the segmental analysis of the upper abdomen and the heart. a situation called dextrocardia (Figure 16. not known. The mortality in this small group was 51% within the first year of life. leading to a mirror-image arrangement. accounting for 2. whereas the stomach and the descending aorta are on the right (Figures 16. since persons with this abnormality are asymptomatic and are not identified until an X-ray. It is tried rather to supply the reader with basic information for a practical approach to suspected abnormal conditions. however. The early prenatal detection of an abnormality of this group has a large impact on counseling the pregnant woman. showing that under normal conditions in situs solitus the stomach and descending aorta are on the left and the liver and the inferior vena cava on the right. the focus was on .1 which analyzed 4390 congenital heart defects (CHDs) detected in the first year of life over a period of 10 years (1981–1989). After completion of lateralization. continuing with the portal sinus. the ‘normal’ and most common condition found is then called situs solitus for the visceral arrangement. In the fetus the true prevalence is. In Chapter 12.3). There are well-defined right-sided and left-sided organs and structures. there is very rarely a situation where all organs are rotated exactly to the opposite side. since some forms. Because this condition is extremely rare.2% of CHDs. isomerism was found in 99 cases. especially because small details can radically change the prognosis. as the examiner can rely only on ultrasound. the right atrium and right ventricle are on the left anteriorly and the left atrium and left ventricle are on the right posteriorly and the heart axis points to the right anterior thorax. According to our experience. the liver and inferior vena cava are on the left. malrotations.3) (mirror-image dextrocardia or situs inversus with dextrocardia.16 Cardiac malpositions and syndromes with right or left atrial isomerism Rabih Chaoui Isomerism and related heart malpositions belong to the most difficult chapters in pediatric cardiology.

The normal sonoanatomy of the upper abdomen with situs solitus (left) and the heart in levocardia (right). R. right ventricle. The fetus is in vertex position and in the upper abdomen a mirror-image rotation is found with the stomach (St. LV.) and the aorta (Ao) on the right side and the inferior vena cava (Vci) and liver on the left side. inferior vena cava. left. The fetus is in vertex position.2 Situs inversus. right.1 The two important planes in assessing isomerism and other cardiac malpositions. left ventricle. ST. Figure 16. . AO. The right ventricle (RV) (with trabeculation) is on the left side and in the anterior thorax. aorta. left atrium. L. The descending aorta (Ao) is on the right side. LA.240 Fetal Cardiology Figure 16. VCI. right atrium. Figure 16. The heart points to the right anterior thorax (arrow). RA. RV.3 The heart in a fetus with complete situs inversus and mirrorimage dextrocardia. stomach.

the arrangement is called situs ambiguus (indeterminate or uncertain situs) and is the most complex form of visceral and atrial arrangement. as in dextrocardia (with situs solitus). i. or fluid accumulation. showing an indeterminate visceral situs. associated with either a bilateral right-sidedness . there are other conditions where the mirror-image arrangement is partial: either only affecting the heart. However. and ectopia cordis. as observed in left-sided congenital diaphragmatic hernia. or asplenia– polysplenia syndromes or isomerisms. as in situs inversus with levocardia. cardiosplenic syndromes.e. or affecting the visceral organs. dextrocardia is present when the heart is in the right hemithorax with the axis pointing to the right. Right and left atrial isomerisms in the fetus Conversely to the above-cited very rare abnormalities with a mirror-image arrangement of the visceral and intrathoracic organs. a fetus with the rare finding of right pulmonary agenesis. in consanguineous couples. levocardia. left-sided intrathoracic masses. or in scimitar syndrome. in agenesis of the right lung (Figure 16. a fetus with left-sided congenital diaphragmatic hernia (ST. there are more common conditions with an incomplete lateralization (heterotaxy) of these organs during embryological rotation. the normal position) is mainly considered when there is a situs inversus or ambiguus. Compared to its normal position in the left chest with a normal base–apex axis of 45° to the left side.Cardiac malpositions and isomerism 241 Figure 16. that a high risk for situs inversus was in a family with a previous child with this condition.2 In these conditions heart defects are more common. mesocardia. This group of defects has many synonyms. Levocardia (heart on the left. there is a tendency to symmetric development of the normally asymmetric organs. stomach) with a shifting of the heart into the right thorax. and in the offspring of diabetic mothers. since on X-ray the heart seems more on the right side. and the conditions are regularly associated with complex cardiac defects. In these conditions. The heart position and axis are important features to check while analyzing the heart. As already stated. such as heterotaxy syndromes. Many children having mesocardia are incorrectly grouped in the dextrocardia group.4). In these abnormal lateralizations of the visceral and intrathoracic structures. levoposition (displacement to the left) is seen when the heart is shifted into the left thoracic cavity with the axis still pointing to the left. to stress that the heart still points to the left side.4 Two examples of dextropositions: on the left. Mesocardia is found when the heart points to the midline of the thorax. cardiac malpositions are divided into dextrocardia. Cardiac dextroposition (displacement to the right) is found when the heart is shifted into the right chest with the axis still pointing to the left. as typically found in right-sided diaphragmatic hernia or some other right-sided thoracic lesions. on the right. Similarly. The heart fills the space of the right lung and the left lung fills the thorax.

situs ambiguus with the stomach (ST) on the right side (R) and an interruption of the inferior vena cava with azygos (AZ) vein persistence. which is the dilated azygos vein. The double-vessel sign is typical of left isomerism. Examination of the fourchamber view reveals the diagnosis: a dilated heart with pericardial effusion. In the four-chamber view the heart shows an atrioventricular septal defect (green arrows) and behind the heart again the double-vessel aorta (Ao) and azygos (compare with Figure 16.242 Fetal Cardiology Figure 16. this is a typical finding in left isomerism. Figure 16.6 This fetus was referred with bradycardia. and behind the heart side-by-side with the descending aorta there is a second vessel.13).5 Fetus with left isomerism showing. in the upper abdomen. These cases of atrioventricular septal defects are not associated with Down syndrome. . M-mode shows the presence of a complete heart block: the atria (A) have a normal sinus rhythm (yellow arrows) whereas the ventricles show bradycardia (50 beats per minute) (red arrows).

The . asplenia being the previous name of right isomerism and polysplenia of left isomerism. and this terminology was abandoned4 (although it is often still used in clinical pediatric cardiology). The constant starting point for classifying these conditions is the anatomy of the atria. The diagnostic approach using ultrasound is based on the approach proposed by Huhta et al7 for neonatal echocardiography. position. The identification of these anomalies and their differentiation may be easier for the pathologist during necropsy. Figure 16. and number do not have definitive diagnostic value. Because the spleen develops as a left-sided organ it was in the past a sign for classification.8 Left isomerism with atrioventricular (AV) septal defect with azygos continuation side-by-side with the aorta. it was found that spleen presence. The left atrial appendage is finger-like and has a narrow base. focusing on the upper abdomen and the relationship between the venous system and the atria. whereas the right atrial appendage is pyramidal in shape and its base is rather broad. This group of developmental defects were therefore called cardiosplenic syndromes. (right isomerism) or bilateral left-sidedness (left isomerism).6 Even for the experienced examiner this group of anomalies is considered a challenge.Cardiac malpositions and isomerism 243 Figure 16. Knowing that the heart develops according to welldefined cardiac segments. The central point of diagnosis is the anatomy of the atria defined by their shape and their appendages. Therefore. However. known as the segmental approach. Ivemark3 noted the association of spleen anomalies with some cardiac defects.7 Two examples of left isomerism with the ‘double vessel sign’ (dilated azygos and aorta side-by-side) with the stomach on the left (left case) and on the right (right case).3). diseases with isomerism are now divided into right atrial and left atrial isomerism. In the fetus the diagnostic possibilities are reduced. and the reliable diagnosis or differentiation between right or left isomerism can be very difficult to achieve prenatally. Van Praagh5 proposed a classification according to the anatomy and connections of these segments. This example shows that the diagnosis is achieved by evaluation of the vessels’ position rather than the position of the stomach (compare also Figure 16.

10 Left isomerism. shown that prenatally the morphology of the atrial appendages could have been suspicious in 19 cases.244 Fetal Cardiology Figure 16. In a frontal view the descending aorta and the azygos vein can be seen side-by-side with different flow directions (left). Since the connecting veins are part of the atrial anatomy. appendages can be visualized in a plane slightly cranial to the four-chamber view. Figure 16. Doppler findings demonstrate the venous pattern (right) (arrhythmia with the associated heart block). If persistence of an azygos vein is suspected. the venoatrial connection is the leading diagnostic sign for . however. but are not identified reliably under many conditions. with the typical bilateral sickle-shape appearance in left and the bluntshape appearance in right isomerism.9 Thoracic cross-section (left) and longitudinal view of aorta descendens and dilated azygos vein. the examiner can confirm the diagnosis by using color Doppler. In a recent retrospective study8 on 30 fetuses with isomerism it was. In its further course the azygos vein crosses the diaphragm and connects with the superior vena cava to enter the right atrium (middle).

The heart showed a univentricular atrioventricular connection with a small (v) and a dilated ventricle (V) connected by a ventricular septal defect (*). Figure 16. Figure 16. the ‘interruption’ of the inferior vena cava in its intrahepatic part and its persistence as the azygos (or hemiazygos) vein (Figures 16. . Except the azygos continuation of the inferior vena cava. TP. The azygos vein is then visualized leading into the superior vena cava or into a persisting left superior vena cava (Figures 16.10 and 16. superior vena cava. pulmonary trunk).10). In this fetus with left isomerism the (dilated) azygos vein is seen to connect to the superior vena cava. It can be recognized by the observation of the aorta and the (dilated) azygos vein on its right or left side (hemiazygos) either in the upper abdomen (Figure 16. Sheley et al10 described it as the ‘double-vessel sign’.13). If the examiner is aware of this sign. the diagnosis was confirmed at autopsy (see Figure 16.11 21 showed this azygos continuation sign.11). arrhythmia with heart block (malformed sinus node) (Figures 16.9.Cardiac malpositions and isomerism 245 diagnosis. two leading signs in left isomerism can be expected: first.9 and 16. However. which might help in making diagnoses. The types of cardiac malformation associated with left and right isomerism are complex. The azygos continuation of the interrupted inferior vena cava has been shown to be present in most cases with left isomerism (> 80%).6 and 16. there is a fourth vessel. showing a considerable overlap. AO.5–16. Left atrial isomerism In this condition of double left-sidedness. right-sided structures such as the inferior vena cava and the right atrium with sinus node are absent or may have developed abnormally. second. In another more recent study in 22 fetuses with left isomerism.8). In the following sections some features associated with each anomaly are enumerated. but also in one false-positive case with right isomerism. to the left of the pulmonary trunk.10). and found it in all eight fetuses with left isomerism that they examined. aorta. he can easily detect it prenatally on real-time imaging and confirm it using color Doppler (Figures 16. After termination of pregnancy because of suspected left isomerism.10). The liver lay centrally.12 This fetus was referred at 14 weeks because of nuchal edema (and beginning hydrops) associated with bradycardia. considered as a typical sign of left isomerism.7) or at the level of the four-chamber view (Figure 16. which is the left persistent superior vena cava (LVCS). Therefore. Furthermore.11 Transverse view across the upper thorax showing the so-called three-vessel view (VCS. there are no heart defects that are pathognomonic for the one or the other diagnostic group. We found a heart block. a stomach (St) on the right side (left) and levocardia (H) with a heart defect.

12. The stomach (ST) is nearer the midline than on the left.13 Necropsy of the 14-week fetus shown in Figure 16.12 These hearts tend to be biventricular.4. Figure 16. The conditions of a heart with left atrial isomerism seem to be less severe than those with right isomerism. Heart block detected in the early second trimester is also . the stomach (St) on the right side. and.14 Right atrial isomerism. demonstrating the central liver (left). after removing the liver. there are no cardiac defects permitting a strict classification into one or other group of isomerism. the heart with two left atrial appendages (arrows). In a longitudinal plane (right) both the aorta (blue) and the inferior vena cava (red) are seen in one plane and are both directly anterior to the spine.246 Fetal Cardiology Figure 16.13 The association of an atrioventricular septal defect with complete heart block is considered to be pathognomonic for left atrial isomerism6 and should prompt careful examination of the venous connections. demonstrating a normal ventriculoarterial junction in almost 70% of cases. a simultaneous visualization not seen under normal conditions. and on many occasions a ventricular septal defect or an atrioventricular septal defect is present. In the upper abdomen the aorta (AO) and inferior vena cava (VCI) are on the same side (here the right side (R)) and the inferior vena cava is anterior to the aorta (arrows).

In late pregnancy we observed a herniation of the stomach into the thoracic cavity (arrows) through the intact diaphragm.16 Right isomerism with the typical sign of the juxtaposition of inferior vena cava (VCI) and aorta as in Figure 16.15 either on the left (left case) or on the right side of the spine (right case).12 and 16. Figure 16.14. Atresia of the esophagus or Figure 16. a persisting bradycardia was found in 12/22 cases. on the right. In this group of complex malformations there are no characteristic features.16. 16. such as interruption of the inferior vena cava or the heart block described for left isomerism. and non-fixation of the gastrointestinal tract leading to various degrees of malrotation (Figures 16.15 and 16. and 16. The most severe complex extracardiac malformation observed in left isomerism is extrahepatic biliary atresia with absence of the gallbladder. Right atrial isomerism In this condition of double right-sidedness. In the study reported above in 22 fetuses with left isomerism. but also the symmetrical liver.16). these include not only the common absence of the spleen. The visceral heterotaxy is more common and severe in right isomerism. and the upper gastrointestinal tract are likely to be found malformed. .Cardiac malpositions and isomerism 247 very likely to be due to left isomerism and not to maternal autoantibodies14 (Figures 16. and anomalies of the upper abdomen are more likely to be found in right than in left isomerism.15 In this fetus with right isomerism the stomach was also found to be central.13).11 The position of the heart can be on the left.17). left-sided structures such as the left atrium. The inferior vena cava is present and is generally on the same side as the descending aorta (Figures 16. the pulmonary veins. The position of the stomach can be on the right or on the left side (the stomach is generally more central and the liver is more enlarged). or in the midline.

predominantly atrioventriculary septal defects and right ventricular obstruction in 62% and 48%. and out of six cases with anomalous pulmonary venous return four were not diagnosed prenatally. Owing to the inferior vena cava connection the atrium on the left is recognized as a right atrium and the atrium on the right receives the (abnormal) connections of the pulmonary veins.248 Fetal Cardiology duodenum can be found as well as the herniation of a midline-positioned stomach (Figure 16. An absence of the coronary sinus is found in 85% of cases.17 Right isomerism in a 13-week fetus discerned by the detection of a cardiac anomaly at routine scan. which prompted targeted examination of the vessels showing the typical sign of the juxtaposition of VCI and aorta. A left persisting superior vena cava is found very frequently (Figure 16. The stomach was found to be on the right side.13 Individual differences.18). Univentricular hearts and right-sided obstructions are more represented in right isomerism (see text). total anomalous pulmonary venous drainage is found in 70% of cases (rarely in left isomerism) (Figure 16. which cannot be definitively ruled out in Figure 16.15). Therefore. should be considered when counseling parents after a prenatal diagnosis. although this is generally less severe than in right isomerism. the connections of the pulmonary veins should be examined carefully to rule out the critical infradiaphragmatic pulmonary venous return.11). In a recent study of 21 fetuses with right isomerism. Cardiac failure and hydrops may occur very early. Among extracardiac anomalies. Figure 16.15 20 had complex cardiac anomalies.19). when right isomerism is suspected in a fetus. biliary atresia can be considered as the most severe finding. Heart defects in right isomerism are likely to be more severe than those with left isomerism. but are more likely to be associated with a univentricular atrioventricular connection and exhibit a much higher incidence of abnormal ventriculoarterial connections (double outlet ventricle. There is a univentricular atrioventricular connection to one ventricle (V). the prognosis may be further complicated by the underlying heart defect.18 Fetus with right isomerism. . Postnatally. Hearts with right atrial isomerism can also show an atrioventricular septal defect. depending on the specific finding. leading to spontaneous in utero demise. These hearts show a much higher frequency of pulmonary stenosis or atresia (Figure 16. but it is often difficult to assess the prognosis from the prenatal scan in these conditions.16 Prognosis in isomerism It is known that left and right isomerisms are associated with a poor prognosis. Fetuses with left isomerism show a poor prognosis in utero when associated with complete heart block and hydrops. malposition of the great arteries).4. respectively Only 12 fetuses in this study showed a juxtaposition of aorta and inferior vena cava. The outflow tract evaluation revealed pulmonary atresia: color Doppler demonstrates a normal antegrade flow through the aorta (AO) and retrograde flow across the ductus arteriosus (DA) into the pulmonary trunk (TP). Among the cardiac defects.

79–94% of all children with right isomerism were reported to die. since the diagnosis of isomerism rather rules out such chromosomal aberrations. to the ductus-dependent pulmonary perfusion in right outflow tract obstruction. which is often associated. 13.20 Summary of the four typical findings in the upper abdomen in situs solitus (top left). however. The inferior vena cava (blue) and its position relative to the aorta (red) are main landmarks. Within the first year of life. Under some conditions. children with left isomerism show a better survival rate than those with right isomerism. The diagnosis of right atrial isomerism is very difficult in utero and should be considered in every fetus with a complex cardiac malformation.13 A long-term risk in patients with right isomerism is also infection due to asplenia. utero. 18.19 Abnormal pulmonary venous drainage in a fetus with right isomerism. especially when cardiac or situs malposition are suspected. .Cardiac malpositions and isomerism 249 Figure 16. The association with chromosomal aberrations such as trisomy 21. or others is extremely rare. and is due to the abnormal pulmonary venous connection. situs inversus (top right). or to the complex chamber anatomy. right isomerism (bottom left). Yates et al17 reported. The severity of the disease usually appears postnatally. a fetus with Figure 16. with or without operation. and left isomerism (bottom right).

Knöpfle G. Batukan C. 4: 129–37. 14: 311–14. 48: 97–108. 14: 381–7. Kamil D et al. Neonatal Heart Disease. References 1. Peoples WM. Perspectives in Pediatric Cardiology. Gutgesell H. Berg C. 15. 26: 538–45. 1307–36. In: Drose J. 7. Gembruch U. Polysplenia: a review of 146 cases. This approach can be difficult to achieve in some cases and even omitted in others. Riemenschneider T. fetal heart block.250 Fetal Cardiology isomerism associated with a 22q11 deletion. Blaas H-G. 18. Macartney FJ. Chaoui R. Drose JA. 3. Azygous continuation of the interrupted inferior vena cava: a clue to prenatal diagnosis of the cardiosplenic syndromes. 10. 11. Left atrial isomerism detected in fetal life. Implications of agenesis of the spleen in the pathogenesis of conotruncus anomalies in childhood. Children and Adolescents. Geipel A. the prenatal diagnosis of these abnormalities may be revolutionized by this new knowledge. Hobbins JC. Van Praagh R. Situs inversus viscerum with levocardia. 26: 234–8. 6. Sharland GK. Smallhorn JF. 44 (Suppl 104): 7–110. 2. Raymond FL. Hals J. In: Emmanouilides G. Therefore. Kohl T et al. Huhta J. Loffredo CA. Fetus 1991. Isomerism of the atrial appendages associated with 22q11 deletion in a fetus. Vol 4. Cardiosplenic syndromes. Silverman NH. without associated anomalies. Devoto M. 14.1 but probably on Xq26). eds. Br Heart J 1982. 5. Mapping a gene for familial situs abnormalities to human chromosome Xq24–q27. Heling KS. 77: 1083–8. the examiner should always rule out isomerism when the following ultrasound signs are found: cardiac or stomach malpositions. 1: 75–93. Hansmann M. Fetal Echocardiography. Ivemark BI. The Epidemiology of Congenital Heart Disease. 12. Ultrasound Obstet Gynecol 2005. 76: 548–9. Bjornstad PG. 9. NY: Futura Publishing. Villegas MD. Fetal echocardiographic evaluation of atrial morphology and the prediction of laterality in cases of heterotaxy syndromes. Sheley RC. 1995: 543–60. The syndrome of left isomerism: sonographic findings and outcome in prenatally diagnosed cases. eds. Pedersen T. 1993.20). Ultraschall Med 2005. The Baltimore– Washington Infant Study 1981–1989. Ultraschall Med 2006. Benson L. In: Freedom R. Nat Genet 1993. Two-dimensional echocardio-graphic diagnosis of situs. Smallhorn JF. 4. Edwards JE. Hartung J. Moller JH. Schwabe M. Moss and Adams’ Heart Disease in Infants. Geipel A. Birth Defects 1972. Nyberg DA. 16. Ballabio A. Jones K. Kamil D et al. Ferencz C. abnormal venous connections. Recent results in families with recurrences of visceral heterotaxy cases showed that the gene for these malformations is probably localized on the long arm of the X chromosome (region between Xq24 and X27. Freedom RM. Conclusion Every fetal heart defect should be analyzed in a segmental approach in order to detect (or rule out) an isomerism (Figure 16. Berg C. 1998: 253–62. 1995. Phoon CK. Philadelphia: WB Saunders. Cook A. Baschat A. Casey B. Geipel A. 8: 4–23. Acta Paediatr 1955. Ultrasound Obstet Gynecol 1999. Cardiac malpositions and abnormalities of atrial and visceral situs. 5: 403–7. J Ultrasound Med 1995. The segmental approach to diagnosis in congenital heart disease. J Ultrasound Med 2005.1. 24: 921–31. Mount Kisco. 17. Pediatr Cardiol 1983. First trimester heart block: a marker for cardiac anomaly. 13. Allen H. Berg C. Ursell PC. . ed. Rubin JD.18 In the near future. Baltimore: Williams & Wilkins. Smallhorn J. Hagler DJ. a complex cardiac defect. Magee CA. New York: Springer. Prenatal diagnosis of right atrial isomerism (aspleniasyndrome): case report. The syndrome of right isomerism – prenatal diagnosis and outcome. Heart 1996. O’Leary P. 27: 225–33. Kapur R. Am J Cardiol 1996. and dilatation of the azygos vein. Syndromes of right or left atrial isomerism. Yates RW. 8.

be it at the atrial. Various . and the tricuspid valve normally positioned. first-line prenatal ultrasonographic studies will refer to routine screening performed on all pregnant women during the early second trimester to eliminate major malformations. that in the presence of rightsided obstructive lesions. recording of flow direction through the ductus arteriosus is a reliable means of assessing the degree of severity of all lesions situated on the right side of the heart during prenatal life. echocardiographic diagnosis. any significant interference with the filling patterns of the rightsided cavities of the fetal heart should alter flow velocity waveforms in the systemic veins. A few cases have been reported in the pediatric literature5–8 and during prenatal life. Finally. the right atrial wall shows widespread muscular degeneration and diffuse fibrosis. ventricular. For each condition. make sure that no other malformation is present. Prenatal ultrasonographic diagnosis is easy since the enormous right atrium markedly alters the four-chamber view. the right atrium appears extremely dilated. however. however.5. the circulatory pathophysiology.3 It must be recognized. Pathophysiology common to all prenatal right heart dysfunctions One of the major implications of the parallel disposition of the two ventricles which characterizes fetal hemodynamics is that the respective output of each ventricle could be different. The second-line sonographer must. realized in a specialized unit by an expert sonographer. finally. diameter of the foramen ovale has to be greater than normal values to avoid any restrictive effect. Doppler investigation has demonstrated that in the human fetus. Actually. and.4 These various hemodynamic adjustments can have major influences on secondary cardiac morphological development and on the clinical condition.10 On anatomic examination.9. and (2) the absence of left ventricular failure.2. prognosis. clinical presentation.17 Anomalies of the right heart Jean-Claude Fouron This chapter will cover significant lesions of the right side of the heart that can be observed during prenatal life. Combined ventricular output will then be predominantly made up of blood coming from the left side of the heart in a proportion dictated by the severity of the right-sided lesion. the ductus arteriosus will channel blood from the aorta towards the pulmonary circulation. Consequently. Normal reference values for foramen ovale dimensions and Doppler velocity pattern throughout human pregnancy have been reported. If this amount is significantly reduced. after a brief anatomical description. For simplicity and clarity. or arterial level. impact on obstetrical and perinatal management will be covered. systemic output can be maintained if the following two conditions are met: (1) adequate size of the foramen ovale. In these circumstances.1 Any significant impairment to forward flow through the right-sided channels. Flow through the lungs will comprise blood coming from either the right or the left ventricle. described predominantly in the adult population. depending on the actual amount of blood ejected by the right ventricle. second-line echocardiography will relate to ultrasonographic study focused on the cardiocirculatory system of the fetus. both before and immediately after birth. It will summarize our experience gained from fetal ultrasonographic cardiocirculatory assessments performed in our Fetal Cardiology Unit from the years 1990 to 2006. as well as on the long-term prognosis of the disease. during the third trimester of gestation. will cause an increase in the amount of blood crossing the foramen ovale to reach the left-sided cavities. right ventricular is 28% greater than left ventricular stroke volume. adequate size means that the Anomalies of the inlet The right atrium Idiopathic enlargement of the right atrium This is a rare entity. Histologically.

. Clinically. plagued by its subjectivity. LA. This Doppler criterion is.5 A search in our prenatal database reveals only one case. however. Significant tricuspid regurgitation causing right ventricular dilatation and dysfunction has.11 Somewhat disturbing is the occasional observation of tricuspid valvular dysplasia associated with right ventricular echogenic foci in fetuses with trisomy 21 (Figure 17. aorta. since both the four-chamber view and the relative position of the great arteries will appear normal globally.13 Isolated tricuspid insufficiency carries a good fetal prognosis as long as the foramen ovale is wide enough to accommodate the volume overload caused by the regurgitant flow. In our group of fetuses with isolated tricuspid dysplasia. the anomaly was well tolerated throughout pregnancy in all but one case. leading to the erroneous diagnosis of Ebstein’s anomaly.1).252 Fetal Cardiology degrees of tricuspid insufficiency have been described. Tricuspid dysplasia must be distinguished from Ebstein’s anomaly on the basis of the absence of tethering of the leaflets to the posterior ventricular wall and. identified at 39 weeks of pregnancy.1 (a) Echocardiographic real-time image recorded in a fetus with an idiopathic dilatation of the right atrium (RA). The dysplasia can be either isolated or. been reported. LA RA RV SVC RA Ao LA FO (a) (b) Figure 17.9. This view shows the markedly dilated right atrium and helps to rule out an Ebstein’s anomaly on the base of the normal attachment of the tricuspid valve. and it was an isolated finding in only seven cases. The situation is completely different in the presence of major associated lesions. right ventricle. In our database. 35 fetuses have been classified with this diagnosis. SVC. The isolated form of tricuspid dysplasia can be easily overlooked by the first-line sonographer. However. With the help of color Doppler investigation. After birth. The color Doppler investigation confirms that the foramen ovale is not restrictive. one neonate was seen in cardiac failure related to atrial tachycardia. however. associated with other malformations usually causing increased right ventricular pressure. criteria based on the extent of and surface area covered by the regurgitant jet have been proposed in an attempt to overcome this drawback. since some tricuspid insufficiency is commonly observed in normal fetuses. the more frequent being stenosis or atresia of the right ventricular outlet. delivered vaginally without problems (Figure 17. the evolution is usually uneventful. superior vena cava. The two venae cavae can be seen entering the dilated RA. which will dictate the management and influence the outcome. from the simple thickening of all three leaflets of the tricuspid valve to loss of mobility related to involvement of the chordae tendinae. Vaginal delivery is always possible when the regurgitation is well-tolerated. more frequently. FO. suspicion of tricuspid dysplasia raised by greater echogenicity of the valve will be confirmed by evidence of significant regurgitation on Doppler interrogation. foramen ovale. Postnatal echocardiographic study confirmed the diagnosis. The tricuspid valve Tricuspid dysplasia This classification encompasses a wide spectrum of anomalies. more frequently.10 The few fetuses reported with idiopathic right atrial enlargement had no sign of circulatory failure. to the septal surface. isolated tricuspid dysplasia is usually a fortuitous finding. since the postnatal fall in right ventricular afterload causes a rapid decrease in the volume of regurgitation. left atrium. During a detailed echocardiographic study. Ao. (b) The same fetus is scanned in a sagittal view.2).12 The prenatal prognosis is also dependent on the associated lesions. RV.

15 (4) the severity of tricuspid regurgitation. in the incomplete form of Ebstein’s anomaly.Anomalies of the right heart 253 TV RA Figure 17.2 Four-chamber view of the heart of a 34-week fetus illustrating the hyperechogenecity of the dysplastic tricuspid valve. the other formed by the infundibulum. showing some mobility only at their tips. From a hemodynamic point of view.4). However. pulmonary circulation will be essentially maintained by retrograde flow coming from the ductus arteriosus. TV. Evidence of a dilated right atrium on the four-chamber view will raise suspicion during first-line sonographic screening.14 the right atrial cavity may appear normal. among them the severity of the tricuspid insufficiency. It is crucial that the first-line sonographer realizes that identification of the malformation early in gestation is essentially based on a four-chamber view which includes both atrioventricular (AV) valves. where only the septal leaflets have a lower attachment than usual. This fetus also had a ventricular septal defect (membranous type) and the karyotype revealed a trisomy 21. (3) the surfaces of the right atrium + atrialized right ventricle over the functional right ventricle + left atrium + left ventricle. at this stage of pregnancy (Figure 17. its anatomical characteristics are specific enough to allow reliable prenatal echocardiographic identification. The second-line sonographer has the responsibility of completing the anatomic and hemodynamic assessments by obtaining the following information. combined ventricular output can thus be maintained by an increase in left ventricular stroke volume.and/or postnatal prognostic implications: (1) the ratio of diameters of the functional tricuspid opening over the annulus. therefore. and also the possible presence of stenosis at the subpulmonary and/or pulmonary valvular levels. made up of parts of the ventricular free wall and septum above the functional opening of the tricuspid valve. Indeed. Ebstein’s anomaly Although this malformation covers a wide spectrum in terms of presentation and severity. which opens against pulmonary and systemic resistances. The leaflets are grossly deformed. (5) the presence and type of right ventricular outflow tract obstruction better assessed on the short axis view at the base of the heart. Basically. we retrospectively studied 10 fetuses with Ebstein’s . the redundant and relatively fixed tricuspid leaflets are both stenotic and regurgitant.3). Anterograde flow through the main pulmonary artery will depend on many factors. In severe tricuspid insufficiency. the four-chamber view may not appear suspicious early in the second trimester since the right atrium may not yet be significantly enlarged (Figure 17. tethered onto the septal surface and inferior wall. (10) evidence of atrial or ventricular arrhythmia. however. The foramen is usually large and non-restrictive. This produces two chambers within the right ventricle: one.4). even in the more complete form. the atrialized portion. which offers lower resistance than the pulmonary valve. significantly increased. Furthermore. (8) the cardiothoracic ratio. Flow through this dilated chamber is disturbed not only by the regurgitant tricuspid jet but also by the fact that ventricular depolarization causes myocardial contraction on both sides of the ‘functional opening’ of the tricuspid apparatus. the criteria of severity proposed for extrauterine life do not necessarily apply to the parallel disposition of the fetal ventricles. In prenatal life. A ballooning anterior leaflet guards the outlet of the newly formed atrialized portion. (2) the degree of displacement (ratio of the distance from the tricuspid annulus to the apex over the functional opening to the apex). even though the right ventricular outlet could be anatomically permeable. (6) the ratio of fossa ovale diameter over the length of the atrial septum on a horizontal fourchamber view. tricuspid valve. In an effort to designate prognostic criteria specific to fetal life.11 all these measurements can be obtained on a vertical four-chamber view (Figure 17.17 (9) the flow pattern through the main pulmonary artery and the ductus arteriosus (appearance of retrograde systolic and/or diastolic flows).16 (7) left ventricular output. Despite all this information. the problem comes from abnormal attachments of the septal and posterior tricuspid valve leaflets. Four of our 18 cases of Ebstein’s disease diagnosed before birth were of this form. the functional capacity of the remaining infundibular portion of the right ventricle. causing marked dilatation of the ‘new’ right atrial cavity. which could have major pre. lack of mobility associated with insufficiency of the tricuspid valve might indeed be the only clue for diagnosis. The right-to-left shunt normally observed through the foramen ovale is. permeability of the outflow tract can be difficult to assess with the Doppler technique since the entire right ventricular ejection fraction could go backward through the leaking tricuspid valve. An Ebstein’s anomaly can be observed on the left side of the heart in cases of ventricular inversion. the chances of a given fetus with Ebstein’s anomaly reaching term without problems remain difficult to establish.

b) At 19 weeks of gestation. (a) This four-chamber view shows a slightly dilated right atrium without clear evidence of tethering of the septal leaflets. (c. The attending team is then faced with the necessity of performing fetal extraction by cesarean section which will.254 Fetal Cardiology LA LA RA TV TR (a) (b) TV LA TV TR RA (c) (d) Figure 17.3 went on to show normal circulatory function throughout pregnancy. TR. the indices of severity of tricuspid impairment based on anatomy and the degree of regurgitation. All six fetuses with a ratio of foramen ovale over septal atrial length greater than 0. the permeability of the right ventricular outflow . factors such as the degree of compression of the lungs by the dilated right atrium. The situation is more dramatic in the presence of hydrops fetalis. the major challenge which will influence immediate postnatal outcome is the need to establish efficient flow through the lungs.3 (a. Irrespective of the condition of the fetus. in all likelihood. Fetuses with Ebstein’s anomaly without cardiac failure or hydrops can usually tolerate vaginal delivery without major problems.5). and abnormal attachments of the redundant anterolateral leaflet are clearly identified. Arrhythmia due to extreme dilatation of the right atrium is another factor which. anomaly. This investigation suggests that the prenatal prognosis could be significantly influenced by the ability of the foramen ovale to decompress the right atrium (Figure 17. The high risk of fetal death in the presence of a restrictive foramen ovale could explain why the majority of Ebstein’s anomalies seen in postnatal life are associated with large interatrial septal defects and right-to-left shunt. The risk of postnatal death is thus very high in this group of infants. which usually appears towards the end of the second or the early part of the third trimester of gestation. in effect. Then. tricuspid regurgitation. even if flow through the foramen ovale is not restricted. d) Same fetus at 34 weeks of gestation. (c) The right atrium is markedly dilated. add the risk of prematurity to that of severe circulatory failure in a fetus already crippled by a major cardiac malformation. should be a cause of sudden intrauterine death. the tricuspid leaflets are dysplastic. a severe tricuspid insufficiency is now documented. (b) Color Doppler investigation disclosed tricuspid insufficiency of moderate severity. (d) On color Doppler mode.

In tricuspid atresia with transposition of the great arteries. tricuspid atresia is associated with either normally related (approximately 80% of cases) or transposed great arteries with.2 0.7 0. survival is possible through accommodation of the entire systemic return by the widely patent foramen ovale. 0. ventricular septal defect.3 developed hydrops fetalis (adapted from reference 16). Because of the parallel disposition of the two ventricles. ❒. depending on which of these arteries arises from the right ventricle. A major overload of the left-sided cavities is then observed. with an increased risk of coarctation of the aorta at birth. Hypodevelopment of either pulmonary artery or aorta will also be observed. All fetuses with a ratio lower than 0. in these cases. overload of the left atrial and ventricular cavities also remains part of the picture. With both normally related and transposed great arteries. in relation to the outcome of 10 fetuses with Ebstein’s anomaly. however. Fetal survival is based on major hemodynamic adjustments. . however. RA S and TV ''A'' RV TV LV Tricuspid atresia In postnatal life. even though it is small.3 0. or less frequently without. Secondary dilatation of the aorta associated with variable pulmonary artery hypodevelopment is also part of the classical picture. the aortic arch and its arterial branches going to the head and arms as well as the coronary arteries are perfused by blood coming retrogradely from the aortic isthmus and the pulmonary artery. depending on the size of the defect. the pulmonary artery is dilated while the ascending aorta and the aortic arch are relatively hypoplastic. On the four-chamber view.Anomalies of the right heart 255 tract. this adjustment is compatible with normal peripheral perfusion throughout gestation. In cases of isolated tricuspid atresia and normally related great arteries. •. Pulmonary flow could even be normal or slightly decreased. The prenatal diagnosis of tricuspid atresia can be easily suspected during routine obstetric ultrasound.5 FO / AS 0.4 0. Δ.4 Cardiac image of a 31-week fetus with an Ebstein’s anomaly. The common feature in all these cases is the relative lack of growth of the right ventricle associated with an enlarged left ventricular cavity.1 0 Fetuses (n=10) Figure 17. This vertical four-chamber view allows reliable assessment of (1) the tethering of the septal leaflet (S and TV). hydropic fetus. fetal demise. Right ventricular hypodevelopment is. size of the foramen ovale (FO) over length of the interatrial septum (AS). favorable outcome. asymmetric development of the two Figure 17. (3) the extent of the atrialized portion of the right ventricle (‘A’ RV).18 All but two of the 26 fetuses seen in our unit had normally related great arteries. (4) the degree of dilatation of the new atrium formed by the anatomical right atrium (RA) + the atrialized right ventricle. The pulmonary circulation is provided by retrograde flow through the ductus arteriosus.5 Distribution of the ratio. less marked in the presence of a ventricular septal defect. the cavity can then be well-identified and the tricuspid valve appears as either an imperforate membrane or fused echogenic leaflets with little or no mobility.6 0. (2) the morphology of the anterior leaflet of the tricuspid valve (TV) which is redundant and loosely attached to the ventricular wall. At the great arteries level. the presence of a large ventricular septal defect will cause passage of flow from the left to the right ventricle and some development of the right ventricular cavity. and the level of pulmonary vascular resistance will all come into play. the widely patent foramen ovale remains an absolute necessity. and the main pulmonary artery or aorta may be within normal limits. This explains why tricuspid atresia is frequently classified under the hypoplastic right heart syndrome.

(5) the sizes of the pulmonary artery and aorta. In reviewing our cases. not only with real-time examination of the inlet of the right ventricle showing the absence of tricuspid valve movement. however.6 (a) Four-chamber view recorded in a 20-week fetus with a tricuspid atresia. for immediate care of the neonate in a tertiary center must be underlined. (c) During the following systole. (3) the integrity of the ventricular septum. ventricular septal defect.6a). blood velocities are seen going towards the aorta from the right ventricle because of the presence of an associated pulmonic stenosis. In the absence of hydrops. In the absence of a ventricular septal defect. however. left ventricle. (b) The color Doppler investigation shows in diastole that blood is entering into the left ventricle through the mitral valve and being shunted into the right ventricle through the ventricular septal defect. the survival of these neonates will rely on the patency of the ductus arteriosus. The second-line investigation will confirm the diagnosis. . The following additional points must be clarified: (1) the size of the fossa ovale and the mobility of the membrane of the foramen. (2) the size and function of the left ventricle. LV.6b and 17.6c). The relative hypoplasia of the right ventricle is obvious in this image. VSD. All these elements should help to establish the prognosis either before birth or immediately after birth. in the presence of a large ventricular septal defect where forward pulmonary blood flow can be efficiently LV VSD VSD RV LA TV TV (a) (b) LV VSD TV Ao (c) Figure 17. The most likely explanation is that cases of tricuspid atresia with relatively small foramen ovale do not survive beyond the first trimester of gestation. which represents the only source of pulmonary flow. and normally related great arteries. (4) the relative position of the main arteries. Arterial oxygen content will indeed be primarily dependent on the respective proportion of pulmonary and systemic venous return that will enter the left ventricle. but also by the lack of flow velocities across the tricuspid valve in pulsed and color Doppler investigations. (6) the size and direction of flow through the ductus arteriosus (Figure 17.256 Fetal Cardiology ventricles is obvious (Figure 17. ventricular septal defect. it is surprising to see that all had good left ventricular function and no sign of circulatory failure. vaginal delivery is certainly possible in a fetus with tricuspid atresia. The situation is not quite the same. The need.

. The four-chamber view can appear completely normal. the appearance of myocardial hypertrophy could decrease ventricular diastolic compliance and consequently end-diastolic volume. and prognosis. The second-line sonographer who makes the diagnosis of mildto-moderate pulmonic stenosis in prenatal life must complete the study and be sure to: (1) look at the integrity of the septal wall.7). This is the main reason why second-line sonographers should always perform a Doppler flow velocity investigation through the semilunar valves in all high-risk patients screened for cardiac malformation. After birth. pulmonary valve. main pulmonary artery. mild-to-moderate pulmonic stenosis has to be distinguished from critical stenosis or complete atresia of the valve. favoring an increase in the right-to-left shunting through the foramen ovale.8 m/s. When tricuspid atresia is associated with transposition of the great arteries. such cases can even develop cardiac failure due to high cardiac output. the thickness and mobility of the pulmonary valve are variable. PV. clinical presentation. at that point.7 (a) Short-axis view at the base of the heart showing the outflow tract of the right ventricle and the pulmonary artery around the aorta. (2) investigate the tricuspid valve to rule out tricuspid regurgitation. right ventricular filling characteristics change and end-diastolic pressures tend to rise. In the moderate form. the size of the aorta. and the neonate may not have significant cyanosis. It is important to remember. The morphology of the right ventricle can be normal. the secondary fall in right ventricular output renders the modified Bernoulli equation based on the peak velocity measurements20 unreliable for the assessment of the severity of the stenosis. Anomalies of the outlet: pulmonary stenosis The presence or absence of ventricular septal defect profoundly modifies the dynamics of obstructive lesions along the right ventricular outlet. cardiac failure. a gradient of 20 mmHg was found through the pulmonary valve. In the mild-to-moderate form. Without ventricular septal defect On the basis of morphologic appearance. The leaflets of the pulmonary valve are echogenic but the infundibulum and the pulmonary artery are of normal size. on both the five-chamber and the shortaxis views at the base of the heart to eliminate ventricular septal defects (a three-dimensional (3D) ultrasonographic modality could help in this search). pulmonary flow is normal or increased. some right myocardial hypertrophy can be present particularly at the end of gestation. Mild-to-moderate pulmonary stenosis can be very difficult to detect on routine obstetric ultrasound because of the relatively minor changes observed on cardiac morphology. and the possibility of coarctation or arch interruption are all elements of concern. however.5 m/s which is above the normal limit of 0. that in prenatal life. disclose abnormal echogenicity of the leaflets. MPA. (3) record RV PV MPA Ao (a) (b) Figure 17. The relationship between the increase in peak velocity through the stenotic valve and the severity of the lesion remains linear as long as the diastolic function of the right ventricle is not altered by myocardial hypertrophy. Careful examination of the pulmonary valves could. An increase in peak flow velocities through the valve can be the only clue to the diagnosis (Figure 17. The extent of this fall in preload will vary according to the gestational age at which the hypertrophy occurs.19 In the presence of moderate to severe stenosis. in these cases. however. however. (b) The pulsed Doppler tracing recorded a few millimeters beyond the pulmonary valve reveals a peak velocity of 1.Anomalies of the right heart 257 maintained through the defect.

22 Major anomalies are also present in the coronary arteries. muscle cell disorganization has been described. the neonate does not require any specific postnatal care. In critical stenosis or complete atresia of the pulmonary valve.23 Sinusoids are sometimes present. characterized by abnormal origin and distribution. The cavity itself is small and almost virtual. A progressive increase in severity of the stenosis is rare. or interruption. the anatomy of the lesions varies according to the status of the tricuspid valve. coronary stenosis. Usually. (c) A pulsed Doppler recording of flow velocities within the same dilated coronary artery seen in (b) (but with the cardiac apex pointing towards the top of the picture) shows that in systole (S) the blood is circulating from the myocardium toward the aorta at the peak velocity of 3 m/s (gradient of 43 mm between right ventricle and aorta). the right ventricle is markedly hypertrophied. connecting the ventricular cavity with the coronary circulation (Figure 17. in diastole (D). .8 (a) Two-dimensional real-time image of the four-chamber view in a 20-week fetus with pulmonary atresia and intact ventricular septum.258 Fetal Cardiology flow through the ductus arteriosus to see whether it is retrograde or fully anterograde.8). The left-sided cavities are dilated. a reverse flow is noted from the aorta toward the right ventricle. the absence of proximal aortic coronary connections. (4) measure the ratio of left to right ventricular end-diastolic diameters as well as the mitral and tricuspid annuli. and (5) measure the size of the pulmonary valve annulus. The prenatal prognosis of mild-to-moderate cases of valvular isolated pulmonic stenosis is excellent. (b) In the same fetus. On histology. Delivery in these cases can be achieved by the vaginal route.21 In the absence of tricuspid insufficiency. a color Doppler interrogation recorded in systole reveals a dilated branch of the coronary artery (arrow) which is draining blood from the right ventricular cavity towards the aorta (Ao) suggesting the persistence of sinusoids within the myocardium. and an elevation of the observed gradient could well be related to the physiological rise in flow through the right side of the heart. It has been shown experimentally that an increase in pressure within the right Ao RV LV (a) (b) D S (c) Figure 17. A hypoplastic and trabeculated right ventricle (RV) is seen with very thickened free walls.

24 A link between these findings and the histopathologic anomalies seen in pulmonary atresia with intact ventricular septum is a possibility which deserves investigation.8). (4) measure the diameter of the pulmonary valve annulus compared to that of the aorta. he must (1) evaluate the severity of insufficiency to establish the type of pulmonary atresia. the diagnosis can be made quite easily at the first-line routine screening. evidence of progressive flow restriction through the foramen ovale. the blood flowing back and forth through the incompetent valve causes some growth of the ventricular cavity and prevents a significant increase in ventricular pressure above the systemic level. the size of the main pulmonary artery should also be assessed.9). Subsequent management of the neonate will be based on echocardiographic evidence of a right ventricular cavity with sufficient functional ability to warrant balloon dilatation of the valve (Figure 17. looking for sinusoids (Figure 17. The presence of sinusoids establishing wide communication between the ventricular cavity and the coronary LV RV TR (a) (b) Figure 17. therefore. . The cardiologist assessing a fetus with critical pulmonic stenosis and intact ventricular septum must study the tricuspid valve morphology and rule out significant leaks. the myocardium appears less hypertrophied and with almost normal organization of myocytes. in pulmonary atresia associated with significant tricuspid regurgitation. using the color Doppler mode.Anomalies of the right heart 259 ventricular cavity of fetal lambs was associated with a marked accumulation of free oxygen radicals within the myocardium.9). (3) determine the orientation of flow through the ductus arteriosus as an important element of diagnosis. the intrauterine prognosis of critical stenosis or complete atresia of the pulmonary valve depends on both the size of the foramen ovale and left ventricular function. vaginal delivery is possible. suggesting that they represented survivors from what was a wider spectrum from which fetuses with restrictive foramen ovale were eliminated due to intrauterine death early in gestation. In these cases. or fluid accumulation in either the pleural or the peritoneal cavity should be an indication for cesarean section. During the immediate postnatal period. since retrograde flow should be expected.9 (a) Real-time image of the heart of a fetus with pulmonary valvular atresia and severe tricuspid regurgitation. this evaluation will influence postnatal management inasmuch as it will help to establish the indication of perforation followed by dilatation of the valve in the cardiac catheterization laboratory (Figure 17. and shows variable degrees of right ventricular hypoplasia. In addition. the right ventricle (RV) appears trabeculated but relatively well developed and the diameter of the tricuspid annulus is only slightly smaller than that of the mitral valve. (2) examine carefully the right ventricular myocardium. In this view. The 18 fetuses seen in our unit with isolated pulmonary atresia or critical pulmonary stenosis with intact septum were all in good clinical condition. pulmonary atresia with intact septum represents a typical example where arterial oxygenation is entirely dependent on patency of the ductus arteriosus. Usually. In both cases. The four-chamber view is indeed abnormal. the increase in rightto-left shunting through the foramen ovale is responsible for enlargement of the left-sided cavities. Any sign of left ventricular dysfunction. By contrast.8 m/s. the intravenous administration of prostaglandin. (5) evaluate the size and function of the left ventricle as well as the size of and flow through the foramen ovale.9). Urgent initial treatment is. (b) Recording of pulsed Doppler in the right atrium showing a marked tricuspid regurgitation (TR) with a peak velocity of 5. Sinusoids are usually absent (Figure 17. The mode of delivery will be decided according to the functional condition of the left ventricle. in the presence of tricuspid regurgitation.

Consequently. The sizes of the pulmonary annulus and artery vary according to the severity of the stenosis. Perfusion of the lungs comes from the ductus arteriosus. With ventricular septal defect Classical tetralogy of Fallot This malformation is a consequence of anterior displacement of the infundibular septum. This latter diagnosis is ruled out by identification of the blind outlet of the right ventricle followed by a small pulmonary artery. In cases of truncus arteriosus. be darkened by the relatively high incidence of chromosomal and extracardiac anomalies in this group. These fetuses are usually able to support vaginal delivery. This good fetal hemodynamic picture can. When the diagnosis of tetralogy of Fallot is confirmed. no indication.10c). surgical palliation will have to be considered. Combined cardiac output remains within normal limits. Cases with decreased or diastolic retrograde flow through the ductus arteriosus in prenatal life can be expected to require specialized care immediately after birth to maintain the ductus open. aneurysmal . however. however. all fetuses with tetralogy of Fallot should be delivered in a tertiary center. be normal. Tetralogy of Fallot can easily be missed during first-line sonographic screening unless adequate sonographic views are recorded. This picture is sometimes misinterpreted as being that of a truncus arteriosus. All of our 69 patients with this classical form of tetralogy of Fallot had no sign of circulatory failure. even in less severe cases. this should be one of the best criteria for severity assessment: reverse flow will be observed through the ductus arteriosus only if right ventricular output is not sufficient to maintain normal flow through the lungs. the pulmonary artery arises directly from the trunk. Doppler investigation is unreliable for this purpose since the amount of blood which flows through the infundibulum is inversely related to the severity of the stenosis. in isolated tetralogy of Fallot. of course. at the level of the AV valves. never described in association with tetralogy of Fallot (Figure 17. however. the subaortic ventricular septal defect is usually obvious. Tetralogy of Fallot with absent or dysplastic pulmonary valve This entity. the septum will look perfectly intact and the view will be described as normal (Figure 17. Its anatomic features are: large ventricular septal defect with an overriding aorta. which appears tortuous. preferably on the short-axis view at the base of the heart (Figure 17. undergo cardiocirculatory evaluation immediately after birth to assess the efficacy of the collaterals and of the ductus arteriosus in maintaining adequate pulmonary blood flow and sufficient oxygenation of arterial blood. The cardiocirculatory condition of these fetuses is usually good for the same reasons as given for the classical form of tetralogy of Fallot. Ideally. we identified eight such cases in our files. The aorta is dilated. the main pulmonary artery is small or sometimes absent. They must. close monitoring of the clinical condition is justified after normal closure of the ductus arteriosus. Significant coronary artery stenosis is another factor which will influence survival. (2) the degree of right ventricular outflow tract obstruction.10d). the aortic arch and isthmus are well developed and coarctation of the aorta is. only visual assessment is therefore possible. If the four-chamber view is obtained posteriorly. In this view.10a). and from aortopulmonary collaterals. although rare. The immediate postnatal condition and management will be influenced by the severity of the stenosis and associated extracardiac anomalies.26 There is. The ventricular septal defect is large.11). for all practical purpose.25 Microdeletion of chromosome 22 may be found in up to 20% of infants with tetralogy of Fallot. for extraction by cesarean section. In this condition. The only change will be the significant increase of the portion of combined output that comes from the ascending aorta compared to the pulmonary artery. overriding the septum (Figure 17. Prenatal ultrasonographic diagnosis is based first on an abnormal five-chamber view showing the very dilated aorta overriding the septum with a large ventricular septal defect. The second element of diagnosis of tetralogy of Fallot with pulmonary atresia is the presence of aortopulmonary collaterals which can be seen in prenatal ultrasonographic studies coming from the thoracic descending aorta (Figure 17.260 Fetal Cardiology circulation is an absolute contraindication to any attempt at opening the right ventricular outlet. Tetralogy of Fallot with pulmonary valve atresia This is considered an extreme form of the disease characterized by complete obstruction of the right ventricular outlet. The clinical condition of fetuses with tetralogy of Fallot is usually good. Characteristically. (3) the pattern of flow through the ductus arteriosus. needs to be identified during prenatal screening. causing narrowing of the right ventricular outlet chamber. in these cases.10b). This form of tetralogy of Fallot is less frequent. The relative position of the great arteries will. Eight of the 69 fetuses also presented a complete AV canal. this is due to the presence of the ventricular septal defect which allows decompression of the right ventricle towards the aorta. the second-line sonographer must specify the following criteria of severity: (1) the ratio of pulmonary over aortic annulus diameters. Prenatal diagnosis of tetralogy of Fallot is established on a five-chamber view where the aorta appears dilated. they are always smaller than the aortic annulus and ascending aorta. because of the septal malalignment.

which is frequently observed. (c) A truncus arteriosus is ruled out and the diagnosis of tetralogy of Fallot established by the identification of a small pulmonary valve annulus (arrow) connected to the right ventricle. The two cases from our database showed no sign of cardiocirculatory failure throughout pregnancy. (b) On the five-chamber view. Close examination of the right ventricular outflow tract will show the dilated pulmonary artery with abnormal or absent pulmonary valves. the subaortic ventricular septal defect is clearly observed as well as the dilated ascending aorta (Ao) and the septoaortic discontinuity. Studies of rat fetuses with tetralogy of Fallot and absent pulmonary valve created by maternal administration of bis-diamine led to the conclusion that neural crest cells are important in formation of the pulmonary valve and that enlargement of the pulmonary arteries and bronchial compression develop in fetal life. associated with pulmonary artery hypoplasia. right ventricular dilatation. There are. suspicion is raised by dilatation of the right ventricular cavity on four-chamber view. appearance of the dysplastic or rudimentary pulmonary valve.28 During routine ultrasonographic screening. Careful investigation with both real-time and color Doppler echocardiography will confirm absence of the ductus arteriosus in the majority of cases.Anomalies of the right heart 261 LV LEFT Ao RIGHT Septum (a) (b) (c) (d) Figure 17.12). The intrauterine prognosis of tetralogy of Fallot with absent pulmonary valves is uncertain.27 Ultrasonographic diagnosis of this form of tetralogy of Fallot is easier than in the classical form. (d) Sagittal view of the same fetus showing the aortic arch which is dilated. absence of ductus arteriosus. finally. Here again. the diagnosis is confirmed on five-chamber view. Doppler investigation of the dilated pulmonary artery will disclose systolic antegrade and diastolic retrograde flows caused by severe pulmonary insufficiency (Figure 17. and. however. dilatation of the pulmonary trunk. in the . (a) This four-chamber view taken posteriorly at the level of the atrioventricular (AV) valves could be described as normal. In the lungs. unlike in the classical form of tetralogy of Fallot. the dilated peripheral pulmonary artery can be clearly seen with color Doppler interrogation. The reduction in diameter usually observed at the level of the isthmus is not present.10 These images were obtained from a fetus with the classical form of tetralogy of Fallot.

in addition. Suspicion of the specific diagnosis will be based on the thinness of the right ventricular free wall with the absence of any apparent contraction (Figure 17.13b). (b) Pulsed Doppler recording in the main pulmonary artery above the valve illustrating the systolic anterograde (S) and diastolic retrograde (D) velocities. On first-line screening examination. but in both cases the same flow pattern was observed in Desc Ao Collat Figure 17. (a) Short-axis view at the basis of the heart showing the pulmonary valve (PV) which is echogenic and dysplastic. the to-and-fro movement of blood between the right ventricle and the dilated pulmonary artery may result in elevated right ventricular end-diastolic pressure. On the other hand. the main pulmonary artery and both branches are well visualized. Note the absence of a ductus arteriosus between the pulmonary artery and descending aorta (Desc. the ascending aorta (AscAo) is dilated.262 Fetal Cardiology literature reported cases that developed polyhydramnios29 and/or hydrops fetalis. only the pericardial and endocardial layers are present with some adipose tissue and a few myocytes. Despite this major myocardial impairment and systemic pressure in the fetal pulmonary circulation.13a).Ao). anterograde flow through the pulmonary valve has been documented in our two cases (Figure 17. Major respiratory problems occur after birth due to the associated abnormal development of bronchi which. the four-chamber view will be abnormal. Echocardiographic diagnosis of Uhl’s anomaly is relatively easy. PV AscAo Desc. leading to a possible increase in venous systemic pressure and hydrops fetalis.Ao D S (a) (b) Figure 17. On microscopic studies. the right ventricular free wall appears thin and translucid. The right ventricular cavity is markedly enlarged and hypotonic.12 Ultrasonographic images taken from a fetus with tetralogy of Fallot with rudimentary pulmonary valve. . It is essentially characterized by the lack of development of myocytes.31 On macrocospic examination.30 Tracheobronchial and esophageal compression by the dilated pulmonary artery is considered to be the etiology of the polyhydramnios. are frequently compressed by the markedly dilated pulmonary artery. Anomalies at the myocardial level Primary anomalies Uhl’s anomaly This myocardial impairment specifically involving the right ventricle has been rarely described in prenatal life.11 The color Doppler investigation of this fetus with tetralogy of Fallot and pulmonary atresia discloses the presence of aortopulmonary collaterals (Collat) originating from the thoracic descending aorta (Desc Ao).

To establish postnatal prognosis. suprasystemic pressure will then be present not only in the pulmonary circulation but also in the right ventricle.31 Morbidity and mortality later in life are frequently related to major ventricular arrhythmia. (a) The short-axis view reveals marked dilatation of the right ventricle (RV) and the thinness of the myocardium. retrograde flow (red color) coming from the ductus arteriosus (DA) is however also documented. One of our two cases died suddenly at home at the age of 14 months. the second-line sonographer will have to assess flow through the main pulmonary artery as well as the size and function of the left ventricle. . Those fetuses who were able to reach term without evidence of circulatory failure should. The real-time picture frequently shows an hourglass appearance of the ductus (Figure 17. It is not known whether adult patients reported with this disease had the full pathological picture since birth. Ultrasonographic diagnosis of restrictive ductus arteriosus is based on both bidimensional imaging and Doppler flow velocity signal through the ductus. assessment of the capability of the right ventricle to maintain pulmonary flow is mandatory. it is possible to observe some anterograde flow (blue color) through the pulmonary artery (PA). characterized by an early systolic shunt from aorta to pulmonary artery followed by a normal anterograde flow in late systole and during diastole (Figure 17. however.14b. and pulsatility index (systolic velocity − diastolic velocity/ mean velocity) < 1.13 Images of a fetus with Uhl’s anomaly.33 An example of these changes is shown in Figure 17. These conditions deserve mention since they are not infrequent. such as indomethacin. peak systolic velocities > 140 cm/s. More frequently. it can become restrictive after maternal administration of prostaglandin synthetase inhibitors. A similar effect on the ductus arteriosus has been reported with the administration of betamethasone. the normal pulmonary arch is made up of the main pulmonary artery and ductus arteriosus. After birth. This pattern is compatible with a late right ventricular contraction (right bundle branch block). With color Doppler interrogation. is good as long as the medication Secondary anomalies Some fetal conditions can cause secondary myocardial changes which predominantly involve the right ventricle.Anomalies of the right heart 263 PV PA DA RV Ductus arteriosus (a) (b) Figure 17. The ductal segment can be absent in certain types of malformation.14a).13b).34 The prognosis. After birth.32 a glucocorticoid frequently given for fetal lung maturity induction. (b) Pulsed Doppler flow velocity recording in the ductus arteriosus shows reverse flow early in systole (above the zero-velocity line) followed by the normal anterograde flow in end-systole and during diastole.9 have been proposed as expressions of ductal constriction. Using the Doppler technique. Another hemodynamic problem is related to frequently associated tricuspid insufficiency. presumably from arrhythmia. Restrictive ductus arteriosus During prenatal life. facilitated because of the fall in pulmonary vascular resistances. the ductus arteriosus. Shortand long-term prognosis is reserved. since cardiocirculatory failure has been described in this condition. used for suppression of premature labor or for reduction of polyhydramnios. Progressive right ventricular hypertrophy associated with tricuspid regurgitation is then observed. diastolic velocities > 35 cm/s. as mentioned in the section on tetralogy of Fallot. in theory. These fetuses must be identified quickly. however. and this anomaly is alternatively named ventricular dysplasia or arrhythmogenic right ventricle. A decision concerning the type of delivery will be based on the clinical condition of the fetus. be able to tolerate vaginal delivery. Any decrease in the diameter of the ductus will cause a gradient between the main pulmonary artery and the aorta. the right ventricular performance is.

36 upregulation of the renin–angiotensin system in the donor. Elevated concentration of endothelin in the recipient. cases of subpulmonary stenosis have been described. both in systole and in diastole.264 Fetal Cardiology is discontinued. causing passage of blood from one fetus (the donor) to the other (the recipient). S.14 (a) Color Doppler flow imaging of the ductus arteriosus of a fetus on indomethacin for polyhydramnios. RV S Pulm. . The recipient twin becomes polycythemic and plethoric.15 illustrates such right ventricular outlet stenosis in a case of twin–twin transfusion. septum. (b) Pulsed Doppler taken above the pulmonary valve of the same recipient twin showing an increase in peak velocity of 1. the pulsed Doppler sample volume encompasses both the pulmonary artery (PA) and the ductus arteriosus (DA). Although this myopathy involves both sides of the heart. Note the thickening of the septum and the free walls of the ventricle. The flow velocity waveforms of the two arterial segments are superimposed and reveal a significant increase in peak velocities through the ductus. Artery LV (a) (b) Figure 17.37 and early DA PA (a) (b) Figure 17. disappearance of Doppler evidence of ductal constriction is usually observed within 24–48 hours after the arrest of medication.15 (a) Four-chamber view of the heart of the recipient twin in the context of a twin-to-twin transfusion syndrome. The twin-to-twin transfusion syndrome This syndrome is observed in monochorionic twins and is characterized by arteriovenous connections within the placenta.5 m/s.35 Figure 17. Pressure rather than simple volume overload is increasingly considered as the most likely explanation for this cardiomyopathy. (b) In this tracing taken in the same fetus. Note the hourglass appearance of the ductus arteriosus. One of the features of such a condition is that the myocardium of the recipient twin tends to become thickened and echogenic.

(d) Simultaneous recording of Doppler flow velocities in both the aortic isthmus and the ductus arteriosus illustrates the clear preponderance of the right ventricular ejection volume: blood flow is retrograde in the isthmus during the second half of systole and the entire diastole. All five cases of aneurysm of the vein of Galen found in our database had massive arteriovenous shunting (Figure 17. This situation is typically observed in intracranial cerebral arteriovenous fistula.16b). These conditions. (c) In the same fetus. the greater are the shunts. or coarctation of the aorta. SVC (a) (b) R L Isthmus DA (c) (d) Figure 17. the earlier will be the time of appearance of neurologic and cardiocirculatory symptoms. . the neonates will show essentially right ventricular hypertrophy on electrocardiogram. restrictive foramen ovale. the right-sided cavities are mildly dilated.39 Diastolic overload of the right ventricle Intracerebral arteriovenous fistula. (b) Because of the direct arteriovenous connection. Some prenatal circulatory anomalies. will be discussed in Chapter 20. but a few cases have been described where progression of right ventricular outflow tract obstruction was observed. mitral disease. This most frequent form of cerebral fistula is characterized by aneurysmal dilatation of the vein of Galen draining a network of cerebral arteries.16a and 17. right.16 (a) Color Doppler interrogation in a fetus with aneurysm of the vein of Galen.Anomalies of the right heart 265 diastolic functional impairment in the recipient twin38 are all elements in favor of the pressure-overload pathogenic concept. Both myopathy and stenosis usually regress after birth. however. Overload of the right ventricle can be observed in all conditions affecting fetal left ventricular function because of the parallel disposition of the two ventricles. Direct connection of an anterior cerebral artery (arrow) into the dilated vein can be seen causing a jet of blood curling around into the aneurysm. such as hypoplastic left heart syndrome. L. At birth. the flow velocities through the superior vena cava (SVC) present an arterial pattern with marked forward flow during diastole. could cause isolated right ventricular diastolic overload while the left heart is normal. left. for obvious reasons. R. The degree of shunting through the fistula is variable but.

(c) On the two-dimensional image of the SVC. In addition to cerebral damage.17 Echocardiographic findings in a 25-week fetus with anomalous drainage of the right pulmonary vein into the superior vena cava (SVC). (a) The four-chamber view is abnormal. Anomalous pulmonary venous drainage: These anomalies are described in more detail in Chapter 29. (d) Color Doppler shows clearly at least two pulmonary veins (PV) (colored in red) coming from the right lung. The common hemodynamic feature of all variants of partial anomalous pulmonary venous connection is the diastolic overload of the right-sided cavities. The aortic blood (colored in blue) is also visualized. Retrograde diastolic flow of blood coming from the ductus arteriosus is observed through the aortic isthmus40 (Figure 17. (b) The M-mode tracing confirms the right ventricular dilatation with a flat septum.16d). A team approach is essential for the management of both fetuses and neonates. Besides causing dilatation of the right atrium and ventricle (Figure 17. this volume overload is transmitted to the pulmonary circulation and ductus arteriosus.16b).16c). showing right atrial and ventricular enlargement. a suspicious vascular imaging (arrowhead) can be noted at the junction of the SVC with the right atrium (RA).17 is an example of an anomalous connection of the right pulmonary veins to the superior vena cava observed in a RV RA LV RV (a) (b) RA SVC PV (c) (d) Figure 17. The prognosis of cerebral arteriovenous fistula diagnosed in prenatal life is somber.16c and 17. .266 Fetal Cardiology The prenatal hemodynamic changes are related to the marked increase in volume of flow drained by the superior vena cava going into right-sided cavities (Figure 17. draining into the SVC very close to its junction with the right atrium. Figure 17. intrauterine cardiac failure is a frequent compli- cation. parallel to the SVC.

Deanfield JE. Wilson GJ. Chemtob S. These associated malformations are responsible for biventricular dilatation. In: Redington 18. Takao A. 28. Sahn DJ. Abnormalities of the coronary circulation in pulmonary atresia and intact ventricular septum. Fetal Diagn Ther 1993. Callan NA. Fouron JC. Phillipos EZ. Noninvasive prediction of transvalvular pressure gradient in patients with pulmonary stenosis by quantitative two-dimensional echocardiographic Doppler studies. Color Doppler is very helpful in this process. Prognosis and outcome of idiopathic dilatation of the right atrium in children: a cooperative study of 15 cases. 65: 20–3. Allan LD. Am J Perinat 1991. The echocardiographic assessment of the human fetal foramen ovale. Chartrand C et al. Allwork SP. 21. Robertson MA. Bender R et al. Prenatal diagnosis of tetralogy of Fallot with absent pulmonary valve. 15. Villain E. Pavenik D. Drblik SP et al. Respondek ML. Kozlj M. Lindesmith GC. 8: 15–17. Sahn DJ. Baltimore: University Park Press. Witsemburg M. Assessment of fetal pulmonic stenosis by ultrasonography. 64: 547–9. J Pediatr Surg 1983. Barber JC et al. 12. Still KD. Pulmonary atresia with intact interventricular septum: diagnosis in the fetus. Kammermeier M. 67: 866–71. Burchell HB. Fouron JC. 4. Am J Obstet Gynecol 1994. Lima CO. 77: 202–1. The prevalence and clinical significance of fetal tricuspid valve regurgitation with normal heart anatomy. Anderson RH. Outcome in neonates with Ebstein’s anomaly. Ninomiya Y. Paladini D. Idiopathic dilatation of the right atrium: postoperative follow-up in a child. Importance of microdeletions of chromosomal region 22q11 as a cause of selected malformations of the ventricular outflow tracts and aortic arch: a three-year prospective study. The anatomic substrate of pulmonary atresia and intact ventricular septum. Stewart P. Pediatr Radiol 1977. 9. Marin-Garcia J. Cook AC. Gersony WM. 27. A propos of a case diagnosed in utero. Jacob F. Hatchwell E. Freedom RM. eds. 33: 1117–19. Webber SA. Momma K. 7: 257–63. Med Clin North Am 1949. 8: 305–8. 17: 167–73. Teyssier G et al. Elzenza N. 14. Anderson RH. Takahashi M. Kasznica J. Idiopathic enlargement of the right atrium. Fetal venous circulation – an update on hemodynamics. Cullen S. Pediatr Cardiol 1992. J Perinat Med 2000. Chita SK. 13. Sameshima H. In the presence of a four-chamber view showing right atrial and ventricular enlargement. AN. Am Heart J 1961. Circulation 1983. Celermayer DS. 74: 1208–16. Wilson AD. Ando M. Fetal cardiac morphology of tetralogy of Fallot with absent pulmonary valve in the rat. 25. Am Heart J 1998. In: Tucker BL. J Am Soc Echocardiogr 1996. Greewold WE. Brawn WJ. Prenatal diagnosis and circulatory characteristics in tetralogy of Fallot with absent pulmonary valve. 26. Congenital pulmonary atresia with intact ventricular septum: clinicopathologic correlation of two anatomic types. Sharland GK. Reinhardt-Owlya L. Arch Mal Cœur 1997. Syamasundar Rao P. Nishibatake M. 20. 129: 26–32. 16. 18: 196–8. 17. Marçon F et al. Prognosis in fetal tetralogy of Fallot. identification and connections of the pulmonary veins are mandatory. 9: 805–13. Zorman D. Edwards JE. Circulation 1986. 13: 1–4. 2. Changes in intracardiac blood flow velocities and right and left ventricular stroke volume with gestational age in the normal human fetus: a prospective Doppler echocardiographic study. Pernot C. Tricuspid valve disease with significant tricuspid insufficiency in the fetus: diagnosis and outcome. . 8. Fouron JC. Sekarski N. 11: 161–3. References 1. J Am Coll Cardiol 1992. Doubilet P et al. 6. Sharland GK. The spectrum of Ebstein’s anomaly of the tricuspid valve. Idiopathic dilatation of the right atrium simulating Ebstein’s anomaly. Edwards JE. J Thorac Cardiovasc Surg 1979. Ludomirsky A et al. 22: 143–6. Pavlova M. Allen RG. Prenatal measurement of cardiothoracic ratio in evaluation of heart disease. Am J Cardiol 1989. Circulation 1990. Plappert E. Valdès-Cruz LM et al. Kan JS. 1984: 217–55. Hornberger LK. Antenatal diagnosis of tetralogy of Fallot with absent pulmonary valve accompanied by hydrops fetalis and polyhydramnios. Hurni M et al. Total supradiaphragmatic anomalous venous drainages are frequently associated with other cardiac malformations such as atrial isomerism with intracardiac shunts. 23. 5. Fouron JC. 82: 1343–51. The Right Heart in Congenital Heart Disease. Sahn D. Tokudome T. Idiopathic dilatation of the right atrium revealed in childhood by dysrhythmias. Rev Port Cardiol 1992. 114: 1415–20. Blaysat G. 6: 52–4. 1998: 25–34. 7. Aeschlimann S. Angelski R. 10. 30. 19. Blanc WA. 24. Factors affecting the prognosis of Ebstein’s anomaly during fetal life. 3: 491–4. J Am Coll Cardiol 1991. Hoeffel JC. 91: 645–9. J Am Soc Echocardiogr 1994. Zuberbuhler JR. Arch Dis Child 1990. Davignon AL. Normal fetal foramen flap and transatrial Doppler velocity pattern. 90: 645–8. J Am Soc Echocardiogr 1990. DuShane JW. Kiserud T. Pediatr Cardiol 2001. 19: 1041–6. Am Heart J 1987. Congenital tricuspid atresia: a classification. Allan LD. eds. 28: 90–6. Kleinman CS et al. Castor S. Ursell PC. 3. DaSilva AM. London: Greenwich Medical Media. 135: 1081–5. These veins are relatively easy to follow in prenatal life because of the fluid-filled lungs. Generation of reactive O2 species in the myocardium of newborn lambs following intra-uterine increase in right ventricular pressure. Pediatr Cardiol 1998. Sullivan ID et al. 171: 1265–70. 29. Arch Mal Cœur Vaisseaux 1998.Anomalies of the right heart 267 25-week fetus. Idiopathic dilatation of the right atrium diagnosed in utero. 19: 420–1. Third Clinical on Congenital Heart Disease: Obstructive Lesions of the Right Heart. Kenny JF. 22. 62: 591–602. 11. J Pediatr 1996.

Circulation 1987. 33. 72: 74–9. Fetal pleural effusion following maternal indomethacin therapy. Lamoureux J et al. Murray HG. 100: 277–82. Pathogenesis of twin-twin transfusion syndrome: the renin-angiotensin system hypothesis. Clinical and echographic features of in utero cardiac dysfunction in the recipient twin in twin-twin transfusion syndrome. Takao A. 39. 40. Pediatr Res 1989. Huhta JC. Flower J. Strand L. Joly D et al. 25: 69–75. 75: 406–12. Sinclair B. . Early intertwin differences in myocardial performance during the twin-to-twin transfusion syndrome. 32. Zosmer N. Momma K. Fetal Diagn Ther 2000. 38. Guiraudon G et al. 33: 536A. Fouron JC. P